Daniel A. McFarland and Charles J.

Gomez

Organizational
Analysis

Acknowledgements

The material presented in this textbook consists of lecture notes that agglomerated into
their present form after nearly a decade of teaching organizational analysis at Stanford University. If there are positive features of the text and the course, then we think it fair to attribute such accolades to the scholars we heavily draw upon. In particular, the theoretical
work of Dick Scott, Graham Allison, Herbert Simon, James G. March, John Kingdon, John
Seely Brown, Paul Duguid, Joanne Martin, Deborah Meyerson, Gideon Kunda, Jerey Pfeffer, Gerry Salancik, John Padgett, Mark Granovetter, Paul Dimaggio, Woody Powell, Arthur
Stinchcombe, Michael Hannan, John Freeman and Glenn Carroll (and many more!) have all
been an inspiration to us and we have relied heavily on their work and its insights. We encourage all the readers of this text to go out and study these authors primary works and to
take their classes wherever and whenever possible.
DM and CG
September 2013.

Copyright Notice

The authors have made a concerted eort to ensure all appropriate attributions have
been made and copyright clearances obtained prior to publication of this work. If you find
any errors and copyright concerns please contact the lead author. We will make special
eorts to correct errors and address concerns as quickly as possible. Similarly, if you
have any comments, or would like to request permission to use this work or a part of it,
please contact the lead author (mcfarland@stanford.edu). And thank you for your interest
in Organizational Analysis!
Front Cover Source http://commons.wikimedia.org/wiki/File:2/23/Hong_Kong_Skyline_Restitch_-_Dec_2007.jp
g
ii

Table of Contents
!
!

Introduction to Organizational Analysis ...... 1

Decision-Making in Organizations............ 16
Coalition Theory ....................................... 38
Organized Anarchy .................................. 52
Organizational Learning ........................... 71
Organizational Culture ............................. 89
Resource Dependency Theory .............. 106
Network Forms of Organization ............. 122
Neoinstitutional Theory .......................... 144
Organizational Ecology .......................... 163

1
Introduction to
Organizational Analysis

Source: http://commons.wikimedia.org/wiki/File:Skyscrapers_of_Shinjuku_2_7_Desember_2003.jpg

Organizational Analysis
In this introductory chapter, you will be introduced
to the concept of an organization. In so doing, it
will become clear that organizations are everywhere and come in many different forms. Their
ubiquity means that many pressing social problems are organizational in nature. Their variability
and complexity require study. And this is why we
need courses on organizations all so we develop
a better understanding of the world we live in and
how to better manage it.
What Is and Is Not an Organization?

Let us begin with our preconceptions and understandings. What is an organization? What is
not an organization? When most of us consider
organizations, we think of hospitals, schools, businesses, stores, companies and factories. But what
about families, various voluntary associations, and
even street gangs? What qualities make something an organization or not?
One of the best writers on organizations has
been Richard Scott, whose work we will draw on
heavily from time to time. Scott defines organizations this way: Organizations are conceived as social structures created by individuals to support the
collaborative pursuit of specified goals (Scott
2003: 11). There is a lot packed in here, so lets
simplify it some. What Scott means is that organizations are groups whose members coordinate
their behavior in order to accomplish shared goals
or to put out a product. Given this, lets reconsider
what is and is not an organization.
At some point, we encounter cases that are
unclear. Some features of the definition may be
lacking, while other features may be present. Take
for example, a social movement. Many social
movements have specified goals, but the social
structure - or pattern by which participants associate - is emergent and can change dramatically
from one event to the next. As we reach ambiguous cases like these, the key features defining an
organization grow unclear they are less of a
group, involve less coordination, and /or are less

goal-oriented. And then in cases that are not organizations, we see all these features no longer apply.
Examples

Organizations

Not
Organizations

Ambiguous
Cases

Qualities

Companies,
schools,
Roles, rules, goals,
families and recurring behaviors,
voluntary
clear boundaries.
associations
Random
collections
of persons,
isolated
individuals
Street
gangs,
friendship
groups,
social
movements

No roles, rules,
goals, pattern of
recurrence, or
boundary.

Less clear roles,

rules, and goals,
porous boundaries
and fluid
participants.

Table. Organization as Concept

Varieties of Organizations

Now that we have some sort of idea what is and

is not an organization, we can start reflecting on
how common and important organizations are. Organizations accomplish most of what society
wants and needs. From socialization (in schools)
to re-socialization (in prisons and mental health
care facilities), from tax collection, public administration, protection and soldiering, to production
and distribution of goods, service provision, preservation of culture, communication, and even recreation. Organizations are the means by which many
of our collective goals are pursued and accomplished. For example, would disaster relief or
schooling be possible without organizations focused on these efforts? Organizations are so common that they have become the medium of modern
social life we cannot imagine existing outside
them. We live in a world greatly made up of for2

mal organizations, their rules, structures, goals,

members, and instrumental efforts. Organizations
are also collective actors (or social entities) that
take action, use resources, own property and enter
contracts. They are groups that have attained
thing status.
Organizations are everywhere and they vary
tremendously. They vary in size such that some
are huge and others are small. For example, IBM
employes hundreds of thousands of employees,
while a community youth organization may be run
out a basement and employ only a few individuals.
Organizations vary by market sector, whether private industry or public sector not-for-profits. They
can even be voluntary associations like unions,
parent-teacher associations, and religious groups.
Their social structures also vary. Some are hierarchical like the military and football teams; some
are centralized dictatorships like perhaps Henry
Ford and Andrew Carnegie managed in the 192030s; others have flat governance structures like
consulting firms, while yet others are horizontally
differentiated into many different divisions and
relatively autonomous units like university departments.
Organizations vary by their context or surrounding environment. Some firms vary by the
temporal context or era in which they are in. For
example, the context for the federal government is
very different today than it was in 1790; and a
time of recession is very different for most firms
than a time of economic boon. Firms also experience regional differences reflecting different cultural contexts. For example, Euro-Disney worked
very differently than Californias Disneyland and
required a different organizational model and approach for relating to the local population. In sum,
the main idea here with environment is that the
same organization may not have the same effect in
a different time, culture, and set of participants.
Organizations are everywhere, they are very
important to the functioning of society, and they
are very diverse. They have also changed a lot
over the last 50 years and have altered the modern
world as a result. For example: manufacturing has
given way to a service industry in the U.S.;
women have become half the labor force; part time

subcontracting has grown; and so forth. The organizational world we live in is changing right
underneath us.
Organizational Problems and Reform

Because organizations are everywhere and

varied, they are often a source of consternation
and social problems. All too often, our problems
are organizational ones and we want to reform the
firms we interact in. Through this course, you will
gain a better appreciation of organizational complexity and the difficulties of redirecting organizations in desired directions. Sometimes coordination and contracts fall apart and need to be renegotiated; schools do not live up to expectations and
need reorganization; a military may be gender biased and need to change; and government regulation fails to prevent corruption. Participants frequently propose and implement reforms in an effort to change an organization. Many reforms fail
long before they are ever implemented. They are
either rejected outright or they are dramatically
adapted to the local context. Those that are implemented, often end up looking like something very
different from what they were planned to become.
Much of my research focuses on educational
organizations like schools and universities, so
many of the reforms I see try to change the nature
of schooling. Most of these reforms fail. In fact
they fail so routinely that a teacher gave me a list
of 45 failed school reforms, adopted in very
piecemeal fashion, that went through his school
over the span of 20 years. Here is that list:

roles, participants, and goals, thereby supplanting

others or shifting attention. Since organizations
are in great part complex systems, the features are
interrelated, and a change in one element can frequently result in problems elsewhere.
This textbook and course will help you think
more deeply and clearly about how organizational
reforms are generated and implemented, and what
factors likely contribute to their success or failure.
Why Understanding Organizations Matters

Table. Failed School Reform Lingo

Many of these are jargoned and hard to interpret, but they often target change efforts on certain
organizational features over others. For example,
some are focused on the social structure. Lead
Teachers #8 is one such attempt. The goal of that
reform is to insert an additional level in the flat hierarchy of faculty roles. So in stead of just having
faculty and then a department chair, there is now
an additional level of Lead Teacher in between.
Other reforms present a technology or schooling
process that caters to a particular goal. For example, Heterogeneous Grouping #13 gives students
an active role in their education, and emphasizes a
goal of equality. The reform calls for group-work
instruction where different task roles are rotated
(e.g., speakers, note-takers, etc). Yet other reforms
attempt to manage pressures from the external environment (#12, 27).
Most of these reforms are developed and
tested in one school and then packaged and applied in many other contexts. Unfortunately, the
local environment of each new context often differs from the original testing ground. As a result,
the reforms goals may not be valued by the local
managers, or the targeted change may disrupt
other valued tasks and missions. In addition, there
is a governance structure in place within most
schools and districts that is threatened by change
efforts (especially those with external origins)
usurping their established coordination patterns.
In short, every reform emphasizes certain rules,

To this point I have presented a working definition of organizations and explained just how
common they are. Now I want to sell you on why
organizations matter: learning about organizations,
reflecting on how they operate, and considering a
variety of means by which they can be managed is
an important skill most everyone today should develop. We live in an organizational society, and
many of the problems we confront are organizational in nature. We need to better understand and
manage organizations if we are to evolve as a society.
This course attempts to provide you with
such training. It is an introductory course on organizations that helps you grapple with the complexity of institutional life. The course focuses on
actual cases of non-profits, educational institutions, government agencies, private firms and the
policies aimed at changing them. The course material is designed for advanced undergraduates, masters students, and PhDs interested in organizations.
So lets cut to the chase what is the utility
of this course to managers, policymakers and analysts? Why should you care? Organizations are
everywhere! You cannot change society or understand much of it without knowing something about
organizations and how they work. Unfortunately,
the social reality of organizational life is pretty
messy and complex. Therefore, we need conceptual frameworks to help us make sense of it. For
example, what should you pay attention to? What
matters? What does not? Where do you begin if
you want to study and change them? This course

offers you conceptual frameworks and tools by

which to do this.
Through this textbook and course, you will
better understand the problems that organizations
-- like schools, universities, non-profits and private firms -- confront. There are so many problems that arise in an organization that it is hard to
relate all of them, but here we can name a few:
Organizations confront problems of defining
objectives (goals).
Organizations struggle to get people to show
up and perform services (tasks).
Organizations worry about the coordination
of lots of people trying to accomplish these
tasks, and even how to coordinate different
tasks with one another (coordination / implementation).
There is always a concern of drawing necessary resources from the environment organizational inputs like money or revenue, materials, knowledge (input).
Then they have to worry about outputs dispensing ideas, products, and funds to the environment (output).
There is also the concern with selecting, training, and replacing members as participants
move thru these organizations (participants).
Organizations even worry about relations outside the firm ties to neighbors and fits with
the surrounding environment (environmental
fit). For example, Walmart cannot just up
and move into any neighborhood!
This course exposes you to a variety of actual
organizational cases and then organizational theories that help make sense of what you have observed. Through this course you will learn there is
nothing more practical than a good theory. Many
of you have organizational experiences that will be
of great value to this course. Think of them as experiences from which you have developed different accounts or interpretations. In most cases,
your accounts focus on certain features of the organizational context, attribute causal force to certain elements and certain actors over others, and
come to certain conclusions as to why things hap-

pened the way they did. Those accounts are in

many ways a folk-theory (or proto-theory). But as
we all know people have different accounts of the
same phenomenon, and the same explanation or
way of seeing organized life cannot be universally
applied. In many regards, it is not enough to adopt
one theory or one perspective on everything in
whatever career you pick, you will confront new
problems and new situations where your previously generated explanation does not apply or
where another perspective altogether is needed.
This textbook and course exposes you to multiple theories of explaining and managing organizations. Why? To help you develop accounts that
are different from the ones you already know. To
help you think in new ways about organizations so
that when you go out and study one or manage
one, you do not just draw on rules of thumb that
will likely never work in a particular case, but
adopt different ways of seeing and thinking about
the organizational phenomenon in focus.
So this course provides you with different perspectives you may not have considered before.
When you look at an organization now, it may
seem unbearably complex and composed of an endless array of features. Through organizational theories, you will learn to listen for different kinds of
music in all the noise. Each theory picks up on different features of organized life and renders them
into explanatory narratives you can use. By implication, my hope is that you will learn different,
and perhaps better ways of managing.
Finally, this textbook and course are designed
to enrich your understanding of organizational phenomena and your experiences within them. You
will not be given a laundry list of advice of rules
of thumb that soon go out of style or fail to apply
to the novel situations you will likely confront.
There are no silver bullet solutions here. You will
be given a set of tools ways of seeing, understanding, and managing the complex reality of organizations. I will leave it up to you, and the actual organizational cases that interest you, to discern which tool (or combination therefrom) best
applies.

Features of Organizations
We will now identify some core analytic features
of organizations. These analytic features give us a
language or terminology we can use to make sense
of firms, their various forms, and their prevailing
problems.
Elements of an Organization

Organizations are complex, so it helps to have a

concept space, a set of things, or elements to focus
on in discussing them. This requires some abstraction from the details of our personal experiences in
organizations. Scholars like Leavitt (1965) and
Scott (2003) identify a finite set of organizational
elements for us to consider and focus upon .

EN
VI

RO

NM

EN
T)

EN
VIR

ORGANIZATION)

ON
ME

NT
)
Social Structure

Social)Structures)
Technology)

Goals)

VI
EN

Par6cipants)

RO
T)
EN

NM

M
ON
R
I
V
EN

T)
EN

Figure. Features of an Organization

(adapted from Leavitt 1965: 1145; Scott 2003:18)

In this diagram, an organization is represented as

having a boundary and being placed in a wider environment. Every organization has certain elements: it has a set of social actors or participants,
a social structure by which they interrelate, goals
or a mission, and a set of technologies or tasks it
performs in order to render inputs into desired outputs. Lets take each of these elements of an organization in turn.
Participants

First. we have an organizations participants.

These are a firms social actors, employees, and
stakeholders. Participants are social actors that
make contributions to and derive benefits from the
organization. For schools these social actors are
adults and children, and they typically assume
roles like administrators (superintendent, principal), teachers, students, staff (from custodians,
counselors, nurses, cafeteria workers, to administrative assistants), and even parents and politicians
connected to the school in various ways. Participants can also be organizational actors, like firms
in a particular industry. If you recall, we noted
that organizations are often things listed in contracts, and considered unitary actors. As such,
firms can be participants in one anothers affairs.
For example, in the technology industry, firms often contractual relations, partnerships, shared
boards of director, and so on, and it is through
these relations that they influence one anothers
affairs.

Second, we have an organizations social structure.

This concerns features that regulate and establish
the usual pattern of relationships between participants. So social structure concerns the persistent
relations existing among participants within an organization. These can vary in form, from some being vertically differentiated with lots of status levels, while others are horizontally differentiated
with many different departments and divisions.
These social structures can vary in their degree of
formality. Formal structures entail clearly prescribed and demarcated social positions while informal structures emerge and are unplanned relations that persist. In a school, the formal structure
might reflect the prescribed roles we briefly mentioned above: principal, assistant principal, department chair, teacher, students, counselors, etc. All
are roles with relational obligations. The informal
structure might be the actual advice relations and
friendships that arise between participants. For example, some teachers may be popular and a locus
of authority even though they lack such a formal

position. Likewise for students: some may hold

undue authority and influence the manner in which
curricula are taught.

ioral coordination. What principles and beliefs

give shape to these structures so peoples behaviors adhere to them? Is it one of authority and control in the formal organizational chart; or is it one
of task adaptation via the informal organization?
Goals

Figure. Formal and Informal Social Structure

(source - http://farm4.static.flickr.com/3303/3527073630_a9abc78619.jpg)

Social structures are more than recurring behavioral patterns they are also cultural systems
that entail normative principles and cognitive beliefs (Scott 1995). In fact, these cultural aspects of
social structure often guide behavioral patterns.
For example, adults in classrooms often follow
norms and ideals concerning how a teacher or manager should interact with others. That is, we have
a sense of better and worse role-performances, and
organizations tend to reward performance that
most coincide with the ideal.
Social structure can run even deeper and reflect cultural cognitive beliefs and understandings.
For example, we find it hard to imagine schools
without teachers and students, and this belief is distinct from our sense of better or worse ways to perform those roles. The belief that every school has
to have those roles is a deeply ingrained belief.
The belief may invoke particular behavioral norms
of teaching (say traditional or progressive), and in
turn, this may partly shape the behavioral patterns
witnessed in an organization like a school. But it
need not do so perfectly. Other social structures
are at play like those of gender roles, class differences, peer cultures, etc., and they can cloud the
clean appearance of prescribed forms of behav-

Third, organizations have goals Desired ends

that participants attempt to achieve through the
performance of task activities (goals of schooling
e.g., technical and moral socialization of youth;
if we focus on faculty in universities like Stanford,
we can see a historical change in what goals are in
place: from one of student training, to that of research production, to one of resource acquisition,
to community service. Many organizations have
multiple goals, and they can come into conflict.).
If we look at concrete missions, they vaguely relate some of these ends. Companies often relate
general goals such as Citi, Levis and HarleyDavidson. But organizations also vary in the extent to which their goals are focused or multifaceted, clear or ambiguous.

Our goal for Citigroup is to be the most

respected global financial services company.
Like any other public company, we're
obligated to deliver profits and growth to our
shareholders. Of equal importance is to deliver
those profits and generate growth
responsibly.

(Source: http://commons.wikimedia.org/wiki/File:5/54/Citiipblogo.gif)

Technology

Fourth, an organization has a technology or a

means by which organizations accomplish work or
render inputs into outputs e.g., tasks. Tasks are
often called technology because machines and
factory lines accomplish many tasks. What is processed varies from material inputs of manufacturing
equipment to people being processed, educated or
coordinated to become more knowledgeable and
active citizens. For example, in schools, the technology consists of lesson plans, curricula, and

courses that transform students (input) into socialized adults (output).

Environment

Last is the Environment: the physical, technological, cultural and social context in which an organization is embedded. For example, what is the environment a school confronts? Schools are often dependent on state and city governments for resources and funds, they rely on trained workers
and teachers from local universities, they depend
on the neighborhood they are situated in for clients
and student populations, etc.
Environments can vary culturally in the
sense that Euro-Disney initially did not work because an American version of Disneyland could
not just be plopped down in Europe without some
changes. Environments can vary technologically
such as having an office in Silicon Valley where
everything is wired for internet access and videoconferencing, in comparison to say my parents
home where they are still figuring out a compact
disc player. Physical environments also matter
consider for a second something as basic as your
firms location in a cold region versus a hot dessert. Very different pressures emerge because of
these distinctive physical environments.
Elements
Description
Actors / Participants Organizational participants that make
contributions to and derive benefits from the
organization.
Social Structure
Persistent relations existing among
participants in an organization.
Goals
Desired ends that participants attempt to
achieve through the performance of task
activities.
Technology / Tasks Means by which organizations accomplish
work or render inputs into outputs.
Environment
The physical, technological, cultural, and
social context in which an organization is
embedded.

Table. Overview of Organizational Elements

Environment Linkages

All of the internal features of an organization

can come into relation with elements in the environment (Scott 2003: 23-24). Lets take each in
turn.
1. Participantenvironment linkage: how porous is the boundary for participants in the
school? Is it a total institution like a boarding school or monastery, or is it a loose commuter campus like a community college?
2. Technology-environment linkage: no organization develops all of its own tasks and technologies. They borrow many of them. Also,
they have to adapt to the norms and pressures of larger occupational structures and
professions. Do schools get most of their curricula from textbook publishers, university
faculty, and practitioners in other schools?
3. Goals-environment linkage: the social value
we attribute goals varies. In some communities the safety of students may be of greater
concern than their achievement.
In one
neighborhood, concerns about suicide will
matter, while in other schools it may be the
provision of equal opportunities.
While
many of these same goals arise across settings, they vary in their salience from environment to environment.
4. Social structure-environment linkage: most
schools look the same in terms of roles, but
different communities espouse different beliefs and norms about how these roles should
be performed. For example, elite schools
may worry more about stress and progressive models of teaching, while struggling
schools may see the best teachers as ones
who meet standards. In fact, in some communities, the ideal form of instruction may
entails wrote learning and traditional modes
of teaching.

All the organizational elements tend to have various relationships with one another. While we
only list the interrelations with environment, it is
feasible to consider the linkages between goals
and participants, such as how goals can lead participants to self-select into a firm, and for it to
form a company reputation and identity. In effect, with each case, one can find that these features of organizations form a system of interdependence. By identifying that system, the analyst acquires a deeper understanding for form
functioning, behavior and management.
Degree of Ambiguity

These abstract elements are seldom clean, simple features in real world cases, however. In
fact, ambiguity is more often the reality we confront. For example, schools are often described
as having uncertain technologies for accomplishing technical and moral socialization. We
have courses or course labels, but its far from
clear that particular tasks and lessons lead to certain desired outcomes and which do so more effectively over others. Also, we have ambiguous
indicators of accomplishing said goals e.g., do
we use achievement tests or citizen tests? Are
these tests biased and inaccurate? Furthermore,
participants can belong to multiple organizations, so the question becomes which organization most influences them. Children spend most
of their day in school, so it is a relatively contained environment in comparison with other organizations. Nonetheless, children bring with
them all sorts of baggage and experiences from
elsewhere (family), and these can influence their
behavior in school.
How Can All of These Elements Work Together
as a System?

Fortunately, Richard Scotts review of organizational research not only identifies organizational
elements, but it also describes how theories in
different eras focused on certain organizational
elements over others and characterized their in-

terrelation in certain patterns (Scott 2003: 2630). In short, he recognized three classes of organizational theory.
The earliest class of theories regarded organizations as rational systems Here, the theories characterized an organization as a collectivity oriented toward the pursuit of specific goals
and whose behavior exhibits a formalized structure. These theories tended to focus on the administrative units of organizations and their process of rational decision-making.
An ensuing class of organizational theories
characterized organizations as natural systems
here, the theories related an organization as collectivities whose participants pursue multiple interests, forged in conflict and consensus, but
who recognize the value of perpetuating the organization as an important resource (they want
to survive). In a natural system, it is the unplanned, emergent relations and coalitions
which matter: the informal structure of relations
that develops among participations is more influential in guiding behavior than the formal structures role expectations and guiding principles.
This class of theories regarded an organization
as an adaptive organism.
Most recently, organizational theories have
come to characterize organizations as open systems here, organizations are congeries of interdependent flows and activities linking shifting
coalitions of participants embedded in wider
material-resource and institutional environments. This class of theory focuses more on the
environment than any other organizational feature.
Now one could argue that these theories
reflect the organizations of their day. But I am
not sure that is the case. Most organizations still
entail all these features and the processes that rational, natural and open system perspectives entail. Another view might be that organizational
theories expanded their focus as our understanding of firms and instrumental social groups
grew. All these features were likely there, and
perhaps shifted some in salience, but to this day
rational, natural, and open system qualities persist in many organizations.

Rational
Single organization, or
administrative unit
(organization as unitary
actor)

Natural
Open
Single organization
Multiple organizations
w/multiple actors and
(organizational field)
divisions (organization as
coalition)

Actors /
Participants

Leaders, organization
(admin unit)

Participants across roles

Stakeholders, employees,
and in direct environment and even mass consumers

Social
Structure

Formal & planned /

hierarchical

Informal & emergent >

formal (external seeps
in/ norms enter)

External world permeated

internal organization
(beliefs enter)

Goals

Specific missions /
objectives

Multiple, conflicting
goals

Survival / legitimacy in
environment

Technology /
Tasks

Maximization / Decision Contingent decisions /

trees / Standard operating Unintended outcomes
procedures
(efficacy)

Less decision, more

emergence &
environmental
determinism (legitimation)

Environment

Ignored

Major role

Primary Unit
of Analysis
Organizing
Concepts

Minor role

Table. Classes of Organizational Theories

(adapted from Scott 2003: 26-30)

Case: Overview of the Adams Avenue School Using Organizational Elements

Adams Avenue School
At this point, I want to discuss the case of a
school reform effort and identify the organizational elements being discussed within it so you
have a concrete sense of their application. I draw
on the example of a school organization because
most everyone has experienced one. Regardless of
how old you are or where you are from, you have
some sense of how a school operates and how an
organizational reform might operate in them.
The case was not written for organizational
analysis, but rather for educators (Metz 1986).
Hence, the writer Mary Metz selects the features she thinks characterize the case. We want to

identify the organizational elements in the case

and see if they help us understand what sort of account it is. What elements are the point of focus?
Which are characterized as having an interrelation
and being changed? Through such an application,
I hope you will begin to see that even when we
take off the book shelf a nearly random case, we
can see how the case draws our attention to particular details and how they interrelate that we see
the beginnings of an organizational theory that
helps explain the case.
Adams Avenue School is a case about the
creation of a magnet school. A magnet school is a
pubic school that offers a specialized curriculum
so as to draw students from across zones of a city.
This magnet school serves middle school students
10

(grades 6-8), and uses a particular curriculum that

will hopefully serve both struggling and high
achieving students. It is meant to build a sense of
community, to bring into the fold lower academic
performing students, and to improve student
achievement.
Before becoming a Magnet School, Adams
Avenue School was a 7th grade annex to Williams
Junior High School (Grades 7-9), an overcrowded
school with an African American population in the
poorest part of the town. Williams Annex was established to relieve overcrowding as well as severe
problems of discipline and under-achievement.
The Williams Annex was voluntarily staffed by
young faculty who lacked seniority at Williams
Junior High. Mrs. Michaels led the annex and she
had a good deal of say in how they developed their
program (But she was not an official principal).
She and her colleagues decided that the annex
should follow a multi-unit plan and be divided into
3 small schools of around 100 students each.
Meanwhile, the district was going through
change, and it was planning a new magnet school
program. The Williams Annex was selected as a
great site for becoming a new magnet, called Adams Avenue Magnet School. Mrs. Michaels was
selected as the principal, and the faculty decided to
adopt an individually guided education curriculum
(IGE). In the IGE curriculum, students proceed at
their own pace and complete a series of individualized tasks showing mastery of the material. Before the school opened, the faculty was sent for
training in this program, but it was tailored toward
elementary school students, so they were illprepared when the school opened to 6-8 grade students. When Adams Avenue opened, the faculty
lacked materials but made due the best they could.
By their second year, the school received funds
and faculty received more training in IGE. According to Metz, the school seemed established and
had a coherent program in place by year 3.
Parent Involvement

affairs and took up a good deal of the principal and

assistant principals time. These parents were not
afraid to tell the teachers what to do and check up
to see that they followed through. In addition,
they campaigned for the school at board meetings
and with the district office to retain an assistant
principal position and to get a larger, better building. As the schools reputation grew, it attracted
the interest of lower-middle-class families and ordinary families. By the third year of operation, the
magnet schools population reflected that of the
surrounding community and had less of a bifurcated population of highly educated families and
working class families. By the third year, the very
high achieving kids of the initially aggressive parents were now in 8th grade and would soon graduate.
Individually Guided Education

The textbooks for the IGE curriculum was

not as clear as it might seem, given it was a specific plan written by a specific group of educators.
Faculty found it far from clear in implementation.
Hence, the principal had broad discretion in how it
was defined. Eventually the curriculum was defined on two sets of requirements. The first specified a number of concrete learning objectives for
each subject in each grade, and students were to be
tested on them before and after instruction was
given, and this progress was to be monitored carefully. Second, children were grouped according to
the progress they had already made (pretest) and
instructed from where their knowledge left off.
These skill groupings were to be fluid and reconfigured when a new objective was introduced.
The school kept its small school layout and
each school had 4 homerooms that travelled
through all the same classes together. Teachers in
each school had an hour a day for common planning and the lead teachers met with the principal
as an instructional improvement committee that
enabled two-way communication between teachers
and the principal.

Upon opening, Adams Avenue attracted

well-educated parents (mostly for the gifted program) who had an influential role in the schools
11

School Character

Metz reports that teachers focused their attention on their work with students. Their energy was
directed toward planning and teaching, running
lots of extra-curricular activities, and so on.
Rather than speak of students in terms of IGE,
they spoke of them in terms of their relationships
with the students. The school was notable in that
potentially volatile relations were not evident, and
instead positive relations persisted between the faculty, parents, and students. There were exceptions
that suggested a harder past (in years 1-2), but the
school was mostly in harmony by year 3.
Classes were heterogeneous in composition,
but as stated earlier, they were internally divided
into groups on the basis of skills development with
relation to each learning objective. Lower skill
groups had more African Americans, but they remained relatively heterogeneous and the interactions between students and teacher with students
were task-oriented and respectful for the most part.
Students themselves reported having interracial
friends and seemed open to heterogenous relationships.
There was a general absence of conflict at
Adams Avenue. Discipline was often a simple matter. Faculty issued yellow cards as warnings, and
then formal referrals to administrators for discipline which was noted in the childs record. Metz
reports that these yellow cards were issued less
than two times a day for all 300 children over the
course of the year, and suspensions totaled less
than 1 out of 10 kids. Disciplinary problems were
more common than these formal indicators suggest, but they were handled informally (and this in
turn reinforced positive relations).
If there was any conflict it was likely between the principal and some teachers. This conflict goes back to the end of year 1 when some
teachers did not strike with the rest (and sided with
the principal). The union leaders were especially
bitter over this.
The program in practice

The teachers and principal followed the two

sets of practices believed to be the core of the IGE

program. Teachers charted progress and the principal checked it, but there was still some variation in
teacher compliance with IGE. Some teachers were
relaxed in their application of IGE. Some rotated
students through the same set of tasks in spite of
being in differently skilled groups, so they did not
have differentiated work. Some produced charts
on estimates of student progress rather than
pretest-posttest scores. These teachers said they
adapted IGE like this because it was a lot of work
(more than regular teaching) and they were unwilling or unable to do all of it. A few other teachers
resisted IGE. They didnt comply as a matter of
principle. They argued their subject matter was
ill-suited to IGE and required fundamentals, or too
many skill demonstrations. But even these relaxed
and resistant teachers were influenced they conveyed clear purposes for each days instruction,
they had relatively well formed understandings of
each kids skills and deficits. Even if they didnt
use the explicit features of the curriculum, they
seemed in-line with the general philosophy and focused on skill development in their subject,
thought carefully how to get that across to varied
kinds of students, and how to track progress. Instruction also involved a lot of field trips, projects,
and a rich extra-curricular experience. This added
a personal element.
IGE Influences on School Character

Metz reports that the imposition of IGE changed

the character of the school especially the relations of low achievers with teachers and students
of different races. That is IGE induced a communal ethos and denser positive relations.
IGE influences on Traditional Curricular Structure

The IGE curriculum removed grade-level differentiation from view. Instructional differentiation
was rendered more individualized, and it removed
both the stigma placed on a student performing at
4th or 5th grade level and enabled accelerated students to work at a level beyond grade level. All
that matters was forward movement for every kid,
not where they were moving forward from.

12

IGE Influences on Reward structure

Adams Avenue used report cards that emphasized

effort and the level at which the student worked in
each subject. Hence, a hardworking student with
5th grade skills levels may receive an I for superior effort and progress while the lackadaisical 6th
grader with 8th grade skills might get an E for
inadequate progress. The honor roll was based on
effort grades, not skill level grades. In this manner, IGEs reward structure worked to equalize social prestige and include lower performing kids
and give them academic legitimacy. This conversely lowered the rewards experienced by high
achieving students, and some teachers worried
these students were not pushed enough to excel
higher.
IGE influences on Classroom Task Structure and Relationships

All the instruction was done in groups based on

skill where the students worked independently.
This meant no one performed before everyone publicly, and achievement and schoolwork was more a
matter of private accomplishment and few opportunities for public embarrassment. Teachers spoke
with students as a group for instruction and then
guided progress individually. Metz reports than
everyone felt they got the attention and assistance
they needed. These relations built into ones of
trust between teacher and student and lessened conflict. They also equalized persons more, deemphasizing initial differences in skulls and this
served to build interracial ties.

units enabled the teachers to know students individually and have a healthy rapport with one another.
Faculty culture and school ethos

Faculty regarded good relations with students and

each other an end in itself and helpful to learning.
With few exceptions, teachers viewed all students
as essentially good children, and they regarded the
mutual rapport as normal. The teachers did not
misidentify with their students.
This faculty culture was rebuilt, passed on,
and renewed. This occurred in several ways during team meetings and faculty lounge conversations. Teachers tended to interject positive comments into conversations that spun in negative directions. New recruits got socialized through
these experiences so the culture was passed on.
Informal leaders respectfully sanctioned new teachers adopting a negative view of students. They redirected attitudes so as to be one of respect and
building students up.
That said, few cultures are uniform. There
were exceptions and Metz remarks on 5 teachers
angrily confronting students.
These teachers
tended not to use group instruction but rather
whole class and recitation. In addition, students
knew who they were and responded to them negatively. That said, Metz is quick to point out that
these teachers were relatively negative, but not noticeably so in comparison to say traditional schoolteachers and in other contexts. Her point is that
the school culture is a fragile construction that
needed to be reproduced and was far from a sure
thing.

Physical space

School location in the downtown area of the city

lent itself to field trips to businesses and the museum district. The building was small, lacked sufficient space for a gym, and its heating was not always certain. In spite of this they did not want to
move to another building. Metz argues the context
put everyone into shared spaces, and created more
of a warm atmosphere. All the teachers reported
that the school's small size and partitioning into 3

Leadership: Principal's influence

The principal Mrs. Michael, influenced the tone

of interpersonal relationships via indirect and informal means, but controlled the IGE curriculum and
its instruction via direct and formal means.
It was not official doctrine to have positive
relations with students. But the principal encouraged it in a variety of ways: in her speeches she

13

Main Story-Line (dominant pattern of inference)

Technology ! Structure in good way in spite of population disadvantage and potential for
divisiveness.

Organizational
Elements in Adams
Avenue School

valued building up students; she wanted relative assessments to occur

(over objective / universal ones); she wanted teachers to do field trips;
she encouraged ethnic pride and was involved in those groups; and
sought integration. She publicly appreciated teachers who led extracurricula and made it a point of giving them institutional resources they
needed for such endeavors. In short, the principals relations with faculty
and the students mirrored that of the school culture. Whether one influenced the other is not clear, but they reinforced each other for certain.
The principal's relation with faculty over IGE was a different matter. The IGE program was imposed from the district and the faculty felt
they had no choice or discussion over it and felt a degree of resistance.
Mrs. Michael's resorted to formal hierarchical authority to implement
IGE. In the faculty meetings of the first two years, she reminded teachers
they had to implement IGE or find a job in another school or district. At
the end of the first year, she even demanded 3 teachers transfer, and this
led to a lot of conflict. Eventually 2 were persuaded to leave and the 3rd
filed a grievance. Faculty were upset some since they felt the involuntary
transfer wasn't too fair, and that many didn't know how to implement IGE
the first year. By the 3rd year, teachers were more comfortable with IGE
and resisted less, and the principal resorted to more positive reinforcement and lessened her use of official powers.

14

References
Teacher resistance

A minority of teachers criticized the principal for

her reliance on hierarchical formal authority to
push through the IGE program. The minority's anger was recognized by the majority of teaches, but
it did not diffuse.
Summary

The distinctive feature of Adams Avenue was the

constructive relationships.
The school implemented the formal IGE program to a moderate degree, and the positive relationships seemed to reinforce the elements of IGE that seemed consistent
with it. For example, the aspects of IGE that rendered negative judgment private were reinforced;
the focus on individual or relative performance
was reinforced; and the effort to nurture individuals and relationships via supportive skills groups.
The pride of slow learners was protected, and special activities built a sense of fun and camaraderie.
The technological (task) arrangement of the school
did not work alone. It required a faculty culture
and school character that assumed respect would
breed further respect.
The lack of training and rush to get IGE going led the principal to use her formal authority
and to push IGE through. The principal believed it
was her choice to do this in response, and it was
not a pressure from the district office per se. This
pressure from the principal led the faculty to be resistant and upset at first. A minority remained
somewhat angry even, but the faculty and principal did find ways to work respectfully and productively together (again, partly a result of the small
schools and positive, collegial ethos). The teachers believed the small schools contributed to their
getting to know their students individually and this
was the secret to their success. They did not notice the contribution of their culture (students too)
or the technology (tasks). Their benign belief (unconscious even) seemed natural to them and the
culture operated at its best effect.

Leavitt, Harold J. 1965. Applied Organizational

Change in Industry: Structural, Technological and
Humanistic Approaches, in Handbook of Organizations, 1144-70, ed. James G. March. Chicago:
Rand McNally.
Metz, Mary Haywood. 1986. Adams Avenue
School for Individually Guided Education. Chapter 4 (pp. 57-103) in Different by Design: The Context and Character of Three Magnet Schools.
Routledge: New York.
Scott, W. Richard, 1995. Institutions and Organizations. London: Sage.
Scott, Richard. 2003 (1981). Organizations: Rational, Natural and Open Systems. 5th Edition,
Englewood Cliffs, NJ: Prentice-Hall.

15

2
Decision-Making in
Organizations

Source: http://commons.wikimedia.org/wiki/File:EXCOMM_meeting%2C_Cuban_Missile_Crisis%2C_29_October_1962.jpg

Decision-Making in Organizations
In this chapter, I will present a general introduction and discussion to decision-making in organizations. In the lecture, I will relate various rational system views of organizations that tend to
focus on administrative units, or leaders of organizations. A simple example of organizational decisions can be found in the following figure showing
a decision tree. The choice is whether to upload a
picture or not onto my Coursera course. A variety
of criteria apply and help us decide. Fortunately,
this particular image of a decision tree was taken
off Creative Commons, and freely viewed, so we
are okay. Nonetheless, it gives you an initial sense
of what we mean by decision-making.
Logic of Consequences - Rational Choice Theory

This week we draw heavily on the work by

James G.
March concerning decision-making
(March 1999; 1994: chapters 1-2). March has
spent several decades studying actual decisionmaking behavior in organizations. He classifies
types of organizational decision-making that helps
situate real world cases of organizations further
particularly the rational and natural classes of organizational depictions Richard Scott relates.
March describes two general classes of organizational decision-making, or logics of decisionmaking as he calls them: the logic of consequence (or rational choice theory) and the logic of
appropriateness (what Graham Allison might call
organizational process model, 1969). The core
distinction between these logics is that one is concerned with choices and instrumental efforts and
the other one is concerned with rule-following and
interpretive activity. Both are intentional forms of
behavior. The former entails means-end rational
action, and the latter entails value-rationality or
duty-driven behavior. Value-rationality contends
that regardless of the cost (or without attention to
them), we often make decisions.

Figure. Decision Tree

(Source - http://commons.wikimedia.org/wiki/File:)

Rational Actor Approach

The rational-actor model is essentially a

model that follows the logic of consequence. Rational decisions basically entails four aspects:
(1) The first is knowing your alternatives.
Here a decision maker asks, what are the
options available to me?
(2) Second, it is important to know the consequences of these alternatives. Here, one
asks, what happens if I take each option?

17

(3) Third, you have ordered preferences or

ranked goals and objectives in terms of
greater or lesser value. Here, you weigh
the value gained or lost by taking each option.
(4) The fourth aspect concerns a decision rule,
or a choice process (Graham Allison refers
to this as an inference pattern). Here,
the choice process is a rule by which an alternative is selected on the basis of its consequence for preferences. Two decision
rules are commonly discussed and reflect
different notions of a rational actor.
a. The first, is an ideally rational person,
traditionally called economic man by
its critics. This individual is typified
by clarity, and knowing, and consistency of preferences and objectives,
they're an ideal form of a rational actor.
b. The second is a boundedly rational
person, and is typified by ambiguity,
and uncertainty, and knowing incomplete information, and inconsistency in
preferences or objectives. So, here we
have individuals who are more like the
real persons we all know.

brella and it does not rain, then weI have to carry it

around all day, and there is a cost to that. And, if
we bring an umbrella and it does rain, then we are
prepared and stay dry. We may even feel pretty
pleased with ourselves!
Lets put these alternatives and consequences
in a matrix below, and then affix values from positive ten to negative ten so as to depict our preferences for them. Let's say that not bringing an umbrella and it does not rain is a positive six. We are
happy to not have to carry the umbrella all day.
By contrast, lets say that when we don't bring an
umbrella and it rains, we get wet and we find that
that terribly disconcerting. We can give that consequence a negative ten. On the other hand, say we
do bring an umbrella, and there is no rain, then we
have to carry it around. Carrying around a useless
umbrella all day is somewhat of an inconvenience,
so lets give that a negative five. And last, if we
bring an umbrella and it rains, then we are kind of
pleased with ourselves for being prepared, so we
value that a plus eight.

Dont&Bring&Umbrella&
Bring&Umbrella&

Rain((40%)(

Net(Expected(U7lity(

+6&

310&

(6*0.6)&&(10*0.4)&=&3.4&

35&

+8&

(35*0.6)+(8*0.4)&=&0.6&

)&
&(0.4
Rain

Examples of the Rational Actor Model

Lets use a very simple example of rational

choice in action, to see these aspects and distinct
types of choice processes. Soon it will be winter
in California, and that means it will be the rainy
season. As a result, many Stanford students will
be faced with the alternative of bringing an umbrella to class or not. Now lets say you see certain consequences to these alternatives. For example, there is a 40% chance of rain, so the consequence is 40% it may rain, and 60% it may not.
Now let's say we see certain costs and benefits to each scenario and we prefer some over others. In the case of not bringing the umbrella and it
doesnt rain, we rejoice because we did not have to
carry it all day (yea!). In the case of not bringing
an umbrella and it rains, we get wet, and it is no
fun sitting around wet all day. If we bring an um-

No#Rain((60%)(

a&
rell
mb
g&U
n
i
r
t&B
on

Brin
g&U
mb
rell
a&

EU=34.0&
Net&
Expected&
UMlity&=&3.4&

No&
Ra

i n &( 0
.6)&

)&
&(0.4
Rain

EU=3.6&
EU=3.2&
Net&
Expected&
UMlity&=&0.6&

No&
Ra

i n &( 0
.

6)&
McFarland&Lectures&

EU=33.0&

Figure. Decision Tree for Umbrellas

Now all of this can be readily illustrated in a
decision tree. The first branch in the tree lists our
options of bringing an umbrella or not. The second branch list the consequences where there is
40% chance of rain and a 60% chance of no rain.
At the end of each branch is the preference or

18

value we affix to these scenarios. In the logic of

consequence model, we calculate the expected utility of each scenario. To do this, all we have to do
is multiply the chance of rain (which is 40%) by
the preference we have for the scenario of not
bringing an umbrella and it rains (which is -10).
That gives us the first value of -4.0. Thats the expected utility of not bringing an umbrella and it
rains on us. But say it does not rain and we do not
bring an umbrella. Then we take the chains of no
rain (%60) and multiply it by the value we affix to
that outcome (+6). As such, the we have 0.6 times
6 = 3.6. That's the expected utility of not bringing
umbrella if it doesn't rain. If we add the two together of not bringing an umbrella in both cases
- then we get the net expected utility of not bringing an umbrella = -4.0 + 3.6 = -0.4.
If we go through the same kind of operation
in the lower branch for bringing the umbrella, we
will find the net expected utility to be 0.6. If we
compare the two, then it is clear that bringing umbrella - given our preferences or our sense of costs
and rewards for each outcome is better than not
bringing an umbrella because we really do not
want to be wet.
Now lets do this for a more interesting case
- dating! Many of you are single and perhaps looking for love. Say you are wondering whether to
ask someone out. Lets consider the scenarios. (i)
You do not ask them out when they would have
said no. That is good, right? You're not embarrassed! (ii) You do not ask them out and they
would have said yes. In that case, you miss out
on someone quite interesting and wonderful. That
is a downer. (iii) You do ask them out and they
say no. That is kind of, mortifying, right? That
may be terrible. (iv) And then, there is the last scenario which is you ask them out and they say
yes. When that happens it is quite gratifying.
How would you value each of these options from
positive ten to negative ten?
It all depends. Are you a high-interest, lowcost person? Meaning, you ask people out all the
time and you do not see much cost to it. Or are
you a low-interest, low-cost person? Meaning,
you seldom ask people out and you do not worry

about it. Or are you a high cost person? Here you

see it as risky no matter what happens.
Lets say you find it mortifying to be rejected, and you are a high cost person. We can depict this in the table you see here. (i) Not asking
someone else and them saying no, hey, that is
good for us. It saved us the trouble, so it is a plus
two. (ii) Not asking them out, and they would
have said yes - that is a downer. Lets give that
a negative eight. Pretty bad, but not terrible. (iii)
But then, asking them out, and them saying no is
just awful. We feel miserable over that, so it is a
negative ten. And last, (iv) us asking them out and
them saying yes is a plus ten and that couldn't be
better. Best of all worlds right there!

Dont&Ask&Out&
Ask&Out&

No#(90%)#

Yes#(10%)#

Net#Expected#U4lity#

+2&

.8&

(2*0.9)&&(8*0.1)&=&1&

.10&

+10&

(.10*0.9)+(10*0.1)&=&.8&

)&
(0.1
ept&
Acc

Do n

t&
&Ou
&Ask

Ask
&O

ut&

Reje
ct&(0
.

p
Acce

EU=&.0.8&
Net&
Expected&
UClity&=&1&

9)&

)&
t&(0.1

EU=&1.8&
EU=&1.0&
Net&
Expected&
UClity&=&.8&

Reje

ct&(0
Ambiguity)or)uncertainty)
.9)&
about)consequences)and)costs?))

EU=&.9.0&

Figure. Decision Tree for Asking Out

If we go through the decision tree again, we
can predict the net utility of each option of asking
someone out or not. Lets even say they are very
attractive so our chances are low at 10%. If we go
through the math again like before where we don't
ask them out and get a yes, that equals negative
eight. Then we multiply that by the probability of
yes at 0.10 (10% chance). As such, negative 8
times .1 = -0.8 expected utility. The opposite of
not asking them out and they reject you has a positive utility of 1.8. So, we have a net expected utility of not asking people out equal to one.

19

ploring alternatives until it is good enough to satisfy it.

In the model below we look at the dating example again from a satisficing perspective. Here
we have to choose from ten different people, and
we have to consider the expected utility of asking
them out and them saying yes. Rather than discerning the consequences for each, we begin with
the nearest persons and move further afield (A to
J), only deciding to ask them out as soon as we
reach a person above our expected utility threshold
of say, three. As soon as we hit that three we have
our point of choice.
In the figure we can see that threshold out
here on the vertical line. A was not good enough,
and then B and C are not either (< 3). When we
hit D, we find they are good enough, so we stop
searching. Because we have not considered every
option, we have not optimized our decision. After
all, H and J, are two individuals who we would
have selected if we had considered them. In fact,
we could have found a more optimal choice, especially in J, who has the highest expected utility.
That is considered a satisficing decision, and it is
an example of, of how boundedly rational models
can be performed in a logic of consequence way.
Order)of)Choices)

By contrast, if we actually ask attractive people out, then given the probability that they will
say no and that we would be mortified, we have an
unexpected utility of negative eight. That's pretty
severe, so of course we just avoid the whole effort
altogether.
Thus far, we have related two simple examples of decision trees. You can extend this to organizations, their types of decisions, and their
kinds of options. For example, if a company does
X, then a competitor or client has a probability of
reacting in a certain way. Later, the Cuban Missile
Crisis will be discussed as an example of this. In
that case there are clear choices, potential consequences, and preferences affixed to each one.
That will bring this closer to a real world organizational case.
In all these cases there is clearly a ton of ambiguity. Weather reports are not that accurate,
plus, we really have little evidence to go on in deciding if someone might be receptive to being
asked out or not. Thus far, the rational actor
model is an idealized model that assumes herculean abilities of decision makers. In reality, most
of us are boundedly rational.
So what would a bounded rationality model
look like? Whats the choice process there? There
an actor is uncertain about consequences and
costs. Moreover, the ordering of preferences is not
so clear. To depict this, Herbert Simon related a
theory of satisficing as a potential alternative, one
that may offer a more accurate description of how
we usually make decisions as boundedly rational
persons. Instead of calculating all the alternatives
(would we ever really ask out everyone in a
room?), we start with one that is most near us e.g., not bringing an umbrella or not asking someone out like we always do - and then we see if that
option has a satisfactory consequence.
In most instances of satisficing behavior, we
think about a choice threshold, and we stop somewhere along our sequential search of options when
we find a choice that is good enough. But if we
do not meet our threshold, then we move on to the
next option down the list. So search is stimulated
by a failure to achieve a goal and it continues ex-

A)

Threshold)

EU=)310)
C)

E)

B)

EU=)34)
EU=)35)

EU=)310)
G)

F)

D)

EU=)3.0)

EU=)0)
EU=)31)
I)

Point)of)Choice)
H)

EU=)3.3)

J)

EU=9.5)
EU=)10)

McFarland)Lectures)

Figure. Satisficing Decision

Logic of Appropriateness

So far, we have discussed the logic of consequence, or rational actor models. But there is a second class of models, or a second class of decision
making, that March relates. He calls it the logic
of appropriateness. Most of the time in organizations, people follow rules even if it is not obviously in their self-interest to do so. For example,
when we follow orders in war and march to our

20

death, it seems hard to see much utility in that! In

those instances, we merely follow rules like duty.
And yet, a lot of behavior in organizations (and social life) is specified by rules - take for example
how we follow rules in tasks, routines, professional standards, norms, or standard operating procedures. When a problem or issue confronts an organization, it often becomes a question of finding
the appropriate rules to follow. Instead of valuing
alternatives in terms of their consequences, rulefollowing matches situations and identities.
Rule-Following

Let's take a moment to think about what this involves. Three factors are involved in the sort of
rule following that characterizes the logic of appropriateness.
(1) Situations are classified into categories associated with rules and identities (roles).
What kind of problem is it? Who usually
addresses it? How has it been addressed
in the past?
(2) Decision-makers have official identities
and roles that are evoked in particular
situations. Who usually addresses this
kind of stuff? Who's the appropriate person?
(3) Decision makers match rules to what they
see as appropriate to their role in the classified situation. They match rules and
identities to kinds of situations. They
say this is an x situation for y people to
manage.
One notices rule-following and the logic of appropriateness being used in organizational decisions whenever people follow traditions (path dependence), hunches, cultural norms, advice of others, pre-existent rules or standard procedures, and
heuristics (like rules of thumb). Decision-making
via rules can be as ambiguous as decision making
by means-end calculation. However, the ambiguity here does not concern consequences and preferences, but rather a lack of clarity and ambiguity in
agreements, experience, imitation, and change. In
addition, rule-following is a less conscious form of

decision-making than means-end rational calculation. Rule-following behavior is intentional behavior, but the type of inference being performed is
frequently implicit and taken-for-granted. It is intended action we do not reflect deeply upon.
When ambiguous, the rule-following process is
less about finding a desired outcome than making
sense of situations and discerning what rules apply
and why (e.g., sense-making and meaningmaking). The primary product of decision-making
may be less the decision outcome, than the decision process establishing social meanings and the
identities of participants. So one can say here, the
decision process or theory explaining organizational dynamics suggests they do not necessarily
arise for reasons of improving consequences but
for engaging in a meaningful process. This process will be most evident in 4th-6th chapters of this
textbook when we discuss processes of organized
anarchy, organizational learning, and organizational culture.
March also alludes to the fact that both the
logic of consequence and appropriateness get further complicated when one considers that most organizations are composed of multiple actors with
inconsistent and often conflicting preferences /
identities. Here the theory of coalitions comes
into play as does the negotiation and bargaining
process (This will be akin to Graham Allisons
Bureaucratic Politics Model and reviewed in the
next chapter). March suggests that a two-stage decision model is often inaccurately depicted: stage
1 is the process of bargaining and coming to consensus; and stage 2 is the decision when understandings are executed. Unfortunately, these two
stages are seldom discrete. There are many compounding decision moments and consensus waxes
and wanes. The setup of a system and its implementation are intertwined. Hence, the world of alliances is not one of precision and formality, but
one of informal, loose understandings and expectations.
Last, in his reference to temporal orderings,
March evokes the theory of Organized Anarchy
(Garbage Can Theory), which is the depiction of
decision making from a fully dynamic perspective.
Here we are just remarking on these theories in

21

passing decision in coalitions and organized anarchies but please note them, as we will come back
to them over the next few chapters.

use battlefield nuclear weapons to defend Cuba if

invaded. Fortunately, war was averted.

Case - The Cuban Missile Crisis and the

Prospect of Armageddon
Now that we have some idea of Marchs logics and passing references to culture, coalitions,
and anarchic decision environments, we can turn
to Graham Allisons study of the Cuban Missile
Crisis (Allison 1969). Why the Cuban Missile Crisis? It has lots of nice qualities applicable to nonprofits and government agencies. In fact, crisis
management is common in organizations. And in
many instances, the stakes of policies and decisions are enormous. Take for example the American effort to reform public schools (No Child Left
Behind Act) and the crisis schools are confronting
in order to meet standards (many are being
closed!). Similarly, many organizations face punctuated crises, like NASA facing the Challenger disaster, or human resource departments coping with
an onslaught of harassment and grievance claims,
or companies coping with deaths or massive
worker turnover. In these circumstances, what do
you do? How can you describe what happened?
How can you successfully manage in such a
situation?
More importantly, the Cuban Missile Crisis
was a huge event. It was arguably the closest we
came to World War III when well over 100 million
people could have died. In fact, John Kennedy,
the president at the time, is quoted as estimating
the chance of failure at 1 in 3 or even 1 in 2. That
is a little too close to armageddon. Because of
this, analysts want to understand how national governments and their organizations maneuvered the
crisis. They want to get a better sense for how to
prevent disasters in the future, and to possibly manage these crises better.
Let me give a brief summary of the Cuban
Missile Crisis so as to familiarize everyone with it.
The events occurred back in 1962, and they led the
United States to be on it highest state of war readiness ever, and readied Soviet field commanders to

Figure. U2 Planes
(Source http://commons.wikimedia.org/wiki/File:U2_Image_of_Cuban_Missile_Crisis.jpg)

A little context may help back in 1962, the

Soviet missiles could only reach Europe while
U.S. missiles could reach the entire Soviet Union.
Upon meeting with Kennedy at a summit, Soviet
Premier Nikita Khrushchev walked away thinking
little of Kennedy as a statesman and thought he
might have the upper hand.
In April 1962,
Kruschev thought the placement of intermediaterange missiles in Cuba could deter a potential U.S.
attack against the Soviet Union. Fidel Castro, on
the other hand, was worried the U.S. would attack
again after its failed effort in the Bay of Pigs
(1961).
Fidel approved Khrushchev's plan to
place missiles on the island and saw them as a deterrent to a US invasion of Cuba.
In the summer of 1962 the Soviet Union began to secretly build its missile installations in
Cuba. The crisis began for the United States
around October 15, 1962 when U-2 reconnaissance planes photographed Soviet missiles under
construction in Cuba. President John Kennedy
was informed of these installations and he convened the EX-COMM, a group of his twelve most
important advisors. EX-COMM met for seven
days and Kennedy decided to impose a naval quarantine around Cuba. Quite a few key actors were
in this group, from Robert Kennedy, who was the
Attorney General; Dean Rusk, the US Secretary of

22

State, George Ball, who was the undersecretary of

state, John McCone; George Bundy, the National
Security Advisor; Ted Sorenson, special counsel to
the president; and Robert McNamara, a very important figure who was pretty domineering in the
meetings, and was the security of defense. And
then Llewellyn Thompson, ambassador at large,
the former US ambassador, to the Soviet Union,
who was the only Russian expert on EX-COMM.

to sea, because Kennedy and his Naval commanders were worried about mistakes, and boarding any
craft that might trigger a nuclear war. Tensions
were pretty high at this point, and Kennedy raised
military readiness to DEFCON two on the 25h.
On the 26th EX-COMM received a letter
from Khrushchev proposing the removal of Soviet
missiles and personnel if the US could guarantee
they would not invade Cuba. On October 27 a U2 was shot down over Cuba and EX-COMM received a second letter from Khrushchev demanding the removal of U.S. missiles in Turkey in exchange for Soviet missiles in Cuba. At this point
the Trollope ploy was done where the United
States responded to the first letter accepting the
conditions and both sides largely agreed. It is
kind of an interesting ploy and an effort to get an
advantage in a compromise situation.
Tensions eased on October 28 when Khrushchev publicly announced he would dismantle the
installations and return the missiles to the Soviet
Union, expressing his trust that the United States
would not invade Cuba. Further negotiations
arose to implement the October 28 agreement and
the US secretly removed missiles from Turkey.

Figure. Naval Blockade

(Source http://commons.wikimedia.org/wiki/File:P5M_VP-45_and_DD-835_with_Foxtrot_sub_

Allisons Three Models

at_Cuban_Missile_Crisis_1962.jpg)

The Cuban Missile Crisis is a case of an international crisis that almost led to war, and is describes the kinds of actions that occurred during
this conflict. Graham Allison does this interesting
thing which is very emblematic of the this courses
ambition in teaching you to apply multiple theories to the same phenomenon, and acquiring very
different perspectives of it. In so doing, he comes
to a deeper understanding of what happened that
informs policy experts and persons in such crises.
Allison presents three models that he thought
were the most useful, prevailing models at the
time: (1) the rational actor model (i.e., logic of consequence model), (2) the organizational process
model (i.e., logic of appropriateness model), and
(3) the coalition model (multiple actors with inconsistent preferences).

On October 22, Kennedy announced the discovery of the missile installations to the public and
his decision to quarantine the island. He also proclaimed that any nuclear missile launched from
Cuba would be regarded as an attack on the United
States by the Soviet Union and demanded that the
Soviets remove all of their offensive weapons
from Cuba. Once the crisis was public, tensions
grew. Kennedy ordered low-level reconnaissance
missions once every two hours. On October 23,
Khrushchev wrote Kennedy stating that the quarantine constituted an act of aggression, propelling humankind into the abyss of a world nuclear missile
war. On the twenty-fourth, Russian vessels turned
away from the blockade so, they saw eyeball to
eyeball, as Dean Rusk said. And then on the
twenty-fifth, the blockade was pulled out further

23

Rational Actor Model

(Logic of Consequence)

When we use the rational actor model, we

assume every choice option or alternative (serves a
purpose, and we predict their consequences and
our preferences for them (costs). Looking at the
table below, we see these basic organizing concepts. Lets deconstruct what they mean for the
Cuban Missile Crisis.
From Allisons retelling, we know the actor
is a unified national actor, or the United States.
The problem that motivates US action is that the
Soviet Unions placement of missiles in Cuba
needs a response. Now, if we then look at action
as a rational choice, then we have to break it down
further. So here, we look at the goals and objectives, and the clear objective here is national security. The options and their consequences, basically
concern the courses of action available and the results that could arise from adopting them.
So, let's think about what those were in the
case. First, one option is do nothing. And there's

Rational Actor Model

a cost here -- The Soviets outflank the early warning system, they reverse the United States advantage of power at time, America loses credibility in
Europe, and so on. Second, we have another option which is, we can make a diplomatic response.
And the cost here is that the UN veto is probable
because the Soviets hold a seat. Time matters, and
the missiles are already deployed, so we cannot
really wait. A third option is that we approach Castro. And, the cost here is that the Soviets are in
control of the missiles in Cuba, so Castro's influence is somewhat moot it seems. The fourth option is to invade, and the costs here are that the Soviets could parallel with an invasion in Berlin, or a
retaliatory strike is possible with nuclear weapons.
A fifth option is an airstrike, and here the cost is
the probability of knocking out all the nuclear
weapons is only 90 percent since they are spread
out all over the island. Moreover, retaliation is, is
highly likely and a massive strike would be
needed to make that option succeed. So there's big
risk there. The sixth option is a blockade. The
cost of the blockade is that they could retaliate
with a blockade of Berlin. The benefits are that
you get extra time, and Khrushchev has time to
think and consider that a nuclear holocaust is possible. And last, a naval engagement in the Carib-

(Adapted from Allison and Zelikov 1999:391)

The Paradigm

Model 1
National government

Black!box! National state

labeled! Generic state

Identified state
Personified state

Basic Unit of Analysis

Organizing Concepts

Governmental action as choice

Unified National Actor
Problem motivates action
Action as a Rational Choice
Goals and Objectives
Options and Consequences
Choice

Dominant Inference
Pattern
General Propositions

Action = value maximizing means towards states ends

Increased perceived costs = action less likely
Decreased perceived costs = action more likely

!
24

bean actually favors the United States in this circumstance.

So, what was the actual choice? If we did a
decision tree of all these things and we looked at
value maximization, it would reveal that the blockage is the solution. But, why? Well, consider the
decision trees earlier. If Armageddon occurred,
then the costs to that consequence are so high that
even if it is highly improbable, it is likely that the
actors will not select it as their choice. Therefore,
through the rational actor model, we can somewhat interpret the series of events in the eventual
decision or choice that was made.

How does an organizational process model

apply? There are multiple organizations involved,
each with identities and standard operating procedures for handling aspects of the problem. The actors are a constellation of loosely allied organizations. In addition, the problem is not confronted
as one thing. The problem is cut up and parceled
out to various organizations (matching and the
logic of appropriateness!). Think here for a moment: if we are limited problem solvers, then organizations develop the capacity to do it better and
by experience. To great extent that is why we rely
on organizations. We cue them to do things they
have always done and are good at. By doing that
we accomplish desired outcomes.
Lets go down the list of organizations here.
What are their missions and capacities? The organizational process model will see each organization
as quasi-independent, and they are going to conduct affairs according to their own missions (e.g.,
the navy, air force, etc.) and capacities. This leads

Organizational Process Model

(Logic of Appropriateness)

Organizational Process Model

(Adapted from Allison and Zelikov 1999:391)
The Paradigm

Model 2
National Government
Leaders!

Basic Unit of
Analysis
Organizing
Concepts

B!

C!

D!

E!

F!

A!

What actually occurs; Range of choice; Structure of situation; Innovation

G!

Governmental action as organizational output

Organizational actors multiple!

Problems - divided up parceled out to various organizations (Matching!)
Organizational missions independence & parochialism!
Action as organizational output
Objectives compliance; Sequential attention to objectives; Standard operating procedures;
Programs and repertoires; Uncertainty avoidance; Problem-directed search
Organizational learning and change
Central coordination and control
Decisions of government leaders
Dominant Inference Action (in short run) = output close to existing output
Pattern
Action (in long run) = output conditioned by organization view of tasks, capacities, programs, repertoires, and
routines
General
Existing organized capabilities influence governmental choice
Propositions
Organizational priorities shape organizational implementation
Special capacities and cultural beliefs; Conflicting goals addressed sequentially
Implementation reflects previously established routines, SOPs, programs and repertoires
Leaders neglect administrative feasibility at their peril
Limited flexibility and incremental change
Long-range planning
Imperialism
Directed change

25

to organizational parochialism where each organizations conducts its affairs according to its own interests and defines success by whether they meet
those objectives. To accomplish objectives, organizations rely on standard operating procedures
(SOP), which means they have built in routines
they tend to train with and follow repeatedly, and
they get good at them. Larger programs are then
clusters or repertoires of SOP (e.g., fighting entails
multiple SOPs). Organizations attempt to reduce
uncertainty by ignoring details, having regularized
contact, and conventionalized means of processing
information. However, this all leads to distorted
information. Organizations also perform problemdirected searches, whereby each search is guided
by available and familiar organizational routines.
Coordination and control across different organizations and their SOP clusters is always an issue
(e.g., how do you get the Air Force and Navy to
coordinate their activity?). Executives merely call
into play different organizations and their SOPs.
Let me give an example: It took a long time
for the report on sighted nuclear missiles to reach
to president. This information was lost in tons of
inaccurate information, and the transfer of the actual message took a long time because it followed
standard operating procedures. The first photos
were taken on September 12, over a month before
the actual report was made to the president. On
September 19, analysis of the photos suggested the
presence of the missile silos. On October 4 they
began to believe there were missiles there. At that
point, there is territory dispute between the Air
Force and the Central Intelligence Agency (CIA).
In addition, there is a mechanical delay fro ensuing U-2 flights over Cuba to confirm these suspicions. Only on October 14 was there a flight that
confirmed their presence of missiles and that is
used to inform the president.
Another example: The EX-COMM leaders
are organizational representatives. When each one
is asked their opinion, they respond as organizational representatives and state what a representative of their organization could do. The Air Force
is a proponent of an airstrike and the Navy a blockade. That is said even when there are clear problems with each proposal. The Air Force could not

guarantee success, and Kennedy (a World Ware II

veteran) did not like the idea of enacting a Pearl
Harbor on another nation. Similarly, the Navy
could do a blockade, but they did it the way they
trained to do it - 500 miles out - instead of 180
miles off the coast as Kennedy commanded. This
was even done after the President got angry over
it. Simply put, the Navy invoked their SOPs.

Figure. Surveillance Photo of Missiles Being

Loaded at Port
(Source http://commons.wikimedia.org/wiki/File:Soviet_missile_equipment_beeing_loaded_at_po
rt_in_Cuba_1962.PNG)

Bureaucratic Politics Model

The bureaucratic politics model asks the following: Is the government composed of multiple
actors with different problems and objectives? Is
the choice an outcome of bargaining games that
unfold over time? Was power and skill a factor
that was involved? What compromises were had?
What overlapping games were being played?
Who were the leaders, followers, staffers, and ad
hoc players?

26

Multiple players were there with different

perceptions, priorities, and focused on separate
problems. For example, the Air Force and Army
had very different views of the atomic bomb. The
Air Force saw it positively due to their success in
using it in World War II. However, the Army saw
it negatively due to their experience of it in Japan
while on the ground. Of course, all of these are
relative judgement. All of these players contribute
to the coalition arrangement, and they are compiled over time into different outcomes. Had different players been involved, the outcome of the
Cuban Missile Crisis might have been different.
The key features of the Bureaucratic Politics
Model are the points of leverage, the personalities,
and various interest coalitions. How players negotiate, posit claims and thwart / work for them, is
how these temporary agreements arise and force a

decision. For example, take the actors and their

stances: Kennedys weak spot was Cuba due to the
Bay of Pigs fiasco. So he was very vulnerable
there. Re-election was key to him and he could
not fail on Cuba again. The military, on the other
hand, wanted to reprise the Bay of Pigs but this
time succeed.
What arose where two coalitions or viable options adopt the blockade or perform an airstrike.
One coalition was formed when the defense secretary voiced that a holocaust could be a potential
result i.e., the President, Robert Kennedy, Robert
McNamara, and Ted Sorenson are all for the blockade. In contrast, the six chiefs of staff including
McCone, Rush, Nitzke, and Acheson, wanted an
air strike. This coalition fell apart due to a lack of
guarantee, the problem of retaliation, and Kennedys concern of mirroring Pearl Harbor. So the
bureaucratic politics model assumes a variety of
views and actors alignments create different camps
that duke it out.

Bureaucratic Politics Model

(Adapted from Allison and Zelikov 1999:391)

The Paradigm

Model 3
National Government
A!!!!!!!!!!!!!!!!!!!!!!!!!!B!
!!!z!!!!!!!!!!!!!!!!!!!!!!!!!!!r!
!!!y!!!!!!!!!!!!!!!!!!!!!!!!!!!y!
C!!!!!!!!!!!!D!!!!!!!!!!!E!
!!!n!!!!!!!!!!!!t!!!!!!!!!!!!!z!
!!!x!!!!!!!!!!!!y!!!!!!!!!!!!!f!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!F!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!p!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!r!

Basic Unit of
Analysis
Organizing
Concepts

Dominant
Inference
Pattern
General
Propositions

Players in positions (A-F)

Goals, interests, stakes and stands (n-z)
Power
Action-channels

Government action as political resultant

Players in positions
Factors shape players perceptions, preferences, stands
Parochial priorities and perceptions; Goals and interests; Stakes and stands; Deadlines
and faces of issues
Power
What is the game?
Action-channels; Rules of the game; Action as political resultant
Governmental action = resultant of bargaining
Political resultants
Action and intention
Problems and solutions
Where you stand depends on where you sit
Chiefs and Indians
The 51-49 principle
International and Intranational relations
Misexpectation, miscommunication, reticence, and styles of play

27

28

Unitary actor or team that confronts a problem, assesses

objectives (goals) with regard to it, identifies options, the
consequences of said options, and then chooses option that
minimizes costs.
Variant: Bounded rationality and satisficing. Recognize
imperfect info, ambiguity, and select first satisfactory option
(good enough).

Summary or Basic
Argument

Action = Maximization of means to ends.

Know alternatives and their consequences for the shared goal,

and select wisely. Improve information and analysis.
Management by consequences.

Management
Strategies

Not salient except as influencing consequences of options.

Environment

Dominant Pattern of
Inference

Actors in hierarchical organizational positions. Cue sequential routines that

accomplish task or solve problem by routines available (supply issue).

Formal roles, hierarchical.

Know SOPs, what problems they go with (matching), and who cues them. Improve
rules and matching with problems. Management by rules.

Action = output close to prior output (path dependence), cueing of SOPs appropriate
to problem.

NA

Objectives compliance to SOPs, match with problem parts.

Organizational positions

Unified team or actor

Goals are defined in regard to problem.

Matching identity and SOPs (solutions) / programs / repertoires to problem.

Maximization of options (solutions).

Dividing up problem, coordinating / activating organizational actors who have special

capacities / SOPs for parts of problem, conducting sequential attention to objectives
(localized searches until problems resolved). Action guided by processes / available
routines.

Exists when the decision is guided by a logic of appropriateness matching problem

to actors with procedures for handling it (routine-process focus).

Organizational Process (OP) / Limited Problem Solver (LPS)

Goals
(what probs to
resolve)
Social Structure

Technology (how
solutions get
decided)
Participants

Key Organizational Elements

Exists when there is a unified actor with consistent preferences,

lots of information, and clear goals (and time calculate).

When does it apply?

Rational Actor (RA)

Summary Table of Three Theories to Date:

lishing teacher accountability. Both phases and

both approaches are different means to solving the
problem of low-achieving schools in Chicago.
The case relates key stakeholders and groups, their
interests and relations, as well as their responses.

Figure. Chicago Skyline

(Source http://commons.wikimedia.org/wiki/File:Chicago_Sears_Tower.jpg/78px-Chicago_Sear
s_Tower.jpg)

Case - Chicago Public School Reforms

The Cuban Missile Crisis is a case where an
organization is forced to make a decision (i.e., a
crisis). Another type of case concerns an organizational reform, and thats the sort of case we will
focus on in this section.
The case we review concerns the Chicago
public school reforms during the time period of
1986 to 2001. For this depiction, we draw heavily
on the accounts of Shipps (2003) and Bryk (2003).
The two works describe the reform efforts under
two Chicago mayors: Mayor Washington in the
early period (1986-1994) and then Mayor Daley in
the later period (1994-2001). As these two mayors
institute different reforms, there are shifts in how
the Chicago public schools are run. The basic
change that occurred was an initial effort at antibureaucracy, and then this was followed by a period of managerialism and reforms aimed at estab-

Figure. Mayor Washington

(Source http://commons.wikimedia.org/wiki/File:Harold_Washington_at_the_commissioning_of
_USS_Chicago_%28SSN-721%29.jpg)

Mayor Washington The First Phase

For the case, we made a summary table that

lists concrete examples of the organizational elements using Shipps and Bryks separate accounts
(Shipps 2003; Bryk 2003). These authors describe
two distinct periods of organizational reforms that
experienced a wax and wane. The first period under Mayor Washington implemented antibureaucracy reforms. The goals discussed concerned killing red tape, decentralizing power,
and empowering local experts in each neighborhood. As the reform moved forward, the political
and economic context changed, and there arose
other problems, like fiscal problems. In addition,
there were questions as to whether the killing red
tape and empowering local experts actually improved student achievement. Shipps argues that
the anti-bureaucracy reforms began to wain and

29

fail because people came to realize there was little

evidence for their success. Moreover, reform efforts at decentralization rendered the Chicago Public Schools inefficient, and it struggled to achieve
coordination in the face of a growing fiscal crisis.
There were a variety of key participants in
this early phase: the democratic legislature, Mayor
Washington, and the local school councils (LSC).
The LSCs were a conduit through which the decentralization of power and the empowerment of
local experts could occur. Teacher unions were
also a pretty powerful group in Chicago, as was
IBEC (Illinois Business Education Committee).
IBEC had been prevalent in local Chicago politics
for a long time. That committee was composed of
business leaders who mostly wanted the education
system to create workers they could use in the local economy. There was also school board nominating committee which nominated principals and
the like.
The social structure of participant interaction
was decentralized during this period. As such,
there arose a coalition across political parties and
interest groups that extended to local district and
neighborhood wards.
The technology and the tasks that were being
applied to accomplish this decentralization was a
reform effort to restructure governance of the education system. Power and budgetary discretion
was allocated to local school councils. This in
turn influenced the nature of educational legislation and standard operating procedures within the
school district during this period. Finally, the environment was one where Mayor Washington, as an
African-American, empowered, AfricanAmericans in the community and there was grassroots involvement in schools. This period was considered to be somewhat of a renaissance for community involvement in Chicago.
Mayor Daley

Toward the end of the decentralization phase

Mayor Washington suddenly dies and there is a
new mayor elected, William Daley Jr. Daley is the
son of a former Mayor who held office for several

decades. Daley is the mayor during the 1994-2001

period that Anthony Bryk writes about. In this period, there is a response to the prior eras shortcomings, and an effort to reform the system so it can
better respond to the fiscal crisis, problems of inefficiency, and the lack of evidence demonstrating
improved achievement. During the 1990s the
goal is to establish greater accountability and more
centralization within the school system. This new
phase of school reform emphasized a form of
managerialism, and it brought in business leaders
who were experts on running organizations efficiently.

Figure. Mayor Daley

(Source http://commons.wikimedia.org/wiki/File:thumb/7/71/Richardmdaley.jpg/703px-Richard
mdaley.jpg)

During the second phase there is a shift in

key participants. The Illinois legislature no longer
had a Democrat majority, and it had flipped to a
Republican one. There was a new mayor, and expectations around him were partly defined by his
fathers legacy as mayor. In addition, the school
system had been centralized so that the legislature
gave Daley the responsibility for running the system well. And Daley delegated management of
daily operations to a CEO for the district, Vallas,
and to the school board president, Rico. Certain
players were still present. For example, IBEC was

30

31

Decentralized power: coalition of strange bedfellows for

governance (across parties and interests)

Local School Councils (LSC) / de-centralized cash usage

decentralized structure, Legislation and SOPs (prior laws)

Larger African American role

Grassroots involvement

Technology

Environment

Reform kill red-tape,

Why goals changed: No
decentralize, and empower
evidence of success, little
local experts.
coordination, fiscal crisis
In: Democratic Legislature, Mayor Washington, Local School
Councils (LSC), Teacher Union, IBEC (IL Business Education
Committee), School Board Nominating Committee (SBNC)

1b (Wane-1994)
Creation of fiscal problems

Social Structure

Key Actors

Phases
ELEMENTS
Goals

1a (1986-Wax)
Anti-Bureaucracy

This time: Part I -- look for evidence of theories in the Chicago case (see handout).

McFarland Notes WEEK 2 Org Behavior and Analysis

2b (Wane-2001)
Problems of implementation

Reform accountability and

Why goal change: Draconian
centralization (tests, no social effect in implementation
promotion)
felt
In: Republican Legislature, Mayor Daley, Vallas (CEO), Rico
(School Board President), IBEC & teacher union leaders.
Out LSC, teacher union members, School Board Nominating
Committee.
Centralized power: coalition of loyalists that push others out.
Same structure of teaching, etc but new heads from outside
(business).
Legislation / appointments Teachers adapt teaching to
regulatory factors: decoupling
centralized structure.
of instruction to survive;
High stakes tests / regulatory
subvert reform goals for local
factors (probation, summer
preservation (cheating).
school, etc) scores go up.
Tests couple technical core to
social structure.
Past: Prior reform failed and
Confound Environment:
created other more pressing
Mayors Coalition makes info
problems.
non-transparent to
Politicians enter: Legislature
environment, manipulates
makes laws and shifts power
press decoupling to survive
to mayor!
and subvert reform goals for
Business enters: Mayor picks political success in
external staff (loyal to him)
environment (hide info! Or
and imposes it on system.
make cloudy!)

2a (1994-Wax)
Managerialism

32

Coalition /
Bargaining

Org Process

Phases
Rational Actor

THEORIES

1a (1986-)
1b (-1994)
Anti-Bureaucracy
Creation of fiscal problems
Actor: Washington & Chicago Public Schools
Problem: low performing schools
Action as RC:
Goal Reform schools and get rid of red tape
Options fight, form coalition of IBEC and LSCs,
Democrats and community activists dominate, etc.
Consequences no consensus, coalition forms and action
taken, legislature rejects (Republican minority rejects) /
no $ acquired, etc
Choice coalition of IBEC and LSCs arises
No data Shipps does not discuss this much - inferred.
Organizations: mayor office, legislature, Union, IBEC, SBNC
Problem broken up: LSCs break up problem of reform to be
handled locally.
Missions vary and LSCs move by their own SOPs without
coordination divergence of standards and costs soar.
Players and positions: IBEC and Community Activists
unlikely bedfellows.
Parochial Priorities: businesses want to implement
Reaganomics, and LSC sought greater control/power.
Goals / Interests: better city for business, better schools for
educators and students.
Deadline: mayor dies, new one elected, etc.
Game: negotiating legislation of power, then implementation.
Problem with decoupling and resource allocation.

McFarland Notes WEEK 2 Org Behavior and Analysis

Business leaders fall back on their own SOPs (managerial

program), not that of education system. Problem identified
using bus SOPs.
Disconnect in view of SOP, or error in using them (grade
equivalence scores), creates problems. Lack of ed experts
makes local level further decoupled and resistant.
Players and positions: Republican legislature and IBEC
propose reform give democratic mayor power
(undermines comm. activists). Mayor co-opts union
leaders
Parochial priorities / goals: IBEC goals are business focused
and concerned with econ model (educ model / community
activism failed), so impose & regulate (~IMF) as
environment / power shifts, so does emphasis on which
goals. Mayor reelection is a parochial goal.
Deadline: elections, contracts, fiscal years, etc.
Game: legislating power then implementation. In latter,
decoupling helps, couplingaccountability of power

2a (1994-)
2b (-2001)
Managerialism
Problems of implementation
Actor: Daley & Chicago Public Schools
Problem: broke, bad schools
Action as RC:
Goal Reform for results and accountability
Options Daley and Business dominate, prior alliance, etc.
Consequences $ gotten & action taken, legislature rejects /
no $ acquired / no action taken, etc
Choice new more centralized coalition forms

still there, but now it had more of a central role.

In contrast, the teacher union leaders were still
there, but now they were sidelined in the decision
making. Therefore, in spite of some consistency
in players, their political fortunes had shifted. Certain players fell out of favor: the local school councils, teacher union, the school board nominating
committee. They had less authority and were circumvented in decision making.
In the second phase, the social structure
shifted from a decentralized to a centralized structure. At the top was a coalition of managerial loyalists. Nonetheless, teachers still did the bulk of
the educational work and were called upon to implement managerial reforms entailing more testing
and accountability. The administrators did not
fully understand teachers or how the school system worked since many of them came from business backgrounds. The hope was that this new
form of managerialism could be imported into the
education system and improve it.
Several technologies - policies that transformed the system from one of local empowerment to centralized managerialism - characterize
this phase. The first of which was a change in governance. The legislature had control over budgets,
and they altered the rules so that budgetary power
and managerial appointments was given to the
mayor. Moreover, the state legislature made Daley
accountable to them. Daley responded by further
centralizing the education system, and as such, the
governance structure was transformed. Second,
new technologies like high stakes testing was imposed, and regulatory factors like probation and
retaking a grade level was placed on underperformers. Underperforming students had to go to summer school if they did not achieve at a certain
level. In response, scores went up, but that was
greatly the result of teachers now teaching to the
test. In addition, it was because many students
were retaking the test and for obvious reasons got
better as the test became more familiar. Nevertheless, high stakes testing served to recouple ground
floor educational efforts of teachers with the leadership efforts of managers. It rendered social structure related to the technological core of teaching
and instruction.

Toward the later period of the reform, there

were some unintended consequences to this. The
teachers adapted to regulatory factors, and they
started to subvert the reform goals to preserve their
jobs, students self esteem, and so on. Bryk describes how more and more teachers begin to cheat
on the test. In short, the workers began to adapt to
the reforms so that they could demonstrate success, even though that success was only in appearance and possibly untrue.
The environment changes in the second
phase. The context was one where a prior reform
(under Washington) had failed, so there is an effort
to correct that. In addition, new problems had
arisen -- pressing problems of fiscal issues. As a
result, politicians enter to deal with this crisis and
the legislature makes laws and shifts power to the
Mayor. Business concerns also enter because the
Mayor picks external staff from that community.
Over time, however, the reform wanes as the environment shifts again.
In particular, public opinion concerning educations reforms change. Initially, the public is for
evidence demonstrating improvement and a desire
to have a system that stays within budget. However, the Mayor's coalition makes information nontransparent. They claim successes even though
there's all kinds of other evidence to the contrary.
They try to manipulate the press in an effort to report good news and to hide the bad. This kind of
decoupling - managing the environment to survive
- even when the depictions are not fully true, begins to subvert reform. Goals for political success
(getting reelected) begin to eclipse those of authentic educational success, and the public becomes
aware of this. Contradiction arise in the press and
opinions shift.
So we have this interesting account of the
waxing and waning of two reform efforts during
this 15 year period. If we combine these phases
into one table it is useful to someone like me, and
possibly to many of you. But it is a pretty dense
table listing all the aforementioned organizational
elements of actors, goals, social structure, technology and environment. For the analyst, the systematic portrayal can be very helpful.

33

Applying Models to the Case

Phase 1 - Mayor Washington 1986-1994

Now that we have described all the organizational elements relevant to this case, we can begin
to consider their analysis. Let's begin again with
the first era of Mayor Washington and apply each
of our theoretical models to the case of Mayor
Washington, and his effort at reform. The first
phase entails an anti-bureaucracy movement that
waxes and wanes. If we apply the rational actor
model, we would focus on particular actors in administrative circles, like mayor Washington. The
core problem that commences the need for a decision is low performing schools. The goal would
be to reform the schools and get rid of red tape,
sine that was regarded as the problem that was preventing achievement and preventing buy-in to the
schools. The various options that they could consider were (a) fight this reform effort, or fight each
other in terms of how resources are allocated. For
example, the legislature could combat the unions.
(b) Another option is to form a coalition, such as
one forged between IBEC and the local school
councils. The Democrats could form a coalition
with community activists and dominate because
the democrats were in charge of local and state
politics at the time. With each option there are
likely consequences. If groups fight, then there is
likely no consensus and it just creates a difficult
environment for every side. If you form a coalition (IBEC and local school councils) and take action, then that makes some sense. There may be
certain kinds of cost to that and you may offend
the legislature in power. The legislature could reject everything. We can only estimate the likely
risks (we are boundedly rational!) and probabilities for success here, and use that in our decision
calculus. But conflict and stalemates have huge
costs, so we probably want to work with the Republican minority. for all we know , they could
filibuster any political coalitions efforts and stop
them dead in their tracks. That said, to some degree, it is reasonable to predict how certain camps
will behave depending on the kind of options before them, right?

The rational actor model seems to begin to

provide some insight into why the observed coalition of IBEC and the local school councils
emerges. It also helps explain why the grass roots
collaboration between businesses and community
level participants arises, and how an antibureaucratic, localized effort begins to take shape.
The rational actor model may be the most viable
model if we believe there is a true consensus, and
one group does not necessarily dominate another.
If we look at an organizational process
model, it is not clear this case writeup affords
much detail. Shipps and Bryk do not discuss standard operating procedures in much detail, so we
have to infer some of this.
If we apply the organizational process
model, we need to consider the organizations involved and what they do. The organizations involved are as follows: the Mayors office, the state
legislature, the union, IBEC, SBNC, and all those
actors you saw in that table before. Seeing this list
we can next imagine how the problem is broken
up and addressed by each group. The problem is
one of low performing schools and red tape that's
rendered schooling difficult to assess (i.e., consistently across schools), decision making inefficient,
and daily operations expensive. Given the actors,
the local school councils break up the problem so
they can handle it locally. Instead of it being a
Chicago problem, it is a series of distinct problems for each district. That's actually how the
situation unraveled. With each district, the goals
vary, and the local school councils adopt their
own standard operating procedures. When you decentralize a problem and coordinate locally, each
of these councils comes up with its own way of
dealing with things. As a result, you have a divergence of standards (one district worries about
achievement, another inequality, another about security and gang violence, etc). Solutions adopted
in one area are often not replicable in another community. As a result, costs soar as the complexity
of coordination across communities grows more
difficult. Decentralization creates local buy-in and
commitment, but it does not result in organizational efficiency.

34

If we apply coalition theory or the bureaucratic politics model, we have a different perspective as well. We have to think about the players
and their positions. So, IBEC and community activists have parochial interests. The IBEC wants
to support local business; the community activists
want their local communities to do well. As a result, they are strange bedfellows in a way (one republican and the other democrat). The community
activists want support for their neighborhoods and
to serve the interests of their local schools. IBEC
is composed of representatives from larger Chicago corporations, so their interests are not so local. Instead, they had the redevelopment of Chicago in mind as a means of not only making
schools better, but so as to have a potential pool of
more qualified employees, that will improve their
business. So, in spite of their different political
leanings, they had shared interests they could form
a coalition around.
Nevertheless, parochial priorities surfaced.
The business leaders wanted to implement Reaganomics, which was a trendy economic policy
of the 1980s -- it entailed less government, less
red tape, etc. Surprisingly, this aligned with the
parochial interests of local school councils that
wanted greater local control over neighborhood
schools. As such, in spite of their differences, the
two found overlapping interests, and this enabled
them to develop a coalition.
There is a deadline in all this, of course. In
the first period, Mayor Washington dies. It was an
untimely death and unexpected, and this led to a
new election and a shift toward recognizing problems with his reforms. In addition, there were different political games going on during this period.
There was a problem with decoupling and resource allocation to some extent. THe decoupling
arose because local school councils were given
power, and it was difficult to coordinate all their
decentralized efforts. As such, governance at the
city and state level decoupled from the local level.
When fiscal concerns arose and requests for accountability emerged, the decentralized system
failed to coordinate and they offered a cacophony
of responses. The decentralized system was great
for political games like power sharing and legisla-

tion of power, but not for the implementation of

top-down reforms or achievement assessments.
Discerning whether schools are performing better
or not required coordinate and standardization, and
the decentralized system was structured to do the
opposite.
Phase 2 - Mayor Daley, 1994-2001

Next, let's consider Mayor Daley's era.

Mayor Daley comes into office the same time the
state legislature in Illinois turns Republican and all
kinds of problems like a fiscal crisis start occurring in the state and city governments. As a result,
they had to decide where to place their resources
effectively.
If we take a rational actor view here, Daley
and the Chicago public school system were centralized actors. The school system has a much more
unitary actor under them than in the prior decentralized era. Unfortunate, the schools are broke and
they are not very good in spite of having decentralized governance structures.
Using a rational
choice model, the goal is to reform the system so
you get results (achievement) and there is accountability (support for what works and none for what
does not). Daley's options were that he could coordinate with business and dominate; he could fall
back on the prior alliance between local school
councils; and so on. And the consequences of
each choice, or each kind of option was different.
With one option, he's sure to get state money and
he might be able to take action. If he sticks with
the prior alliance with local school councils, then
the state legislature will likely reject his efforts
and hold back the allocation of education funds.
As such, it was clear he had to form a new kind of
coalition and make a new decision here. And the
choice was obvious, to go with a more centralized
coalition with business leaders and a means of accountability that was more efficient and managerial than the prior local decentralized school councils efforts.
The organizational process model provides
a different perspective. The business leaders fall
back on their own standard operating procedures.
That's why they came up with managerialism as an
approach. They didn't adopt the perspective or the

35

standard operating procedures of the school system. Instead, they adopted the kind of procedures
they had become accustomed to in managing their
businesses. The schools on the other hand hold a
different view and a different set of standard operating procedures. There is a disconnect in these
perspectives and procedures that creates tensions
and problems. The educators didn't understand the
standard operating procedures that business leaders wanted to impose on the education system. In
addition, the lack of educational experience in management led to kind of decoupling of understandings and unfamiliarity with educational routines
used to keep schools operating and teachers happy.
From an organizational standpoint, this later period of mismatch across organizations and or, organizational routines from different kinds of leaders helps explain the troubles encountered in the
managerial era.
Finally, a coalition or bargaining perspective
also highlights certain qualities of the case. From
a coalition-bargaining view, we see that certain
players and positions matter more than others.
The Republican legislature and IBEC propose reforms and they form a coalition with the mayor.
As a result of his participation, they afford him
power and resources. This centralization of power
completely undermines the community activists.
They're pushed out. In addition, the mayor coopts the union leaders in various ways (see Bryk
2003).
Parochial interests come through though.
Members of IBEC press the interests of the citys
business and economy. They feel that an educational model and community activism failed, so
their new model should work. They centralize and
regulate, and they emphasize distinct goals of
achievement and efficiency.
With each phase there's an election. The
new mayor falls back on their parochial interests
(re-election and getting power), and then as the reform moves forward, other kinds of parochial interest come into play and compete. As such, the bargaining model offers a more dynamic, political
characterization of the reform process. Organizations arent just actor optimizing or organizations
following rules, it's an a loose confederation of ac-

tors and organizations with shifting interests dependent on the timing and particular leaders involved.
A key factor in the coalition / bargaining
model is timing or deadlines. Here, of course,
deadlines are dictated by election cycles, contract
renewals, and fiscal years when budgets get done.
Those all have schedules with punctuated effects
on relationships and actors interests.
With each phase, the game shifts. In the initial period, the objective is to gain power. As
such, the new Republican legislature comes in and
wants to change the system for the better by centralizing authority. If they give money to the
mayor, then they can hold the mayor accountable.
If he fails, he will not be re-elected. In the later
phase, decoupling actually helps. If administrative
efforts are carefully linked to the ground level reform efforts, then the mayor is accountable. So
are the legislatures. But if things do not go well
and they cannot find results that validate the accountability model, then they start to hide it. And
this is exactly what happens., Through interactions
with the media and press reports, the mayors office and CEO of Chicago schools try to withhold
certain kinds of information that might show the
model was not effective. They do this for parochial interests -- i.e., the mayor is trying to get reelected, and so are other politicians. And so, it is
not just a matter of meeting the goals, or enacting
operating, standard operating procedures that fit
different groups. Now it becomes a matter of
adapting the implementation process to different
purposes and the reporting of their accomplishments for different purposes. Some of these purposes fit some actors more than others, and at certain times more than others.
In sum, each of our three models has applicability here. If we line them up, we can see how
they compare and which one seems to explain certain phases of the reform era more than others.
But now we come a big question -- which explanation works best?
This is a good case to use the forum on.
There, we can ask things like, does the rational actor model work better in a centralized phase like
Daley's? Or, does the rational actor model only su-

36

perficially apply? Or we can ask if most of the decisions followed an organizational process model
of heuristics and routines?
Rather than me telling you what the right answer is, I think it is best to leave it up to you and
see how you grapple with the issue. Trying to implement these theories to actual cases and seeing
evidence for one over the other is an exercise in
itself, as is arguing one theory works best under
particular circumstances and phases, or two theories complement one another in some way to afford a richer understanding of how Chicago public
school reforms manifested and died.

References
Allison, Graham T. 1969. Conceptual Models
and the Cuban Missile Crisis. The American Political Science Review 63, 3:689-718.
Allison, Graham, and Philip Zelikov. 1999. Essence of Decision: Explaining the Cuban Missile
Crisis (2nd edition). New York: Addison Wesley
Longman.
Bryk, Tony. 2003. No Child Left Behind,
Chicago-Style. In Peterson, P. W., and West, M.
The Politics and Practice of School Accountability, pp. 242-268. Washington, DC: Brookings Institution Press.
March, James G. 1994. A Primer on Decision
Making: How Decisions Happen. NY: The Free
Press.
March, James G. 1999. The Pursuit of Organizational Intelligence. Oxford, UK: Blackwell Publishers.
Shipps, Dorothy. 2003. The Businessmans Educator: Mayoral Takeover and Nontraditional Leadership in Chicago, in Powerful Reforms with
Shallow Roots, ed. Larry Cuban and Michael Usdan, pp. 16-34. NY: Teachers College Press.

Figure. Decentralized and Centralized Networks

(Source - Figure. Decentralized versus Centralized
http://commons.wikimedia.org/wiki/File:7/78/Decentralization.jpg)

37

3
Coalition Theory

(Source: http://commons.wikimedia.org/wiki/File:Sikh_Coalition_Im.JPG/)

Review and Coalition Theory

Thus far in the course, we have covered three organizational theories: Rational Actor, Organizational
Process, and Bureaucratic Politics. We applied these theories to cases like the Adams Avenue School, the Cuban Missile Crisis, and the Chicago Public Schools Reform efforts in the 1990s. If we line up these theories
side by side (see below), we can see how they compare. Much of this is review, so I will not cover it in detail
here. For the purposes of this part of the chapter, I just want to highlight how each theory implies certain management strategies.

Review of the Three Theories Covered so Far

Summary Table of Three Theories to Date:

When does it
apply?

Summary or Basic
Argument

Rational Actor (RA)

Exists when there is a unified actor with
consistent preferences, lots of
information, and clear goals (and time
calculate).
Unitary actor or team that confronts a
problem, assesses objectives (goals) with
regard to it, identifies options, the
consequences of said options, and then
chooses option that minimizes costs.
Variant: Bounded rationality and
satisficing. Recognize imperfect info,
ambiguity, and select first satisfactory
option (good enough).

Organizational Process (OP) / Limited Problem Solver

(LPS)
Exists when the decision is guided by a logic of
appropriateness matching problem to actors with
procedures for handling it (routine-process focus).

Coalitions /
Bureaucratic Politics (BP)
Exists when there are multiple actors with inconsistent
preferences and identities, and none of whom can go
it alone without assistance of others.

Dividing up problem, coordinating / activating

organizational actors who have special capacities /
SOPs for parts of problem, conducting sequential
attention to objectives (localized searches until
problems resolved). Action guided by processes /
available routines.

Focus on the players occupying various positions;

their parochial interests (their conceptions of
problems and solutions); their resources (expertise,
money, people) and stakes in game; and bargaining
processes between them that establish agreements /
coalitions.

Key Organizational Elements

Technology
(how solutions
get decided)

Maximization of options (solutions).

Matching identity and SOPs (solutions) / programs /

repertoires to problem.

Bargaining, or playing the game (within its rules), or

political maneuvering.

Participants

Unified team or actor

Organizational positions

Players in positions

Goals (what
probs to resolve)

Goals are defined in regard to problem.

Objectives compliance to SOPs, match with problem

parts.

Parochial priorities, goals/interests, stakes / stands.

Social Structure

Formal roles, hierarchical.

Actors in hierarchical organizational positions. Cue

sequential routines that accomplish task or solve
problem by routines available (supply issue).

Coalitions enemy/friend

Environment

Not salient except as influencing

consequences of options.

NA

Deadlines and wider array of stakeholders.

Dominant Pattern
of Inference

Action = Maximization of means to ends.

Action = output close to prior output (path

dependence), cueing of SOPs appropriate to problem.

Action = result of political bargaining.

Management
Strategies

Know alternatives and their consequences

for the shared goal, and select wisely.
Improve information and analysis.
Management by consequences.

Know SOPs, what problems they go with (matching),

and who cues them. Improve rules and matching with
problems. Management by rules.

Bargain with players (log-roll, horse-trade, hinder

oppositions coalition formation, etc). Learn others
interests / weaknesses so you know how to manipulate
and win. Direct management of relations via
bargaining.

39

Review of Management Approaches Rational

Actor, Organizational Process, and Bureaucratic
Politics

As a manager using a rational actor approach, you will want to consider alternatives and
their consequences. You will want to improve the
quality of information you receive so you can
make a wise decision based on the consequences
you expect each option to have.
As a manager adopting an organizational
process approach, you will need to know what organizations are involved, what standard operating
procedures they have in place, and then assign
them pieces of the problem they are best suited to
address. Your job is to match pieces of the problem to organizations capable of addressing them
effectively.
As a manager adopting a bureaucratic politics approach, you are more of a negotiator. You
will identify the key players, learn their interests,
identify points of leverage and weaknesses so you
can successfully bargain with them, and then make
exchanges to acquire their support. You will work
relationships and alignments to your advantage.
So each theory implies a different sort of
managerial strategy. With that in mind, lets consider a new case and use it as a thought experiment
for trying out these managerial styles. With every
new example we consider, hopefully you will form
a more concrete sense on how to apply theories to
real world cases.
In this chapter, I want to take the example of
Hurricane Katrina, which hit the city of New Orleans in 2005. Hurricane Katrina was the costliest
natural disaster for the United States (estimated at
81 billion dollars in damage) and in its wake, over
1800 people died, and 80% of New Orleans was
flooded. Lawsuits were filed afterwards against
the U.S. Army Corps of Engineers who designed
and built the levee systems that failed, and there
was an investigation into the responses of federal,
state, and local governments, resulting in the resignation of Federal Emergency Management
Agencys director, Michael Brown, and New Orleans Police Department Superintendent Eddie
Compass. Several agencies performed well and

were commended the US Coast Guard and the

National Hurricane Center.

Figure. Hurricane Katrina

(Source
http://commons.wikimedia.org/wiki/File:6/67/FEMA_logo.svg;http://commons.wikimedia
.org/wiki/File:thumb/9/90/Hurricane_Katrina_GOES_August_29.jpg)

Figure. President Bush, Governor Blanco and

Army Corps of Engineers
(Source: http://commons.wikimedia.org/wiki/File:BushVitterBlancoPoint.jpg)

I know it is an event that has already happened, and there has been a great deal written
about it online. I also know we have 20-20 vision
in hindsight. But it is often hard to acquire adequate amounts of material on a case that reflects
the depth of knowledge that leaders and analysts
might have on an organizational crisis and decision they are immediately involved with. The information we can glean on Katrina from reading
online material begins to approach what I think involved participants or experts might have on the
ground floor. But the point is not to achieve perfection here. It is pedagogical. I want you to get
used to applying theories as lenses to cases.

40

Therefore, ;ets role-play and imagine we are

mayor Nagin -- what would we do? After reading
this textbook, you might want to put in your
toolkit as many theories as it can hold. After all,
peoples lives are at stake and you want to do the
best job possible:
Which theory would you use to help you prepare for the hurricane?
Which would you use to help you manage
the situation after it hit?
Applying the Models to the Case of Hurricane Katrina

As a rational actor, Nagin would consider

the problems and his goals with relation to them:
e.g., the storm is coming and will likely flood the
city and create problems he can only partially address. He has various options to treat the problem
do nothing, build up the levees better, evacuate
before, evacuate after, serve/protect all the while,
and drain/rebuild afterwards. He would think
about all the other actors involved (FEMA, Gov,
Army Cops of Engineers, Red Cross, Police, Fire
Dept, National Guard, etc). As rational actor, he
would assume his staff and others are on the same
side because they too see the costs or consequences of flooding and lacking a good response
any death toll is too much. By relating the consequences of various options (or not taking the ones
proposed) and identifying how the least cost in life
is accomplished, he should be able to get everyone
to mobilize and respond in an optimal way.
But we know people do not always have the
same goal, nor are they always motivated by consequences. Some actors and organizations may
think the walls will hold, others will think 10 or
even 100 deaths will not need a response. At the
other extreme, they may be so overwhelmed with
the flooding that they will act on other instrumental grounds (for example: the National Guard may
be flooded themselves!). Nagin may have to invoke identity expectations, notions of duty, - the
logic of appropriateness! From an organizational
process standpoint, he needs to start partitioning
the problem up so the appropriate organizations

with experience and SOPs are assigned to each

part the city has evacuation plans, etc that he can
commence and coordinate. He knows police and
fire departments will assist there. But will their
ability to perform SOPs remain if they are overwhelmed? What if their homes and families are
flooded too? Will they privilege their family identity? (So perhaps having police/fire family protection plans set is a very good idea as well as
drills to prepare fire/police for the worst) Also,
Nagin might know that some of these SOPs work
better in some neighborhoods (rich) than others
(poor) and can allocate more where needed to
make it work better.

Figure. Organizations Involved

(Source - http://commons.wikimedia.org/wiki/File:FEMA_logo.svg/;
http://commons.wikimedia.org/wiki/File:Hurricane_Katrina_President_Bush_with_New_
Orleans_Mayor.jpg
http://commons.wikimedia.org/wiki/File:Flag_of_the_Red_Cross.svg
http://commons.wikimedia.org/wiki/File:National_Guard_emblem.gif)

That said, as mayor, we might predict it is

likely the city be overwhelmed anyway, so we
need to appeal to other organizational actors who
coordinate a wider array of participants and relevant SOPs e.g., Governor Blanco, FEMA director, and the president of the United States. In looking to these seemingly peripheral organizations,
we begin to adopt a bureaucratic politics model.
We ask these other actors to commence SOPs that
are under their jurisdiction. We might even explain our reasoning via the Rational Actor Model
(costs). Here we would hope to get National
Guard support in evacuating remaining citizens
(helicopters), delivering needed supplies, and maintaining order. We will also need to appeal to the
Army Corps of Engineers so as to be ready with
equipment to repair any walls after the flooding
occurs.
But we know this may not work actors and
organizations have parochial interests. From a bureaucratic politics perspective, we know the Na-

41

tional Guard will have its own problems if

flooded; the Army Corps of Engineers will not
want to be blamed for faulty walls; FEMA will not
want to look inept or totalitarian; and the Governor will not want to have her authority circumvented by outside organizations. Hence, we can
bargain with them. What do we bargain? We can
offer our public statement that they have worked
appropriately and diligently (not neglectfully, or
worse, prejudicially), etc. In short, our theories
offer you ways of organizing and ways of getting
coordinated action. They are descriptive and feasibly prescriptive if you so wish.
All this is a caricature of course, but hopefully it gets you to think more about how to apply
theories to cases. Many of you are welcome to
consider this case in greater detail and how our
theories might apply. There is a multitude of information about Hurricane Katrina online. It is a case
well worth analyzing and especially since many
more hurricanes will hit the gulf coast and eastern
seaboard of the United States in years to come.
The same could be said for earthquakes, tornadoes, tsunamis, etc. Through the careful study of
cases and application of organizational theories, it
is feasible to plan better and improve our management of these recurring problems.
Why Theories Matter?

If you were an analyst or manager, why

would you want to learn these theories and apply
them? I see at least three huge benefits imagine
you are called into an organization to help them
with a problem. Your training in organizational
theories gives you a few useful skills: First, you
have a broader range of experiences. You know
other histories, examples, companies, and accounts that are different from your own personal
experiences.
Second, you have a systematic way of thinking about an organization and its problems.
Whats likely to happen is that the employer
brings you into their office and explains their problem: We have a problem with how the employees
relate to each other, and there seems to be a man-

ager who is really trying to drive a wedge between

everyone. You will hear that and understand that
this is a problem with regard to social structure
and that the current interpretation is that the conflict is intentional or driven by a particular actor.
Now you probably dont want to use such academic jargon to relate this to them, but you can recognize that this is an issue they see as focused on
certain aspects of the organization and has one
kind of explanatory logic applied to it. By relating
that back to them, you help them better understand
what it is they think they are seeing.
As an analyst trained in this class, you are
able to allude to other facets of the organization
other actors, their beliefs, influences from the environment, technologies, competing goals, etc. You
can also offer another form of explanation that
actors are just following SOPs and there is a conflict between those emanating from different units
of the organization. In this manner, you help the
client see things in a different way and most
likely, in a more useful way as well.
Most every organization seldom wants an
outsider to come in and tell them what to do. If
they do, it will likely fail in implementation. They
will want you to help them figure out what is going on so they can propose solutions on their own.
You can help with that process, and by placing
them as central actors in the decision process they
are more likely to adopt some kind of solution and
reform that resolves at least some of their issues.

Coalition Theory
In the second chapter of the course you got a
good sense for how the rational actor perspective
and organizational process perspectives differed.
Those theories nicely corresponded with Marchs
notions of decisions by the logic of consequence
and decisions by the logic of appropriateness. You
probably walked away with a less clear understanding of the bureaucratic politics model. This chapter will give a more elaborate depiction of that
model, and focus on its core process of exchange
and coalition formation. Within organizations,

42

you will frequently confront coalitions of interests,

and you will come to realize that collective action
and organizational reforms are impossible if you
do not build and manage a coalition to get things
done. Therefore, this is the theory of week 3: Coalition theory. To relate this theory, throughout this
chapter we will draw heavily on the writings by
James G March (1962, 1994: chapter 4) and Kevin
Hula (1999) concerning coalition formation.
To this point, we have covered three theories,
and each has certain shortcomings. For example,
the rational actor view assumes people have the
same goal and that is seldom the case in reality.
Also, many people are not motivated by the consequences of options. This leads us to an organizational / rule-following view, but this perspective
fails to take into account how peripheral organizations can matter, and how many of the routines being suggested and enacted have parochial interests
behind them. There is politics and change, and
rule-following is too static and path dependent to
catch it. Finally, we have the bureaucratic politics
model and here we see more nitty-gritty politics
driving decision coalitions. Unfortunately, we did
not discuss thoroughly enough how interests are
negotiated and how collective decisions are
reached. This week we will spend more time
elaborating and explaining how coalitions can be
managed. We will zone in on coalition dynamics
and its core process of exchange and negotiation.

Other coalitions can be interest group based,

where a variety of groups (or even distinct religious sects) coalesce around an issue of mutual
concern.There are even organizational coalitions,
where different agencies and organizations coordinate their provision of services due to a great deal
of overlap.

Figure: Senado de Chile

(Source - http://commons.wikimedia.org/wiki/File:Senado_de_Chile.svg)

Examples of Coalitions

Lets start with something simple. What are

some examples of coalitions? The most common
ones are political and international coalitions. For
example, in Chile, there are many political parties
as shown in the diagram of circles. Some of the
parties find mutual interest and gain from working
together, so they form a coalition like the coalition for change (all the blues) or the coalition of
parties for democracy (most of the rest).
Another example might be seen in this decent tree of Christian political parties in New Zealand. A coalition ends up being a temporary alliance into a unified party, but it doesnt last long.

Figure: New Zealand Political Parties

(source http://commons.wikimedia.org/wiki/File:ChristianPoliticsNZ.png)

We have already read some discussions of

coalitions in chapter 2 of the course both March
and Allison discuss them in various ways. In most
instances, coalitions are social systems led or organized by mutually inconsistent decision-makers
(Cyert and March 1963). Here it becomes difficult
to see the decision-makers as a team. Instead we
have a power-struggle or tenuous coalition that describes the decision process.

43

Looking back to our table of Allisons theories and the outline of the Bureaucratic Politics
Model, we see there are multiple players in different positions. There are various factors that shape
their preferences and stands e.g., their particular
interests, stakes specific to them, the goals of
someone in their position, and the deadlines rushing them to decide. Each actor has certain resources or power (things people want), and they
enter a game of exchange or bargaining according
to rules. The end decision or organizational action
is the result of bargaining across these actors.
March describes coalitions in a nice way that
compliments Allisons Bureaucratic Politics
Model and Hulas notion of coalitions (Allison
1969; Hula 1999; March 1962, 1994). His focus,
however, is more on the central organizing process
of a coalition the process that creates and sustains a coalition. March argues that scholars describe coalitional decisions or coalition formation
as following one of two processes (among others).
The first process which I want to relate as background, not as something you need to learn has
been called a power-struggle, and it has sometimes
been analyzed and operationalized as a simple
force model. A force model is an extension of the
expected utility calculations we did for individuals
last week, except this time we have multiple actors
with different preferences (or different values
placed on the outcome) and they are given different weight in making decisions. To calculate the
decision a coalition makes by the force model, we
simply add up the expected utilities weighting
those of more important or powerful actors.
For example, take our old example of the umbrella and the expected utility of bringing it or not
(or even the dating example of asking someone
out or not). If we calculate that for each of you
and then weight your score by the relative power
you have, then add everyone up we should have
our collective decision. The problem with this procedure is that power is depicted as a stable personal trait we can actually measure. But power is
not a personal trait (think triads [dependent notions of rank]) and it changes over time. It is also
a tautological model and explanation: when people
get what they want, power is seen as explaining

why they got it. Plus, power can refer to many different things persons acquire. As such, it is hard
to measure as one construct everyone agrees upon.
So the force idea of just extending our pure Rational Actor Model to sets of people does not
really apply well.
Exchange Model

A better way to study power-struggles is

through exchange models (Emerson 1962). Here,
collective choices are produced by voluntary exchanges e.g., trading and bargaining. This is the
central organizing process of Allisons Bureacratic
Politics Model and Hulas Incentives Theory or
what you and I will just call coalition depictions.
The process of exchange is relatively simple:
(1) Every actor enters into a voluntary exchange relation regulated by rules (Allisons rules of the game).
(2) Participants bring resources to the table
(e.g., money, property, information, skills,
access to others, rights, knowledge, etc
~similar to Allisons notion of power
Please also note that Scott [2003] describes these powers or control over resources as resting in various roles and positions in an organization -- ownership,
management (capital), worker expertise,
bridging roles, and rule-establishment of
external actors like regulative agencies).
(3) The process of choice is one where mutually
acceptable trades are arranged (within
the rules ~Allisons bargaining).
(4) Each actor trades trying to improve their position, fulfilling their preferences / identity
as best they can until no more legal or mutually acceptable trades are possible (this
presumes bargaining with no time limit
but Allison and Hula add the reality that
we often face time-limits).
The exchange process as described above is the
core of the bureaucratic process model and to coalition formation more generally.

44

How to Manage and Win at Exchange

As a participant in exchanges, your ability to fulfill

your preferences / identity depends on a few
things:
The first is your ability to control the rules.
Do you write the laws? Who defines the rules of
the game? Take for instance, Robert Caros account of Robert Moses in The Power Broker an
unelected bureaucrat who ends up controlling politicians. He does this writing himself into legislative rules and as the person who controls the transit construction budget (arguably the largest chunk
of the city budget). Through this, he controlled
the election of new officials, helped promote them,
and he tore up the whole city and made the modern day New York city landscape (Caro 1975).
Another example of how to manage and win
exchanges is to control resources. Do you have
resources others want? Do others have the resources you want? If you depend on others, or
they depend on you, then there arises powerdependence relations. You may need to exchange
far more of your unvalued resources for those you
really need. Such leverage puts certain parties at a
disadvantage.
The third thing is control over preferences
and identities. Here, you want to transform persons wants so others demand the goods you provide. You want to create demand and define others preferences, or render their identities (in)compatible with the core interests of the coalition.
So to control exchange, follow the old saying
get rich, seize a hostage, and build a better
mousetrap! (March 1994: 150)
Exchange Process in Coalitions

Now that we have a better sense of the exchange process and how to manage and win it,
lets go back to coalitions and explain how they
work (keep in mind that exchange is the generative process of a coalition). Keep in mind that,
that exchange is still the generative process of coalition formation. That has not changed. All we
wanted to talk about now is the larger context of

multiple exchanges or a larger group. Here, coalitions are social systems wherein decisions are
made and reforms are pursued within a context of
potential conflict. That means, coalitions entail actors with mixed preferences and identities that do
not always align. They are often juxtaposed so
they require bargaining. Second, the objective of
members is to form a coalition capable of making
decisions favorable to them. This is obviously difficult because of all the internal inconsistencies.
Participants have parochial interest, and this is obviously difficult, because of all the internal consistencies within the group. The third is that people
therefore have to make exchanges, deals, agreements as to what decisions we made by the coalition. This follows the core process we described
above of exchange. And finally, resources are extracted through such coordinated action and distributed to competing coalition members. This is
what members get in return for joining a coalition.
The resources Hula will cite are strategic incentives, information, and symbolic benefits - Something we will discuss more in the next lecture.
These four characteristics help us understand
the nature of larger coalitions and how the process
of exchange sustains them. However, some important questions follow: Who will be in the coalition? And how are the spoils divided?
This can follow both the Logic of Consequence (which seems primary here) and the Logic
of Appropriateness. As an instrumental actor, you
join the minimal winning coalition so you can reap
the most rewards. As a rule-follower you seek coalitions that match your identity and the standards
you adhere to.
If we look back at the Bureaucratic Politics
model, we will see all the same features I am relating here were coarsely related there. It is just that
here I have tried to anchor the description more in
the process of bargaining or exchange. Above we
described some of the means of controlling exchange which can be extended to controlling a coalition. However, most coalitions will require negotiation and bargaining more than anything else.
Hence, within the context of a coalition, the manager or developer of a coalition is primarily con-

45

cerned with the interaction process by which exchanges are negotiated.

Interactions that Create and Destroy Coalitions

What kinds of interactions create coalitions?

There are a variety of interactions, and they range
from horse-trading, bribing, persuasion, and making threats, to management of information, logrolling, forming alliances, and joining associations. Managers of coalitions really need to focus
here! Lets take a look at log-rolling. What is
log-rolling? It is a coalition of individuals largely
indifferent to each others demands. For instance,
when there is a faculty meeting and we have a
vote, I agree to not care about a hire in policy research if they do not care about a hire in basic research. It is an exchange of acquiescence. Logrolls are particularly attractive for single-issue participants with weaker feelings on most issues.
(Later, we will see Hula call these hanger-ons).
But if you violate log-rolling, suddenly people will
make a stink over something they normally do not
care about and that can create issues of trust and
all kinds of problems. But the point here is that in
order to manage a coalition, you need to think
about a series of exchange logics that have different kinds of allocations and contingencies to them.
And above we list a variety of them. So, coalitions are a dynamic accomplishment through
forms of exchange. They entail all sorts of wrangling and bargaining.
The dynamic nature of coalitions means they
are often under threat. There are variety of things
that threaten them. For example, ambiguity is
good thing for coalitions (March 1994). When issues get cleared up or resolved, members tend to
leave a coalition. As such, clarity and resolution is
not always good for a coalitions survival. Second, outcome optimism is often needed (recall Allison suggesting one look up/down/sideways and
build support?). You have to over-estimate positive consequences of coordinated action when you
are trying to bargain for a coalition. This often
leads to post-decision disappointment and danger
of dissolution during implementation. And then
finally, members often exaggerate their support.

When the rubber hits the road, or when the coalition actually begins to adopt and implements
things, it begins to fall apart. All these weakly
aligned people find that once the initial formation
is hard, they no longer care to continue offering
support.
Coalitions, therefore, exhibit an odd dynamic. They start strong and end weak, or worse,
fall apart in implementation. Building them requires constant bargaining (e.g., log-rolling, wrangling, horse trading, etc). Maintaining them requires ambiguity and control over resources until
implementation is complete.
Case - Kevin Hula and Lobbying

In this part of the chapter we look a little

more closely at Kevin Hulas book on interest
group coalitions and their lobbying efforts in the
United States Congress (1999). Why lobbying? It
is not focused on a single organization or within
one, but it seemed highly salient to those of you
wanting to be leaders and social reformers. Most
social reforms in the United States or any democracy require legislative decisions. And much of
that starts with lobbying and interest group coalitions that succeed in influencing and establishing
laws.
Hula uses an exchange model, much like we
discuss in the prior lecture and is similar to Richard Emersons notion of social exchange theory
(1962). Participants engage in exchange for some
benefit. With lobbyists, free riding is less relevant
because they have already made the decision to be
involved with a cause in some form. Hence, the
issue is more about selecting a level and type of
involvement, not whether they get involved or not.
Coalition brokers then work incentives to get people to participate in different ways so as to effectively accomplish their interest.
So why do other people aside from the lobbyist - join a coalition? Lets look at this more
carefully: Hula gives multiple reasons why groups
would join an organization.
First, he argues that groups benefit from being able to reference an explicit policy or goal to

46

which they and others agree on and can say they

are for or against. For example, in the Hula text,
he discusses CEF (committee for education funding). CEF is a coalition with a broad goal, and
many of the member groups parochial interests
can be subsumed under it and listed (e.g., specific
programs). By subsuming a more particularistic
goal under that of an umbrella goal, they can strive
toward the larger one partially accomplishing their
narrower aims. Moreover, their narrow focus may
turn others off, and a broader coalitions goal may
be useful to their efforts.
Second, by joining a coalition early, you can
shape its agenda and platform. Most issues get
ironed out earlier in a coalitions formation than
later, and precedent exists for previously worked
out conflicts of interest (why open that can of
worms again-kind of thing). Information is a selective benefit of membership it fits parochial interests. Members want to know of any future
threats to their perceived interests. This is especially of value to smaller groups with small staffs.
Therefore, it helps to get on committees that make
decisions, that deal with new issues, and know the
latest about bills being proposed on Capital Hill
(in congress).
Finally, there are symbolic benefits. Showing something is an important issue is not the same
as making it one. Many organizations see something as important and join to show that, but they
do not have the resources to devote core membership activities to it. Plus it looks good to say you
were involved and busy. So reasons for existing
are reinforced. It serves higher ups in the organization who are more concerned with company affairs. And you can also claim credit when something goes well . For example, imagine coauthoring, but you are the 5th author who does not do as
much. You can claim responsibility and rewards
from that collaboration or coalition. Therefore,
joining symbolically can be seen as paying off a
debt (reciprocation), setting an example, etc.
Some kind of symbolic benefit of membership
comes from it.
Now that we have some idea why members
join a coalition, we can start to ask and explore
why members vary in their commitment. The in-

centives a particular group responds to in joining

the coalition will strongly influence the ultimate
role the group will play in the coalition structure.
Understanding whether a group joins a coalition
for strategic reasons or selective benefits helps determine whether it will become a 'core member,' a
'specialist/player,' or a 'peripheral, tag-along member' of that coalition. (Peripheral groups arent
free riding because all groups have entered into a
transaction and the other participants have agreed
to the legitimacy of the exchange.)
Lets look at the types of members and how
Hula describes their level of interest, goals, resources, and commitments to the coalition. First,
lets look at the core members in the first column
(or Founders). Core members view the issue as
very important and are interested in a broad range
of issues on it. Their goal is an overall strategic
victory on the issue. They bring to the coalition a
high level of time, money, reputation, expertise
and membership. And they commit to the coalition more than other members.
By contrast, the players (or specialists), care
about their specific goals and attempt issuehoning. The usually bring enough resources to get
a seat at the negotiating table, so they often bring
expertise on a specific issue as their political capital (piggy-back effort). In-so-long as their specific
issue is at stake, they stay on with the coalition.
Then there are the tag-alongs. They have the
least interest, and their goal is to acquire coalition
byproducts. They bring few resources, but they
are willing to let others use their name. Notably
this model of coalitions ends in an almost Hobbesian view only the most central, powerful actor
invests most into the coalition, while other less
powerful actors invest much less (king and citizens in Leviathan). The core players are interested
in getting the bill passed; the players wanted a
paragraph; and the tag-alongs wanted a picture for
their newsletter. Each lobbyist defined his or her
essential interest, and a symbiotic relationship was
formed.
I have been studying how research centers
and new academic departments get created here at
Stanford, and I see many of these distinctions.
Large interdisciplinary research centers seem to

47

have core members hell-bent on addressing an

broad goal and specific issues regarding it, and
they invest all they have into the effort. But in
building the center, they need to draw in other adherents, many of whom only have a specific interest, like performing a particular research project
that relies on expertise of some subset of faculty in
the center. Because of this, they join and lend
their name and reputation, and they even lend their
expertise on issues related to that, but they do not
attend all the events, nor do they work hard at forging the larger research community. And then there
are the tag-alongs or affiliates who are tangential
to the center and not dependent on it for much of
anything, but they have related projects. Here, the
center can invite them to be an affiliate, and use
their name. In some cases, this can result in some
minor research funds going their way, an school
newspaper article on their work, and recognition
by colleagues (all by-products). But these members seldom attend or do much of anything to promote the community. Nevertheless, they give the
impression of a larger, respectable, collective effort. Interestingly, a similar process arises when
forging a new department, but even more so as a
variety of goals and tasks must be accomplished
from establishing a program to establishing funding of faculty positions.
What do these commitments mean for the
maintenance of a coalition? Hula uses this nice
Rousseau quote to articulate this concern on page
43 (1999): when hunting, those pursuing a deer
will be willing to share; those pursuing a rabbit
will not share. In the coalition, core members
want nothing less than the stag and players will
jump for the rabbit if they can. Because of this, a
coalition manager needs to make sure the broader
goal is the route to an occasional rabbit.
For example, when I run a large research project focused on a larger issue, I try and encourage
the methodologists and computer scientists (or specialists) to send of papers to conference proceedings and methodology journals. Our larger research question is not methodological, but many
of these specialists hope the collaboration on a
new topic will help them innovate their methods
along the way. Hence, I have to point at rabbits

along the way. It is feasible one can view this

variation in commitment in another, more Machiavellian way, however. Opponents can target less
committed individuals and pick off members of a
coalition you can show them alternative legislation where their issue is subsumed (amendments!),
etc. Tag-alongs are of course the least committed
to a coalition. They wont commit much energy to
it, but they join to get selective benefits of information and symbolic clout. Tag-alongs are the third
group in the woods discount hunters along for
the beer and company.
So members can have different goals, levels
of interest and commitment to the coalition. Asymmetries are allowed at different levels because different exchanges are had. Leaders must be able to
welcome tag-alongs and differentiate real players
willing to go the distance. The danger for tagalongs is that they might feel betrayed in the end,
or used, thereby enacting a revolt. This occurs in
Hulas case of proposition 187 (pp. 49) the 1994
passage of the California bill aimed at illegal aliens. Trust was hard to establish when there was
little commitment to the coalition.
So given all this how do you develop and
manage a coalition? Earlier, I talked about managing exchange. Now the concern is going outward
and managing this larger, fleeting group of exchanges and their alignments in some kind of consistent way that meets your (cores) interest. First,
as the manager of this loose coalition, you want to
think about and identify all the interested actors /
organizations in the environment. Consider related issues, etc. Who would be interested? Second, ask yourself why they would be interested and
whose side they would be on. - Friend or foe kind
of stuff - and keep in mind a friend of a friend is a
friend, and the foe of a foe may also be a friend.
You do not want to mobilize any opposition, and
only garner support. However, you may want to
consider possible responses to oppositions (e.g.,
like targeting their specialists and tag-alongs).
Keep in mind that staff members have histories and inter-group linkages you can draw upon.
These can be effective conduits for coordination.
For example, a former employer might be a better
connection than a former employee up-chain con-

48

nections are likely better than down-chain ones.

Some people even belong to multiple coalitions.
Use them as well for information. From this a history and wealth of contacts develops -- they can
efficiently identify potential partners and adversaries, and their relationships can serve as points of
action and information collection (or receptacles). With more linkages, you need not develop
lasting coalitions because you always have access
to new members and their resources. In the field
of education, there are fewer links and developed
networks, so long-term coalitions are relied upon
more heavily. Nonetheless, the basic rule in coalitions is one of immediate usage of ties as their
cache is now. Commitment is fleeting, exaggerated, and ambiguous. Last, as our earlier discussion made clear work exchange, by bargaining
and negotiating. Now that you know interests, alliances and options, you can begin to horse-trade,
log-roll and so on. You can negotiate and work
the coalition into the shape you need.

49

50

Exists when there is a unified actor with

consistent preferences, lots of
information, and clear goals (and time
calculate).
Unitary actor or team that confronts a
problem, assesses objectives (goals) with
regard to it, identifies options, the
consequences of said options, and then
chooses option that minimizes costs.
Variant: Bounded rationality and
satisficing. Recognize imperfect info,
ambiguity, and select first satisfactory
option (good enough).

Action = Maximization of means to ends.

Know alternatives and their consequences

for the shared goal, and select wisely.
Improve information and analysis.
Management by consequences.

Management
Strategies

Know SOPs, what problems they go with (matching),

and who cues them. Improve rules and matching with
problems. Management by rules.

Action = output close to prior output (path

dependence), cueing of SOPs appropriate to problem.

NA

Actors in hierarchical organizational positions. Cue

sequential routines that accomplish task or solve
problem by routines available (supply issue).

Formal roles, hierarchical.

Not salient except as influencing

consequences of options.

Objectives compliance to SOPs, match with

problem parts.

Organizational positions

Unified team or actor

Goals are defined in regard to problem.

Matching identity and SOPs (solutions) / programs /

repertoires to problem.

Dividing up problem, coordinating / activating

organizational actors who have special capacities /
SOPs for parts of problem, conducting sequential
attention to objectives (localized searches until
problems resolved). Action guided by processes /
available routines.

Exists when the decision is guided by a logic of

appropriateness matching problem to actors with
procedures for handling it (routine-process focus).

Organizational Process (OP) / Limited Problem

Solver (LPS)

Maximization of options (solutions).

Dominant Pattern
of Inference

Environment

Goals
(what probs to
resolve)
Social Structure

Technology
(how solutions
get decided)
Participants

Key Organizational Elements

Summary or Basic
Argument

When does it
apply?

Rational Actor (RA)

Summary Table of Three Theories to Date:

Bargain with players (log-roll, horse-trade, hinder

oppositions coalition formation, etc). Learn others
interests / weaknesses so you know how to
manipulate and win. Direct management of relations
via bargaining.

Action = result of political bargaining.

Deadlines and wider array of stakeholders.

Coalitions enemy/friend

Parochial priorities, goals/interests, stakes / stands.

Players in positions

Bargaining, or playing the game (within its rules), or

political maneuvering.

Focus on the players occupying various positions;

their parochial interests (their conceptions of
problems and solutions); their resources (expertise,
money, people) and stakes in game; and bargaining
processes between them that establish agreements /
coalitions.

Exists when there are multiple actors with

inconsistent preferences and identities, and none of
whom can go it alone without assistance of others.

Coalitions /
Bureaucratic Politics (BP)

References
Allison, Graham T. 1969. Conceptual Models
and the Cuban Missile Crisis. The American Political Science Review 63, 3:689-718 review 3rd
model from last time.
Caro, Robert. 1975. The Power Broker (especially ch. 33, pp. 703-754). Vintage Press.
Cyert, Richard and James G. March. 1963
[1992]. A Behavioral Theory of the Firm.
Prentice-Hall, Ch. 3-7.
Emerson, Richard. 1962. "Power-Dependence Relations." American Sociological Review 27:31-40.
Hula, Kevin W. 1999. Lobbying Together: Interest Group Coalitions in Legislative Politics. Washington D.C.: Georgetown University Press (chapters 1-5, 7, and 9 [pp.1-77, 93-107, 122-135]).
March, James G. 1962. "The Business Firm as a
Political Coalition," Journal of Politics 24: 662678.
March, James G. 1994. A Primer on Decision
Making: How Decisions Happen. NY: The Free
Press. Chapter 4, Pp. 139-174.
Scott, Richard. 2003 (5th ed). Goals, Power, and
Control, Chapter 11 (pp. 291-324) of Organizations: Rational, Natural and Open Systems, 5th
Edition, Englewood Cliffs, NJ: Prentice-Hall.

51

4
Organized Anarchy

(Source: http://nopsa.hiit.fi/pmg/viewer/images/photo_2801526484_6f0480b4a2_t.jpg)

Organized Anarchy Model

This chapter introduces you to the basic features of decision making in organized anarchies, or
what some call a garbage can theory of organizations. During the week on coalition theory, I ask
students in my classroom to assume the role of different organizations that have a contradictory stake
in an issue like that of the Milwaukee Voucher Program (Quinn and McFarland 2006). Then every
group has a pair-wise encounter with each other
where they can apply a variety of exchange techniques so as to try and forge a dominant coalition
behind certain solutions like universal vouchers,
targeted vouchers, magnet schools, more funding,
class size reduction or do nothing. Every year, the
student groups do a great job of playing to their organizations parochial interests and manipulating
other organizations into joining some sort of collective resolution. But a lot of what they experience
goes beyond what coalition and exchange theories
of organizing capture. There is a far more chaotic
and dynamic quality to their discussions and decisions that seemed more consistent with an organized anarchy model.
Organized Anarchy - Introduction
What do I mean that the decision process resembled an organized anarchy? Well, for example, some of them have a hard time coming up
with their groups platform and identity (whats the
platform of lower income parents in Milwaukee?).
Also, some of the groups proposed solutions
changed over the course of bargaining some initially proposed universal vouchers only to promote
targeted vouchers in the end. Almost all of the
groups thought in terms of an identity and what
that entailed. And they also thought about others
identities and interests when trying to manipulate
the situation in their favor.
Problems seemed to be brought up in a much
more dynamic and contingent manner. Some
groups brought up problems that fit their interests
(e.g., problem of choice for Republicans; problem
of equity for African Americans; problem of

achievement for businesses), while others mentioned several problems (educators). And then
they presented the problems in different orders.
The same occurred for solutions. Groups created additional solutions to those arising in the Milwaukee case (e.g., sliding scale Vouchers). Some
solutions they never took up (do nothing). None
of the solutions and problems seemed to arrive as
set pairs. Instead, the solutions were matched with
multiple problems and those connections were negotiated. Each group tried to make a case for why
another groups problems could be addressed by
their solution. As such, the bargaining was in connecting solutions and problems in a way that convinced other groups.
The debates and decisions also followed a
temporal dynamic. Some of the students got up
and went to the restroom and their voice was lost
in pushing for certain problems and solutions.
Some pairs of groups took longer to finish their exchange and were rushed to make a deal before
their time was up and that seemed to affect decision outcomes. Some groups even back-tracked
on prior deals when they saw a better solution and
coalition emerge. Many students felt the ordering
of pair-wise meetings greatly affected which bargains arose and which were dropped.
Garbage Can Model
A lot of what I have described pertains to an
organized anarchy view of organizational
decision-making, or what some organizational
theorists call the Garbage Can Model. This theory was proposed by Cohen, March and Olsen
(1972), and throughout this chapter we will draw
heavily on their conceptualization (March 1994:
Chapter 5).
Most organizational theories underestimate
the confusion and complexity surrounding actual
decision-making. Many things happen at once;
technologies (or tasks) are uncertain and poorly understood; preferences and identities change and are
indeterminate; problems, solutions, opportunities,
ideas, situations, people, and outcomes are mixed
together in ways that make their interpretation un-

53

certain and connections unclear; decisions at one

time and place have loose relevance to others; solutions have only modest connection to problems;
policies often go unimplemented; and decision
makers wander in and out of decision arenas saying one thing and doing another. Organizational
decision making often looks like a mess!
With ambiguity, the story of decision-making
moves away from conceptions of order concerning
reality, causality, and intentionality to conceptions
of meaning. Here, decisions are seen as vehicles
for constructing meaningful interpretations of fundamentally confusing worlds (logic of appropriateness!), not as outcomes produced by a comprehensible environment. Hence, as we increase the complexity of decision situations so they more closely
reflect reality, they become meaning generators
instead of consequence generators.
Given this chaos, is there any way to theorize so as to get beyond interpretive, detailed, contextualized accounts? Cohen, March, and Olsen
(1972) describe organized anarchy in a relatively
simple model that describes the more chaotic reality of organizational decision-making. In Garbage
Can Theory, an organization is a collection of
choices looking for problems, issues and feelings
(problems) looking for decision situations (choice
arenas) in which they might be aired, solutions
looking for issues (problems) to which they might
be the answer, and decision makers looking for
work (1972:2; Italicized text added). One can
view a choice opportunity (or meeting with decisions) as a garbage can into which various kinds
of problems and solutions are dumped by participants as they are generated.
Taken in broad perspective, Garbage Can
Theory (or GCT, as I will be referring to it) suggests the following possible metaphor for decision
making within an organization:
Consider a round, sloped, multi-goal
soccer field on which individuals play
soccer. Many different people (but not
everyone) can join the game (or leave
it) at different times. Some people can
throw balls into the game or remove
them. Individuals, while they are in the

game, try to kick whatever ball comes

near them in the direction of goals they
like and away from goals that they
wish to avoid. The slope of the field
produces a bias in how the balls fall
and what goals are reached, but the
course of a specific decision and the actual outcomes are not easily anticipated. After the fact, they may look
rather obvious; and usually normatively reassuring (March and Olsen
1976. Ambiguity and Choice in Organizations. p. 276).
What is an example of an organized anarchy? Robert Birnbaum uses GCT to describe the
American college or university. He describes the
university as a prototypical organized anarchy
and especially the faculty groups like departments
and the academic senate. He views them as not
decision-making organizations, but meaningmaking ones (439):
Organized anarchies need structures and processes that symbolically reinforce their espoused values, that provide opportunities for
individuals to assert and confirm their status,
and that allow people to understand to which
of many competing claims on their attention
they should respond. They require a means
through which irrelevant problems and participants can be encouraged to seek alternative ways of expressing themselves so that
decision makers can do their jobs. They
should also be able to keep people busy, occasionally entertain them, give them a variety of experiences, keep them off the streets,
provide pretexts of storytelling, and allow socializing (Weicks The Social Pyschology of
Organizing, p. 264).
So here we have this understanding that organized
anarchies are a context for meaning making not
consequence generators. That is kind of an interesting, aspect of organized anarchies - that we
need these contexts within organizations so that
we feel like we have reasons and identities for be-

54

ing there and for addressing all kinds of concerns,

many of which may not be consequential. The
places we see organized anarchy are meetings (faculty meetings) and those kind of settings.
Now that we have some sense of where organizational anarchy resides, and and the kind of
general world it is, we can begin to identify their
characteristics. How do we know one when we
see one? What qualities do they have? The most
common things people reference are...
(1) Ill-defined goals, problematic preferences
and inconsistent identities. Within organized anarchies it is unclear which problems mater and which do not.
(2) Unclear technology. It is unclear what the
consequences are for each proposed solution or alternative; it is unclear how to
solve problems because the proposed solutions lack evidence.
(3) There is fluid participation. Within organized anarchies people come and go. There
is participant turnover.
(4) There are quasi-independent streams of
problems, solutions, participants and
choice opportunities. Meetings come
and go on their own schedule; and participants enter and exit depending on theirs;
problems seem to be noticed and related
in ways independent of the persons present or the possible solutions; and solutions seem to hang around, waiting for a
problem that suits it some day.
When these qualities arise in a choice arena, some
form of organized anarchy is likely occurring.
Many of these features also seem to be interrelated in the process of choice. That is, organizations make choices by attaching solutions to
problems, subject to chance, timing, and who
happens to be on the scene. Take for example,
faculty senates. A decision situation (or choice opportunity / arena) is like a garbage can into which
various kinds of problems and solutions are
dumped by participants who attend the meeting.
In such a meeting, decisions happen when problems, solutions, participants, and choices coincide.

The timing is right, and solutions are attached to

problems, and problems are attached to choices by
participants who happen to have the time and energy to see them through. In short, Garbage Can
Theory is about the social construction of meaning
attached to a choice.
Now that we have a general sense of organized
anarchies, lets look more carefully at their particular features. First, they entail (1) Choice opportunities (~what John Kingdon calls policy windows, see Kingdon 2003). These can be meetings, committees, and so on where the opportunity
and capacity to make a choice are possible. These
choice opportunities and policy windows can be
seen as garbage cans. The meaning of a choice
derives from how the trash is organized within a
can - or the mix of problems, solutions, and participants.
Second, organized anarchies entail (2) distinct
flows. Imagine three continual streams of trash
flowing through each can. It is all chaos in the
garbage can, but order is in the larger flows and
their confluences. Each stream flows relatively independent of the other. That is, problems get generated in public opinion (e.g., educational crises
like school shootings, national and international
exam reports, etc), solutions are constantly generated by academics and vetted even when their problem is not recognized yet (e.g., character education
and heterogeneous groupwork), and participants
come and go for other reasons (e.g., school boards
turn over, teachers come and go with tenure or
leave the profession altogether).
Lets look at each of these streams in turn:
The first stream is one of issues or problems (p1,
p2, p3 ~ Kingdons problems). These do not
need to be real problems or even the most important ones. They need to be perceived as such by
the participants in the choice arena. The second
stream is one of solutions (s1, s2, s3 ~Kingdons
policies). These pertain to ideas, bills, programs, all solutions [old and new], standard operating procedures that are revisited and even
changed. And they dont need to pertain to any existent problem. They can lead or lag problems.
The third stream is one of participants or actors
(a1, a2, a3 ~Kingdons participants and as

55

stream, politics). In the government arena, politics determines what actors show up, what interests are represented. Even if a decision is good for
a congresspersons constituents, they may pass up
on the meeting due to political concerns).
So there are these three streams, but they
mean little until a choice opportunity arises. All
too often, the opportunity just is not there. There
is no meeting, most people lack access to it, etc.
And even if there is a meeting, the right confluence of flows may not arise. The right problem
and solutions enter, but all the wrong participants
are there and the decision lacks energy and momentum. This is why timing and finding the right moment matters so much!
Now the outcome of choice arenas can vary.
In many cases, you can hold a meeting and no one
can agree on a problem or solution. One idea after
another is shot down and thrown away. On many
occasion, no decision gets made. In other instances, the solutions adopted do not address a
problem. This can arise in two ways. The first is
by Oversight: sometimes choice opportunities arrive and no problems are attached to them. Why
might this happen? It can happen if all problems
are attached to other choice arenas. In these instances, people make choices and select solutions
before problems reach the meeting. Later, we will
show you such a case where the school board and
the administrators of a district cannot attend meetings about a desegregation court order and its implementation because they must focus on other
concerns like a teacher strike.
The second means by which an adopted solution fails to affix to a problem is by Flight: Here
problems are affixed to choice opportunities for a
while and exceed the energy of the decision makers attuned to them. Hence, the original problem
may move to another choice arena (like another
meeting or department). In these instances, people
wait for the problem to go away in order to pick a
solution. So, in these cases you will see later, people table a decision or send it off into a subcommittee. In both of these instances, the problems do
not get attached to a solution.
Of course, the case we are most interested in
as managers of organized anarchies is when a

problem actually gets resolved: these are instances where problems are brought up in a choice
opportunity or meeting, and the decision makers
attending that meeting bring enough energy/ability
to meet the demands of the problems. Here a
choice is made and the problem is resolved.
Each garbage can, choice opportunity, or
meeting, has different access rules. In particular,
every choice arena has an access structure or social boundary of sorts that influences which persons, problems, and solution can enter or not. The
loosest structure allows for unrestricted access.
All the problems, solutions, and people are allowed to enter, and this creates more energy, but it
also allows problems, solutions and participants to
interfere with each other. This increases conflicts
and time devoted to problems you get greater anarchy! Another structure entails hierarchical access. Here, important actors, problems and solutions are given priority access. For example, big
decisions may occur in executive meetings, while
unimportant issues are addressed by the rank and
file employees.
Finally, there is specialized access. This occurs when special problems and solutions have access to certain meetings. For example, in my
school, the costs students incur when printing their
papers on school printers may be an issue that
goes to the schools technology committee, while
journal costs might be brought up in the library
committee. Therefore, certain specialists have access to certain choices that fit their expertise (e.g.,
engineers with technology concerns).
The diagram below shows the differing access structures for participants.

Figure: Access Structures

(Adapted from Cohen et al. 1972:5-7)
56

Another constraint influencing access to

choice arenas are deadlines. Deadlines characterize temporal boundaries and the timing of decision
arenas and what flows enter them. Here there can
be constraints on the arrival times of problems.
For example, there are seasonal problems like the
flu or cold weather. There can also be constraints
on the arrival of solution, such as when we propose and implement 1 or 5 year plans. And there
are constraints on the arrival of participants, such
as that defined by the timing of work days, school
years, tenure cycles, and so on. There are even
deadline constraints on choice opportunities or
meetings, such as the meetings dictated by yearly
budget cycles and student admissions.
All of this compounds and characterizes decisions in organized anarchies. Decisions arise from
the interaction of constraints (access structures
and deadlines) and the time-dependent flows of
problems (or issues), solutions, and participants
(decision-makers).
To this point, I have covered a lot of concepts in a short amount of time. Lets take the example of a faculty meeting again and work
through the features I have mentioned and see
what they look like. I think this will afford you a
more concrete sense of what the concepts mean
and how to see and apply them in various cases of
organizational decision making on your own (or
rather, meaning-making where a decision might
not get made!).
Lets begin with some of the problems that
might flow in an academic environment. One
problem might concern space usage we have
more people than we have space at Stanford, so it
might be relevant (p1). Another problem could be
the need for additional money or resources (p2)
and whether the school has enough grant money to
function well. Other problems might concern a student advising issue (p3), or even a research center
losing staff (p4), or concerns about the university
endowment and how it lost 1/3 its value in the recession (p5). So those are our potential problems
swirling in the environment. The figure on the
next page captures this space.
The blue circle is the choice arena or faculty
meeting. Which actors or participants attend?

Lets say it is an executive committee meeting

where access is hierarchical, and therefore only the
dean and associate dean can enter (a1-2). And finally we have various solutions: s1 could be a solution concerning minority recruitment; s2 could be
a plan to increase masters student enrollments; s3
might be a new tenure policy; and s4 might be an
idea to find new donors for the school.
Now all of them might not enter the choice
arena, and the meeting agenda might have a certain order and have a finite timeframe of 1 hour,
thereby imposing a deadline. So lets think about
this diagram and what we see:
(i) Lets look at p1. It does not really seem
to go anywhere and not decided on before
a solution enters (decision by flight!).
(ii) p2 on the other hand connects. Or
rather, it is linked to s1, a1, a2. They get
enough energy to be decided upon (i.e.,
decision by problem resolution).
(iii) p5 is also linked when they discuss p2,
but the actors never see the endowment
decline being solved by increasing enrollments. So the faculty who attend agree
that the problem of not enough resources
can be solved by increasing MA enrollments thereby increasing the funds gotten via tuition. So that is the choice decision that occurs. p5 is ultimately unconnected to a solution. So it is another decision by flight.
(iv) And then p3 and p4 is never even
brought up before the meeting ends. So
the deadline affected its discussion.
(v)p1 through p5 could have affixed to s1,
but no actors latched onto it. A plan for
minority recruitment could then be regarded as having underwhelming support.
If it had been picked without connection
to a problem, then we would say it was
decided on via oversight.
Hopefully you now see how these streams
collide in the garbage can, and how their ordering
and deadlines matter.

57

Problems

Participants
(who attends)

(space needs) p1
a1 (dean)
($ needed) p2

p1

(ctr decline) p4
(endowment!) p5

a2 (assoc dean)

a1

(std advising) p3

p2
a2
p5

s2

a3 (fac memb1)
a4 (fac memb2)

s1

(minority recruitment) s1
(increase MA enrollment) s2
(new tenure policy) s3

s1 s2 s3 s4
Solutions

(plan to find new donors) s4

Decision Situation
Managing Organized Anarchy
With all this in mind, we come to the question of how to manage organized anarchies. If we
see an organization that resembles a garbage can,
how do we approach it?
Several types of reactions can emerge.
First, you can try and be a Reformer: eliminate garbage can elements from decisions. Reformers create greater systematicity, order, and control. In a
way, this is what Daley and Vallas did in the Chicago public school case centralize, rationalize,
fix streams and access, etc.
Oppositely, you can be an Enthusiast: here
you try to discover a new vision of decision making within garbage can processes.. This is sort of
what March & Birnbaum argue people should do
in choice arenas like the faculty senate. Here, the

manager needs to realize the planning is largely

symbolic and an excuse for interaction, and sensemaking. It is a way to make people feel like they
belong and to learn about views and identities.
The arena is more for sense-making and getting
observations than making decisions. Also, the
manager can view temporal sorting as a way to organize attention. The order can indicate what is
more of a concern for collective discussion. An
enthusiast will focus on the flows of problems and
solutions and regard them as a matching market
where energies and connections are mobilized.
Recognizing who is present, where links / time
and energy are sufficient, and then pressing the
case is how youd approach it. Last the enthusiast
would see advantages in flexible implementation,
uncoordinated action, and confusion. Its ok not to

58

decide at times, and to make choice arenas into a

space of meaning-making.
Last, you can be a Pragmatist and try to use
garbage can processes to further your agenda (idea
being that organized anarchies are susceptible to
exploitation). Here you can time the arrival of solutions knowing attention is scarce. As such, you
can set the meeting agenda and work the order of
issues put ones you want discussed up front. Put
last the ones everyone knows need to be passed
but you do not want discussed so you can rush the
decision. Be sensitive to shifting interests and involvement of participants. You can be opportunistic and when certain people are not there, press on
issues and solutions you care about that they
would oppose if they were present. Or, you can
abandon initiatives that are entangled with others
if streams get tangled and the opposition is present, move on. If an agenda arises that does not
suit your interests, overload the system to protect
your interests: bring up other problems and solutions, slowing the process and making it more complex. Otherwise, you can provide other choice opportunities (other meetings) to attract decision
makers and problems away from choices that interest you. In this way you open up time for the issues you are concerned with.
In sum, you have options on how you want
to confront organized anarchy situations. Understanding how these arenas operate afford you different levers to try and hopefully the ones related
here give you a sense of how to start. I hope you
find the organized anarchy model useful. I find it
especially helpful because it renders pathologies of
choice theoretically consistent. All too often, real
choice arenas are messy and this theory embraces
that mess and affords us a framework for making
sense of it.
I find garbage can theory especially helpful
in explaining all sorts of meetings where there are
ecologies of choice and where problems and solutions are fluidly discussed. It fits the policygovernment world, research and development
groups, crisis management situations, and most
any distributed, decentralized social system trying
to deal with issues.

Examples of Organized Anarchy

We will now cover a series of examples and
applications of organized anarchy. Hopefully with
each example, you will see greater relevance and
form a more concrete understanding for how this
theory can be applied. I have three examples I
want to discuss. The first concerns the case of San
Francisco Unified School Districts effort to undergo desegregation in the 70s as told by Stephen
Weiner. I want to show you how that case can be
elaborated using the garbage can framework laid
out in the last lecture. Following that, I want to
discuss John Kingdons book Agendas, Alternatives and Public Policies. Kingdon writes a nice
summary of Garbage Can Theory and its application to the policy world and how legislative
agenda setting is performed. It is a great read that
I hope all of you will experience. Last, I will discuss the recent case of Title V in the No Child Left
Behind Act. This last case concerns an federal act
to reform the American primary education system.
I recount this briefly, using materials most people
can find online.
I understand many of you will not be familiar with some of the particular cases I am relating,
so I will try to afford a bit of overview and summary so you get their gist. The point of the examples is to get you thinking as an analyst and manager by applying theory to cases. It might be a
good exercise for many of you to try applying
these theories to cases of your own choosing. Just
view the ones I relate here as models and caricatures that you can apply, extend, and elaborate further.
The San Francisco Unified School District
The case I want to discuss first was written
up by Stephen Weiner, and it concerns San Francisco Unified School Districts desegregation plan
adopted in the 1970s. Here is the general story: In
the 1960s SFUSD experiences white flight, where
the white middle class families start leaving public
schools. At same time, desegregation court cases
emerge in the Southern United States and later

59

Participants (a1-6)

Problems (p1-10)
Problems that never
enter but draw SFUSD
to other arenas
p7 Tch-Std Boycott
p8 LatAmerOrg Sues
p9 Financial Probs
p10 Teacher Strike

Actors that never make it into the

choice arena:
a4 SFUSD consultants & admin
a5 working minorities
a6 working men
Problems that enter arena:
p1 Integrity of comm schl
p2 Bilingual ed
p3 Busing ! white flight
p4 SES integration
p5 Deseg 2ndary
p6 Deseg primary

s1 Tristar (3 zones
bussing / more deseg)

Actors that enter arena:

a1 community int grps
a2 fed consultants
a3 CAC (MC-WF)

s2 Horseshoe (7 zones,
respects comm / less deseg)

Solutions that enter (s1-24)

Solutions (s1-24)
Many solutions were proposed and
discussed, but few connected with
energy

Choice Arena for

Citizens Advisory
Council (CAC)

Deadline!

arise in more Northern and Western states. No action is taken by SFUSD

during this period and the NAACP (National Association for the Advancement of Colored People) warns the district it is too segregated. SFUSD
develops a desegregation plan that is immediately rejected in committee
due to cross-town busing fears. They fear such a plan would be hard to
manage and would be unwanted by the districts stakeholders. Instead, a
citizens committee forms and develops a desegregation plan for only 2
elementary schools.
In 1970 the NAACP files a suit demanding all 102 elementary
schools within SFUSD be included in the desegregation plan. The US district judge would not rule until the Supreme Court ruled (arguing SFUSD
made a small effort with 2 schools and therefore showed good faith). In
the meantime, the judge advises SFUSD to devise a desegregation plan.
SFUSD appoints one staff member and 3 committees: Staff Committee,
Certified Staff Committee, and Citizens Advisory Council (CAC). The
third committee has the most energy and committed members to this
cause.
In 1971 the US Supreme Court rules SFUSD must desegregate its
elementary schools and must devise a plan in 2 months. So it is a case of
partial decisions and little or nothing happening a pretty common occurrence when it comes to policy and school district reforms! Can GCT apply here and help us understand the process of relative indecision?
60

Figure. School Board Meeting (not SFUSD)

(source http://commons.wikimedia.org/wiki/File:Beverly_Hills_Board_of_Education.jpg)

In many regards this is an instance of organized anarchy. First, it is ambiguous as to what desegregation means. The problems and preferences
for desegregation are unclear and it is ambiguous
how to accomplish desegregation. How does one
know desegregation has been accomplished? In
effect, there is an unclear solution and an unclear
technology or means of bringing it about.. Moreover, there is a tight deadline and the participants in
this case keep changing judges turnover, different committees form and dissolve, etc. Only the
threat of a lawsuit creates a choice opportunity!
So the case of SFUSD has many qualities that suggest it is a case of organized anarchy.
Lets identify the problems mentioned in the
case as related by Steve Weiner. The figure on the
prior page identifies the problems, solutions, and
actors involved with SFUSD desegregation. The
focal arena is the Community Advisory Committee, since it is the arena in which a decision is ultimately made. The key problem for this arena is
that of desegregating the elementary schools p6.
At the outset, the participants were not sure what
integration should look like. They eventually
adopt a state standard that is very strong. All the
schools need to have a racial compositions within
15% of the district average.
A bunch of problems enter the CAC choice
arena and are interrelated by participants:
p1 - Keeping integrity of school complexes
p2 - Bilingual education needed
p3 - Bussing disliked by whites (white flight)
p4 - SES integration wanted

Figure. Teacher strike (not SFUSD)

(source - http://commons.wikimedia.org/wiki/File:London_Cuts-Demo_5704.JPG)

p5 - Desegregating secondary schools

p6 - Desegregate elementary schools
(the key problem!!)
Other problems arise but they are not taken up in
the CAC:
p7 - Teachers and students boycott schools in
disrepair (budget woes)
p8 - Lawsuit filed by Latin-American organization (demand for bilingual education)
p9 - Financial problems are apparent with
teacher contract disputes
p10 - Teachers go on strike
So while the courts demand SFUSD go through
desegregation, they are contending with a variety
of other issues and problems. Many are quite severe and draw necessary attention and resources.
A variety of participants are also involved,
but only some of them enter the choice arena that
takes up the problem of desegregation.
a1 Community interest groups
a2 Federal consultants
a3 CAC
a4 SFUSD consultants and administrators
a5 working minorities
a6 working men
Of these groups, the federal consultants are outsiders with little understanding of constituent concerns and who cannot always attend. Because
meeting times are scheduled during the day, the

61

most active CAC members tend to be white middle class women (stay at home moms), while working men and minorities are unable to attend due to
their day jobs (less energy to devote it). And finally, the SFUSD consultants and administrators
are drawn away by other problems that do not enter the choice arena for desegregation (a4 attend to
p7-p10). Only a1-3 attend the meetings.
At the actual meetings, these participants
raise and discuss a variety of solutions: twentyfour of them to be exact (too many to list). Here
are a few:
s1s24 Twenty-four solutions developed
and narrowed down to two.
s1 - Tristar (3-zone plan written by technocrats)
s2 - Horseshoe (7-zone plan less drastic)
What is not considered is the solution of simple
cross-town busing.

If we put it all together we begin to see what

happens in the CAC arena. Certain actors get
pulled away (a4) to other problems arising in other
choice arenas, while other actors just cannot make
the meeting times (a5 and a6). In the arena, the
CAC is composed of mostly white, middle class
females. Their attention and energy is on p1 sustaining the integrity of community schools and this
is related to p3 how busing might lead to white
flight. They see s2 the horseshoe plan as partially addressing the desegregation order (p6) as
well as the problem of sustain community schools
(p1). By contrast the federal consultants see s1,
the tristar plan as the best because it most fully addresses the desegregation order, but they do not
connect the solution to the problems other participants find salient in the choice arena.
In a way the diagram sums up the decisions that
arose and how the deadline affected the outcome.
The deadline of the court decision pushed prevented other problems and participants from fully

Kingdon and What Becomes Part of the Governments Agenda.

Figure. United States Capital Building

(Source - http://www.flickr.com/photos/brad_holt)

62

entering the discussion and decision. The case of

SFUDs segregation plan could have been different had there been a different deadline, different
meeting times, and different problems interfering.
Kingdon and Government Policy
Lets next turn to the Kingdon text (2003).
Kingdon does a nice job of summarizing some of
the major tenets of organized anarchy. He does
this in his focus on American health and transportation policies that arose during 1976-1980 presidency of Jimmy Carter. Kingdon asks: Why do
certain issues become part of the governments
agenda while other issues do not? Kingdons research finds that policy proposals are not necessarily written in response to a particular event.
Rather, at any given time, there exist a multitude
of proposals ready to go and waiting for the best
opportunity for their introduction. An ideas time
comes via a process of organized anarchy. Lets
look at how Kingdon regards federal agendasetting as such a process.
He does so by first asking who are the participants? -- Lets start by identifying the various
participants in Washington, D.C.:
Within government there is
Congress: Upper and lower house, plus
congressional staff they have scheduled
election cycles of 2 and 6 years so there is
some turnover.
The president, plus the cabinet, staff, and
his political appointees. The President has
a large say in agenda setting but less control on alternatives. His election cycle is
every 4 years, and turnover then is likely
even if he is re-elected.
Last there are civil servants: bureaucrats
who have longevity and expertise. They
turnover less frequently.
Outside the government:
Interest groups: lobbyists, labor, professional societies, public interest/advocacy
organizations, etc.

Academics and other researchers

Media
Voters
General public/constituents

So you have all kinds of other actors and participants that can affect the legislative process and
they turnover somewhat rather variably.
Next what is the process of policy formation? In what ways can we consider how a policy
originates and develops? Here, Kingdon considers
a few different models by which scholars have
characterized policy formation. The first concerns
origins. Where did the idea and policy come
from? How did the idea spread? The assumption
here is that it started somewhere and got taken up
more and more. We have an initial origin and if
we follow that origin, we will have some understanding for its development. A second view is
that of rational choice: We saw this earlier in the
course. Here, the view is that we define the goals,
identify alternatives, and choose the optimal alternative e.g., the policy in question. Therefore, its
adoption should be based on predictions of the policys consequences. A third view is that of incrementalism. Rather than starting from scratch, new
policies build on existing policies. Changes are
made at the margins and what we see today is an
adaptation of prior ones.
Kingdon argues that each of these descriptions has some value, but they do not describe the
process of policy formation as completely as Garbage Can Theory (GCT). Kingdon asks how does
agenda setting resemble an organized anarchy?
Lets take a step back like we did in the
SFUSD case and see if it fits the criteria. First, we
ask, is it a context of problematic preferences (inconsistent, ill-defined)? And here, the answer is
yes - Action is often taken before identifying preferences. Participants even disagree on their preferences and priorities. Second, we ask is there unclear technology? Kingdon says how the government attempts to solve problems is often unclear.
There is not a clearly defined way to desegregate
schools, eliminate the achievement gap, end child
poverty: its not like making widgets (2003:85).
Third, there is Fluid participation and there is a

63

good deal of turnover in personnel. Moreover, the

importance of participants does not match their job
description and the executive branch is often involved in legislative processes. Participants outside the government enter and exit the decision
making process all the time, and access varies. In
sum, the federal government would seem to be an
organized anarchy, as defined by Cohen, March
and Olsen.
Kingdons adaptation of GCT conceptualizes
three independent streams of problems, policies
(solutions), and politics (participants).
These
streams converge (couple) at critical points. It is
this process that sets the agenda. He sees the
streams as somewhat independent. For example,
problems flow in and out of focus in the news and
for legislative actors. Policies are generated and
sit around for years, circulating without a home.

No Child Left Behind

Participants come and go. And the opportunities

for decisions (i.e., choice arenas or garbage cans),
arises at different times.
The independence of these streams is a key
point I want to reiterate: policy solutions can be
developed whether or not they respond to an actual
problem. The political stream is not necessarily
dependent on identified problems. And as Kingdon says on page 88:
Advocates develop their proposals
and then wait for problems to
come along to which they can attach their solutions, or for a development in the political
stream...that makes their proposals
more likely to be adopted (Kingdon, p. 88).
These three streams must converge when a policy
window is open. That is, only when the conditions
are right will an issue find itself on a policy
agenda. If you have the chance, read Kingdon as

Figure. Department of Education Building at Launch of NCLB

(Source - http://www.flickr.com/photos/dchousegrooves/)

64

he does a wonderful job applying theory to this particular instance of agenda setting. Rather than rehash his application of GCT to particular instances
of agenda setting, I want to apply garbage can theory to a new case many of you might not be familiar with in this way I can afford you numerous
examples so you see how the theory can be applied in many instances, not just one.
No Child Left Behind
My last example will concern a recent policy
decision: Title V of the No Child Left Behind Act
the Promotion of Informed Parental Choice And
Innovative Programs (or NCLB). Briefly, NCLB
is the name of the 2001 reauthorization of the
1965 Elementary and Secondary Education Act
(which was part of President Lyndon Johnsons
War on Poverty). When originally passed, the
primary focus of the Elementary and Secondary
Education Act was on improving the education for
economically disadvantaged students who met federal definitions of poverty. Over time, the Elementary and Secondary Education Act was expanded
to include bilingual education, education to indigenous communities, education in correction facilities, magnet schools, foreign language programs,
midnight basketball, and migrant education.
The Elementary and Secondary Education
Act has been reauthorized several times since its
original passage in 1964, usually for approximately four- to six-year periods. President George
Bush signed the No Child Left Behind Act of
20011 into law on January 8, 2002. Title V provides federal grant support for Innovative Programs and Public Charter Schools. It also adds a
new incentive program to help charter schools
meet their facility needs. Included in this section
is a provision that provides transportation and
other support that allows students attending
schools that do not meet adequate yearly progress for two years to transfer to a charter school
or other public school.
So, how would we use Kingdons model to
describe how Title V entered the agenda and ultimately became law? First we would look at the

problem stream. At any given time, a set of problems may rise in prominence and capture the attention of governments, often not because of political
pressure but because of systematic indicators that
purport to prove the existence of a problem. That
is, problems may not necessarily be true problems. They merely have to be problems in the
minds of some subsection of the public in order to
be considered.
What problems could Title V purport to solve?
! Failing schools with no sign of improvement.
! Lack of innovation in public schools (charter
schools may be an incubator of innovation)
Public
! There is a lack of competition.
schools are not pressured to improve.
! Unequal opportunity for lower income children (these families have fewer options because they cant afford private schools. Charters are free public schools of choice.)
! Charter school funding (Claim by charter
school proponents that they receive a disproportionate amount of per pupil funding from
the state).
In most cases we would agree that these problems
are probably true. However, I want you to understand that it does not necessarily matter if you
think it is true or not. What matters is that a subsection of a population does - that there is energy
behind it, and actors are affixed to these kinds of
problems.
What are some of the indicators to this problem?
! International comparisons (USA behind)
! Achievement gap literature (by race, income,
urbanicity - disparities exist)
! Government evaluations and other studies
show many problems in schooling
All these indicators suggest the problems of our
education system are more than our biased view,
and exist beyond our own opinion.
What is the publics perception of this problem?

65

Summary of the Problems, Alternatives,

Politics, and Open Policy Windows.

Figure. Bush Signing NCLB

Figure. NCLB Symbol

(Source - http://commons.wikimedia.org/wiki/File:No_Child_Left_Behind_Act.jpg)

(Source - http://commons.wikimedia.org/wiki/File:Nochild.jpg)

66

! Public opinion is that schools in general are

failing (constant media bombardment of this)
! They see their own schools as a little better
than most (bias)
! Market forces make sense to people. They
like the idea of choice and think it could lead
to improvement.
! Bush presents rhetoric picked up in media:
soft bigotry of low expectations
In sum, there are streams of problems in the environment that relate to the No Child Left Behind
legislation.
There is also a stream of solutions (policies)
that is occurring. To see these, we next look at the
competing policy alternatives being proposed to
address the problems above. Within governments,
specialists including lawmakers, staffers, advocacy
groups, researchers, and academics concentrate on
developing policy proposals: Ideas are floated,
bills introduced, speeches made; proposals are
drafted, then amended in response to reaction and
floated again (Kingdon, p. 117).
So lets look at this more closely. What are
the policy alternatives that speak to the problems
identified above? First there are school vouchers.
Here a student might get so much money from the
state, they could use that money to apply to and attend another school of their choice (public or private). Another potential solution is to promote
charter schools, which is somewhat like promoting
vouchers, but here students are limited to public
schools. One could view public school improvement as a policy and solution. But how? Here the
issue is unclear technology. One could focus on
improving instruction (e.g., teacher preparation
programs, professional development training, or
new curriculum). Another way would be to structure the schools better like seen with some forms
of ability grouping, class size reduction, extended
school days, etc. Another would be accountability: where one assesses adequate yearly progress
or conducts annual testing with rewards and punishments, much like NCLB adopts. There are
other less ambitious solutions too - like simply
throwing money at the problem and existing programs. Or you can ignore the problem and play

the blame game. One could argue it is not the role

of federal government to mess with schools and it
is the responsibility of the states, cities, districts,
schools and school teachers.
All these are viable alternatives, and you just
need to remember in Kingdons model, policy
does not necessarily follow problems. These policy alternatives in many cases were developed independent of the problems we have identified. In
fact, much of NCLB, including accountability provisions, was developed under Clinton.
The third feature of NCLB we would look at
are the participants (politics) involved. The political stream described corresponds to Cohen, March
and Olsens participants/decision-makers stream.
Even when a policy solution attaches to a problem,
passage is not guaranteed. Political factors such as
partisan concerns, ideological distribution of policymakers and interest group lobbying can work
against any proposal, no matter how complementary it may be to a policy problem.
In the case of Title V, the reauthorization of
Elementary and Secondary Education Act was
signed in 1994 and was scheduled to expire in
1999. Congress and the Clinton Administration
began work on the reauthorization process in 1999
and in 2000 but failed both attempts to finish the
work. Education was a central component to candidate G.W. Bushs platform. And when Bush entered office, one of his first actions was to send to
Congress a broad outline of his education proposal. He vowed to Leave no child behind
which was hard to argue against on rhetorical
grounds. There was little Congressional criticism
of the final version of the bill (it passed 87-10 in
the Senate and 381-41 in the House) and received
support from even some of the most liberal members, including Representatives George Miller and
Barbara Lee and Senator Ted Kennedy.
Recall from our discussion of solutions or
policy alternatives, above, that school vouchers are
an alternative.
Although original versions of
NCLB contained voucher proposals for private
schools, this was given up in order to make the necessary concessions for the Democratic support required for passage. In other words, the political
environment was accepting of the provisions of

67

NCLB as it was passed. Since that time, there has

been some criticism (mainly around funding issues), but the public is still supportive of the general measures of the law.
The final feature of NCLB we would look at
is the Policy Window, which concerns deadlines
and the convergence of streams. Weve discussed
the three streams of problems, policy alternatives,
and politics. But these streams must converge
while a policy window is open in order for legislation to move. NASA has a launch windowa
time period in which a particular rocket must be
launched. If they miss the launch window, NASA
has to wait for the next one before it can go. The
same is true under Kingdons model. There are
particular times in which a policy window is open.
The policy window is not indefinitely open.
There are deadlines which constrain the amount of
time problem-alternatives have in order to be implemented. Decisions typically must be made by
the end of the legislative session. Failure to do so
means that the process would have to begin from
scratch at the start of the next session. In addition,
legislatures are systems composed of decision makers that can change from one election to the next.
A favorable set of decision makers may disappear,
to be replaced with a new set of decision makers at
the start of the next term who may be less willing
to support the provisions of Title V.

dency in its particular form and not well before under a different guise and during Clintons era.

In the case of Title V, The Policy Window was

open when there was a
Republican majority in Congress
Republican president
Frustration with public education
Promising start of the charter school movement
Strategic use of language by proponents of
NCLB
Success of state accountability laws (CA,
TX, others)
But most of the time, the policy window is closed.
So if we put all 4 features of Kingdon together, we
see the following table and understand better how
that legislations time occurred under Bushs presi-

68

69

Unitary actor or team that

confronts a problem, assesses
objectives (goals) with regard to
it, identifies options, the
consequences of said options,
and then chooses option that
minimizes costs.
Variant: Bounded rationality
and satisficing. Recognize
imperfect info, ambiguity, and
select first satisfactory option
(good enough).
Maximization of options
(solutions).

Summary or Basic Argument

Know alternatives and their

consequences for the shared
goal, and select wisely. Improve
information and analysis.
Management by consequences.

Not salient except as

influencing consequences of
options.

Environment

Management Strategies

Formal roles, hierarchical.

Social Structure

Action = Maximization of
means to ends.

Goals are defined in regard to

problem.

Goals
(what probs to resolve)

Dominant Pattern of Inference

Unified team or actor

Participants

Technology (how solutions get decided)

Exists when there is a unified

actor with consistent
preferences, lots of information,
and clear goals (and time
calculate).

When does it apply?

Rational Actor (RA)

Summary Table of Five Theories to Date:

Know SOPs, what problems they

go with (matching), and who cues
them. Improve rules and matching
with problems. Management by
rules.

Action = output close to prior

output (path dependence), cueing
of SOPs appropriate to problem.

NA

Actors in hierarchical
organizational positions. Cue
sequential routines that accomplish
task or solve problem by routines
available (supply issue).

Objectives compliance to SOPs,

match with problem parts.

Organizational positions

Matching identity and SOPs

(solutions) / programs / repertoires
to problem.

Dividing up problem, coordinating

/ activating organizational actors
who have special capacities /
SOPs for parts of problem,
conducting sequential attention to
objectives (localized searches until
problems resolved). Action guided
by processes / available routines.

Exists when the decision is guided

by a logic of appropriateness
matching problem to actors with
procedures for handling it (routineprocess focus).

Organizational Process (OP) /

Limited Problem Solver (LPS)

Bargain with players (log-roll,

horse-trade, hinder oppositions
coalition formation, etc). Learn
others interests / weaknesses so
you know how to manipulate and
win. Direct management of
relations via bargaining.

Action = result of political

bargaining.

Deadlines and wider array of

stakeholders.

Coalitions enemy/friend

Parochial priorities, goals/interests,

stakes / stands.

Players in positions

Bargaining, or playing the game

(within its rules), or political
maneuvering.

Focus on the players occupying

various positions; their parochial
interests (their conceptions of
problems and solutions); their
resources (expertise, money,
people) and stakes in game; and
bargaining processes between them
that establish agreements /
coalitions.

Exists when there are multiple

actors with inconsistent preferences
and identities, and none of whom
can go it alone without assistance
of others.

Coalitions /
Bureaucratic Politics (BP)

Time when your solution is raised (to

coincide with right participants and
cycle of problems) to maximize energy;
abandon entangled initiatives; know
how to overload system for policies you
detest; and generate choice opportunities
that work to your interests
(access/timing). Indirect managing of
situations.

Action / decision = result of streams

collision in choice arena.

Deadlines and other choice arenas (e.g.,

decision in current arena may be means
of access to another choice arena)

Access rules segmented, hierarchical,

or democratic.

Problems stream determined by public

opinion, prominence / vocalness of
problems in firm, etc.

Confluence of multiple streams, such

that solution is connected to problems
and enough actor-energy to see it
through.
Participant stream shaped by political /
career cycles & unplanned departures.

Focus on choice arenas (when choice

opportunities / windows arise); the
distinct and decoupled streams of
problems, solutions, and participants;
and their access rules to the arena
(whether structural or timed).

Exists when solutions are unclear,

participants turn over, and
preferences/identities are inconsistent.

Organized Anarchies /
Garbage Can (GC)

References
Birnbaum, Robert. 1989. The Latent Organizational Functions of the Academic Senate: Why Senates Do Not Work But Will Not Go Away? Journal of Higher Education 60 (July/August) 4: 423443.
Cohen, Michael D, March, James G. and Olsen,
Johan P. 1972. A Garbage Can Model of Organizational Choice. Administrative Science Quarterly
17(1): 1-25.
Kingdon, J. W. 2003 (1995). Agendas, alternatives, and public policies, second edition. Longman.
March, James G. 1994. A Primer on Decision
Making: How Decisions Happen. NY: The Free
Press. Chapter 5, pp. 175-218.
Weiner, Stephen S. 1976. Participation, Deadlines, and Choice Chapter 11 (pp. 225-250) in
Ambiguity and Choice in Organizations. (eds)
March, James and Johan Olsen. Bergen: Universitetsforlaget

70

5
Organizational
Learning

Source: http://commons.wikimedia.org/wiki/File:Organizational_Learning_and_KM.jpg

Organizational Learning
In this chapter I will describe the theory of
organizational learning and what it entails. Before
I begin, I want to briefly revisit our theory of organized anarchy by retelling how one of our inclass exercises went. As a capstone experience of
organized anarchy, I try to create a garbage can
situation for students to experience in class. Every
year, I call a meeting with the students and ask
them to discuss the course and its grading policy.
I even tell them I will adopt a new class format
and grading procedure if they can all agree on one
and convince me it will improve the learning experience. To help with the process, I ask them to
identify various problems (i.e., a problem stream
is created). The problems they identify are as follows: there is too much reading; lecture materials
go by too quickly; not enough time for individualized projects; not enough time for group projects
and discussion; etc. I then ask students to create a
list of policy changes they would like in the course
(i.e., a solution stream is created): often they ask I
allow them to rewrite papers; sometimes they ask
that everyone gets 10 points added to their grade;
or that my lectures be posted online; or exemplary
papers be shared.
All too often, there is little connection between the problems and solutions they select. For
example, the first two solutions of rewriting papers for a better grade and giving everyone 10
more points do not address any of the problems
they listed earlier. And the last one: what does the
posting of exemplary papers solve? Only the solution of posting lectures actually addresses a problem they list that my lectures go by too quick.
Anyhow, the next thing we do is discuss each solution. Very quickly we see the energy affixed to certain solutions, but then in discussion, it dissipates
as people identify additional problems the solutions may incur. For example, what does the policy of giving everyone 10 points do if they are
graded on a curve? Someone may even notice that
if everyone gets an A that it creates another problem: how will I write recommendations for students hoping to get into doctoral programs, or jobs
if everyone got an A? What distinguishes them?

Same for the other solutions. If I give students my

lecture notes, does it mean they will stop doing the
readings? If I give them exemplary papers, will
they merely follow that format and not be creative? If I allow for group projects and group
grades, it may not be fair because some people do
more of the work than others. And again, how can
I write individual recommendations for graduate
schools and jobs if students have group grades?
With each solution, new problems are affixed
that render them less feasible. In fact, that is often
the kind of discussion that arises in organized anarchy. It is also a tactic used by people trying to prevent the group from taking up a particular solution. In many ways, this is why there has never
been a drastic change in my classs grading policies. Students not just me! raise new problems
with every proposed solution. Moreover, we have
a deadline pressing on us. We only have 20 minutes of class time to decide. As a result, we never
discuss all the solutions and only the most outspoken students concerns get voiced.
The ambiguous nature of solutions, their connection to new problems, and the lack of time all
compounded to render ambitious reforms minor
(amended legislation, if you will!). And simple solutions can also quickly seem complex. In the
end, the class tends to agree on minor changes: students can revise their papers once, they can do individualized projects, and lecture notes are posted
after we meet as a class.
Most of you have experienced organized anarchy like this firsthand - you just never realized it
until now. Go to your next meeting where a bunch
of equals with different opinions try to make a decision. Watch the process unfold, and try and remember your lesson from organizational analysis
many of you now have the capacity to enjoy and
understand the process anew, and possibly even
redirect the discussion in ways that meet your interests!
An Introduction to Organizational Learning

Of course, this chapter is not about organized

anarchy, but rather about organizational learning.
In this chapter, we ask - What is the organizational
72

Source - http://www.flickr.com/photos/usfwssoutheast/8392695372

Organizational
Learning - Practice,
Understanding, and
Organizational
Memory

learning perspective? In the most general terms, the organizational learning perspective concerns adaptation and learning from experience. But
how does an organization learn? Organizations learn by encoding inferences from history into organizational structures (so best practices into
rules, routines, and roles), people, technologies (curricula), and culture
(norms, beliefs) that guide behavior. That is, organizations reflect on
what works well or not, and then encode that knowledge into its organizational elements (participants, technology/tasks, social structure) so it can
remember.
It is important to emphasize that organizational learning occurs at
the organizational level. There is no doubt that individual and team learning are related, but we need to keep in mind that it is the formal organization and firm that is making efforts to learn from experience and pass on
that knowledge to its employees in the hopes of constantly improving performance. In my discussion of the organizational learning perspective, I
draw on the writings of many writers from John Seely Brown and Paul
Duguid (1991; 2000), to James March (Levitt and March 1988; March
1991; 1994; March et al 1991), Linda Argote (1999), Lucy Suchman
(2007), Julian Orr (1996), and others. I merely want to afford you a general framework you can get your mind around and apply in the organizational settings you participate.
One text in particular, Brown and Duguid (2000), contrast organizational learning with an organizational process model (if you recall, this
was Allisons Organizational Process Model where organizations are
73

viewed as following routines and standard operating procedures). Brown and Duguid describe two
characterizations of routines or SOPs On the
one hand they are ostensive rules applied as a
guide and computer program (SOP ~ organizational process model); on the other, they are enacted practices (the heart of understanding or
knowledge). According to Brown and Duguid, a
manager of organizational processes will get a
company to streamline their SOPs to those concerned with the core task and then spell them out
so they are clear. They remove SOPs that are redundant, those that are in conflict with each other,
and those that are pointless.
A good example of a pointless rule might
be what we term blue-laws in the United States.
These are laws created many years ago that are
still in the legal texts even though they no longer
apply nor are they enforced. For example: in Kansas there is a law saying you can not eat snakes on
Sunday; in Connecticut you can not eat pickles on
Sunday; and in Massachusetts, Cows can not graze
in the Boston Commons. The organizational learning perspective agrees that organizational processes and SOPs matter, but it focuses on the practice of these procedures, and argues it is through
their practice that they have meaning, relevance,
and effect (and conversely, it is their lack of practice that make routines irrelevant and forgotten).
In fact, many organizational procedures can
not be looked up in a book or manual. And even if
they can be found in a manual, merely reading
about them does not result in understanding and
knowledge for most persons. Finding the right routine is hard (it may not even exist), and enacting it
well is even harder as each new situation will differ from the one before. You constantly have to
adapt rules and procedures so as to fit changing
situations and actual work experiences. Without
practice and experience, you have no real knowledge about working.
Take the example of self-defense routines.
They are learned as a routine, but then they are
practiced in bouts and used in relation to other routines. That is, they are not just read in a book, but
are practiced and applied before the student becomes an expert fighter. As such, organizational

learning differs from the organizational process approach in that it regards experiential learning
learning by doing (not learning about) - as the central means to making complex organizations work.
Now arguably, learning of this sort may not matter
much for simple tasks like procurement, shipping,
receiving, warehousing and billing as this band
of operations have really well-defined processes
with measurable inputs and outputs. But experiential learning will matter dearly for management,
and research and development where life is less
sequential and linear, where inputs and outputs
are unclear. Here, making sense, interpreting, and
understanding are points of contention and highly
valued. To get at this, one needs to look at the actual activity and practice within routines and work
processes.

Example of Blue Laws

http://commons.wikimedia.org/wiki/File:Ontario_Sunday_Laws.jpg/

74

Source - http://www.flickr.com/photos/kretyen/2460134871

Xerox and
Organizational
Practices

So enacted practice is a route to understanding, shared knowledge,

and expertise. Brown & Duguid give the example of computer help-lines
and Xerox machine repair experts to illustrate this point (Much of it is
drawn from work by Julian Orr and Lucy Suchman). This line of research finds that the machine manuals often dont tell you what you need
know, no matter how much you codify things. You can write out procedures for every problem a Xerox machine can have, and it is still terribly
inefficient (and painful) to ask people to read those manuals as a means to
becoming experts on repairing or using Xerox machines. Instead, a great
deal of professional understanding comes from practice actually doing
the repairs and work yourself. This type of learning is central to many
professions: think about how doctors learn in through residency training,
lawyers by internships, teachers by student-teaching, emergency personnel by simulations, etc and hopefully many of you will understand these
theories via case-applications!
There are several characteristics about enacted practices that distinguish them from ostensive rules you read in a textbook: First, practices
are inherently collaborative and interactional. Practice entails collaboration that leads to an indivisible product. For example, in the case of
Xerox machines, it involves talking to clients, interacting with machines,
and fixing them so they afford the output desired.
Second, practices are shared and understood through stories. As persons perform an activity, they develop accounts or stories. And these are
understandings of what happened and why. In many cases, these are like
75

formalisms represented arguments (we do this

all the time in tables, figures, models, stories, etc).
They can be readily remembered, passed on, and
accessed by others. They not only tell of specific
information but principles of causation and process! As such, the type of knowledge and its representation has a special link to memory.
Third, practices entail improvisation and adaptation via use. A central aspect of organizational
learning is individual adaptation and learning to
apply a rule. We improvise rules and routines so
they can be applied to the world as encountered
(we relate particulars of the world to general schemas of the organization). Even if organizations do
not recognize this process of adaptation and improvisation, it continues to happen. The Xerox representatives learned tricks to get by and understand a problem. Same for teachers and students:
they adapt lessons to the situation, tell the same
joke to different effect, etc. There are Endless
small forms of practical subversion taken up in the
name of getting work done (Suchman 1996:416).
Encouraging Organizational Learning
If successful practice and knowledge involves improvisation, then how do we encourage
their occurrence and transfer within an organization? That is, how do we engineer an organization that learns? There are many things you can
do:
1. First, you can value improvisational efforts.
If an organization ignores or devalues improvisation and rule-adaptation, then those
adaptations will happen anyway and as a
form of resistance to the formal organization.
Do not penalize improvisation, but look for
decoupling between routines and their improvised enactment where does that occur?
Where do the standard operating procedures
say one thing and personnel do another?
You should focus and revise those routines!
2. Second, create collaborative practices by
which useful improvisation is generated and

transferred. You should embrace improvised

practices and develop a means of noticing
which improvisations worked well and then
try to pass them along to others. For example, in the Xerox case, they had a help desk
that took calls from clients struggling with
their machines. Rather than ask the employees to look each question up in a manual,
they had experts placed at the same help
desk as novices such as giving them neighboring seats or the cell phone numbers of
Xerox experts In this way, they can overhear
experts, ask quick questions, and acquire
their built up organizational memory of adaptations and improvisations of the manual that
works. The valuation and learning of successful improvisation allows for continual
improvement and organizational memory to
be passed through participants. This example presages a third means of enabling organizational learning.
3. Put successful adaptations and knowledge
into organizational memory. How do you
retain the knowledge generated in collaborations? How do you pass on what works
well? Improvisational knowledge has an informal quality, a short life, and it fades from
memory! Then people reinvent the same
fixes again and again from scratch. Hence, it
is important to develop means of passing the
knowledge on and remembering it.
What social organization encourages improvisation and the generation of knowledge and understanding, and then makes sure it gets shared and
stored? An organization that supports collaboration does. It affords lateral linkages and opportunities to discuss work practices. Also, an organization that develops a practitioner database does: recognize and value practical knowledge creation,
and help members use it so it need not be rediscovered and reinvented every time.
So again
Xeroxs structuring of the help desk by placing
novices and experts together helped pass knowledge on to new personnel. In addition, in todays
technological world, we can use listservs and gen-

76

erate practitioner knowledge repositories that may

become an accessible database to others. Examples of this can be found at quora.com or even
stackoverflow.com. There you can post any kind
of question, but some of them are technical and/or
practical. For educators an example could be
found at curriki.org or tes.co.uk, where curricula
and lessons are posted.
Organizational Learning In-depth

For the remainder of the chapter, I want to

discuss various topics that are discussed from an
organizational learning perspective. The first topic
concerns learning curves. Here, organizational
learning experts ask, how do we know organizational learning has occurred? Lets use a school as
our example, since it is an organization most of
you are already familiar with.
What are some indicators schools could use
to denote that they improved and accumulated useful knowledge? Some examples might be test
score gains, attendance, and office referrals. We
might want to see teachers becoming more efficient at getting students to learn (say teacher time
per pupil [hrs] per instruction type x cumulative
gains). This is sort of what we would want to see.
Notably teachers might be inefficient at first giving students much time for little test score gain,
but then this
can improve.
The&Rela3onship&Between&Labor&Hours&Per&
Teacher
time per pupil (hrs)
Labor&Hours&Per&Vehicle&

Vehicle&and&Cumula3ve&Output&

Cumulative
test score gains
Cumula3ve&Output&

Figure - Learning Curves

One can even imagine curves like this for different instructional formats or even curricula. It

might be easier to envisage this for manufacturing

say for producing cars or airplanes and reducing
the number of complaints or recalls. But the same
could be done for schools in terms of the number
of arrests or office referrals here on the y-axis.
And then one can envisage this for multiple firms
or schools, discerning which has a steeper learning
curve, suggesting that the organization is reflecting on its performance and developing means to
improve or foster expertise.
How do you generate gains in learning
curves? What might be some tricks that can generate more effective/efficient ways to teach class lessons in a school? I raise this because a manager of
organizational learning will need to consider
means by which participants learn and improve.
For example, they may want to focus on improving
personnel such as getting fresh talent, preventing
initial startup costs via a mentoring program with
experts, or improving recruitment so as to attract
better talent. The manager may want to improve
work routines such as getting better-designed
tasks, removing stale ones, and allowing tasks to
become familiar (efficient). The manager may
also want to afford opportunities to discuss routine
improvements and document that. They can also
consider ways of improving their technology
such as getting better designed textbooks, develop
a better physical layout, and so forth.
Even if you do generate gains, it is important
to keep in mind that learning curves often plateau.
All too often, we learn to resolve the simplest problems first and acquire large gains. Then the improvements peter off and we get bogged down in
complex issues that have smaller gains. Hence, as
a manager of organizational learning, you may
want to shift your firms focus after learning begins
to plateau. For example, a school reformer may
want to move on from one reform to another, estimating learning curves and deciding when to put
in place organizational memory / stable procedures
to ensure gains are retained, but then switch to
some other concern.
Another problem with learning curves is that
they are only as good as the indicator used. Firms
tend to improve on the indicator and ignore other
issues. For example, it is common knowledge that

77

in schools with high stakes testing that teachers

teach to the test. But it is not always clear these
tests measure what we hope some tests measure
only a narrow band of intellectual development
and too strong a focus on any single exam may correspond with less time on other intellectual endeavors (like music, sports, physics, etc.).
Another important topic for the organizational learning perspective concerns organizational forgetting and memory: Why / How does
forgetting and remembering happen in organizations? What are the conditions of knowledge depreciation and knowledge storage? Again if we
consider schools, it is clear they are primarily forgetting organizations since what works well as an
instructional innovation in one classroom is seldom spread among other classrooms. But why?
How do they forget?
There are multiple reasons organizations
forget. Often it is because exogenous factors create distractions and prevent practitioners from recording what works (teacher strikes, lawsuits, odd
schedules, other events, etc). It is also because old
knowledge frequently becomes obsolete with new
audiences (old tricks no longer apply). Teachers
get out of date I need to learn the latest method
in order to stay current. A big reason schools (and
faculty) forget is because their personnel work in
relative isolation. Everyone gets their own room
and has little time to share what works well or
does not work well with their colleagues. I might
do something well, but no one ever hears about it.
Turnover is also an issue it can lead to loss of expertise. For example, in poor American schools
teacher turnover is dramatic. And even our treatments of sending in temporary teachers (like Teach
for America) results in little organization memory
and in the contexts that most need it!
So how can organizational memory be
stored within an organization like a school?
Technology/curricula are great for storing knowledge about successful practice, but it is not easy to
access and tends to remain relatively static.
(Wikis, annual reports are all visible, interactive
forms of knowledge, but they may not get picked
up very often or widespread in an organization.)
Successful tasks and routines can be encoded into

the organization but they are less stable than curricula. For example, team teaching might be a new
routine that has great returns in some cases, but
how it is enacted may vary greatly and the same
positive return may not be observed elsewhere.
Personnel (faculty) are great storage units and
transfer vehicles, but they leave and take knowledge with them if they leave! Organizational memory is not just a database of ideas, it is a database
of knowers with experience. You need ideas and
cultivators of them who are in the know, so retaining key personnel who train others is a very important means to engineering organizational memory
and especially if the sort of knowledge needed is
tacit or implicit and hard to codify and make sense
of in rule-books. Last, Cultural features like stories and community ceremonies can be great
means of preserving organizational memory, but
they might be prone to forgetting (as oral culture
can be) and it might focus too much on particular
individuals (exemplars and pariahs) than situations
so you might want a database created by working people. And again, with computing we now
have the capacity to collect and store practitioner
knowledge.
As I said before, websites like
Quora.com and curriki.org are feasible models to
use. There a searchable repository and demo presentations and materials can be stored.
Communities of Practice

Most discussions of organizational learning

mention the formation of communities of practice
and how they can facilitate and influence knowledge creation. When we consider communities of
practice, we are really asking about the microsocial processes that create new knowledge and/or
adapt and combine old knowledge in new ways.
From an organizational learning perspective, the
general argument is that in order to acquire knowledge, you need to enter a community of practice as
an apprentice. This community of practice is a
world entailing work, learning, and communication among people of a common working identity.
In this world of people who practice and identify
with what they do, you will hear stories and talk

78

about practice, and participate in applying and

adapting routines.
In a community of practice, learning is a
demand-driven, identity-forming, social act. As
such, it creates cohesive groups of persons working on the same task (~persons share bonding, cohesive forms of social capital). Here knowledge
can travel rapidly and be assimilated easily, but it
can also be coordinated or negotiated, and then
communicated in applied ways. By entering the
community, the participant enters strong, reinforcing bonds (bonding capital) that generate conformity and shared identity. This means members
identify with the organization, and it becomes
grounds for interpreting and judging, and reflects
an understanding.
A good example of entering a community of
reinforcing relations is when a person learns chess
and becomes increasingly involved in a chess
league. Most of us can not just learn about chess
from a book. We find it much faster and easier to
watch chess players play and then try playing it
with them by assuming the role of a chess player
ourselves. Over time, we may then enter leagues,
increasing our interactions with other chess players and develop further expertise (a ranked, master
chess player!). Over time, we may even become a
core member and take on the identity of chess
player as one central to our selves it may even
become a profession we embrace as our own.
The leap you need to make is in recognizing
that the same can be said for consultants, lawyers,
teachers, etc. Communities of practice are possible in a variety of organizations. How might we
generate a community of practice? Lets take the
example of a school again. Our goal is pretty
straightforward here we want to create a social
structure that encourages learning and remembers
what works well. To do this, we might want to Instill collaboration in a safe environment that allows for risks. We would want to provide training
to the entire faculty (not just part). We would
want to encourage meetings that entail sensemaking without decisions (remember garbage can
theory?). We want to encourage frequent communication whereby standards and procedures can be
learned. For example, we might want to denote

lead teachers and use them as experts in contact

with new teacher apprentices; we might want to
create mentoring and classroom observation opportunities; we might want to encourage storytelling /
cases from individual experience and organizational self-appraisal. Last, we would want to think
about ways to remember individual and organizational practices and knowledge (database), ways
to create a knowledge base (what people need to
know to do their work well), and how such knowledge can be distributed and interpreted (lots of
meetings concerning practice).
Now that you have some sense of what a
community of practice entails, how it is an asset to
a firm, and you have some ideas how to foster
their creation. But communities of practice are not
a panacea. Merely forming one will not result in
an optimal learning organization. COPs have certain shortcomings that we need to remedy or at
least supplement! COPs provide collaboration
without reach. Groups are also often homogeneous (heterogeneous has high startup costs). This
generates local maxima (not global) and multiple
equilibriums. Groups often only reach local solutions instead of best ones. And they are susceptible to groupthink (bias and uniformity that harms
organization). Negative social capital can also be
an issue: tight groups with wrong attitudes and
poor knowledge can be a disaster!
Networks of Practice

One needs to look outside the local community to form bridges with other communities and
prevent group think. To overcome these shortcomings, organizational learning theorists speak of networks of practice and knowledge transfer. Networks of practice (NOP) are like professional communities (secondary groups) where people may
never get to know each other but adopt similar
practices, similar resources, and similar identities
(technician, sociologists, etc). Here knowledge
about practice can travel rapidly and be assimilated readily. The reach of knowledge is greatly
expanded. Whereas members of a COP learn by
doing practices together, in the NOP, member

79

Source - http://www.flickr.com/photos/86530412@N02/

Networks of
Practice

learn about ostensive rules by way of books and inter-organizational networks (Learning by talking and sharing).In contrast with COP, NOPs have
reach. They span COPs. The inter-COP linkages is viable because members share identities. This allows actors to communicate in relatively similar
ways (info sharing across groups that bridge capital).
How might we generate a network of practice? One way is to Headhunt for experts in other firms. You poach talent, so to say. Another way is
to send your personnel off for training in a new technology (Bootcamp or
summer school!). Many firms will perform reverse engineering of a product
they will look at another firms product and take it apart looking to understand how it can be made for their own COP. Firms can also build NOP by
making sure people transfer across units (people across departments), products, and even organizations. Last, firms can employ people who bridge
communities of practice and facilitate knowledge transfer across them. In
schools, one can find this with professional development leaders who work
at multiple schools trying to retrain teachers.
However, just like COPs, NOPs have shortcomings:
1. No community is had, only reach.
2. More learning about than learning to be.
3. Local adaptations are less of an emphasis.

80

In many ways, NOPs and COPs need each other.

It is in their combination and integration that many
organizations develop practices by which they can
continually improve and strive toward global optima of performance.
Exploration, Exploitation and Learning Traps

Many scholars regard organizational learning

to be a varied process, and potentially dysfunctional. For example, James G. March writes
about organizational learning can proceed by a
process of exploration and exploitation, and either
route can result in learning traps or suboptimal
forms of decision making (March 1991).
When March discusses learning by exploration he means the process of searching, generating variation, risk-taking, experimenting, play,
flexibility, innovating, etc. (the process of generating new practices). In some of the case materials
by Louis and Kruse, we see them describe schools
and school reform as frequently stuck in an exploration mode. There are lots of great ideas but none
of it really sticks or matches what we need! (Louis
and Kruse 1998: 31).
When March discusses learning by exploitation he refers to the process of refining, choice,
production, efficiency, selection, implementation,
and execution (the process of eliminating inferior
forms). Here, the organization attempts to improve by repeating the same task again and again.
Notably, a firm that constantly explores can not
really get good at a task as it has not really practiced it much. Conversely, a firm that constantly
exploits, gets good at performing one task, but it
does not see new ways to enact it.
This leads March to reflect on learning traps
that many organizations encounter. One suboptimal form of learning arises from what is called a
failure-trap. Here, an organizations failure can
lead to exploration. And because most exploration
often fails, the firm can get trapped in a negative
feedback loop of failed explorations! An opposite
form of learning trap is called a competency
trap, and it arises from positive feedbacks. Competency traps can arise in two variants:

(i) If feedback is positive, you stay in exploitation

mode (e.g., a short-term, local solution) and
never search for a better solution (e.g., a longterm, global solution).
(ii) The more you become proficient at a rule /
practice, the better you get at it, so you are
more likely to use it again and again. Positive
feedback makes the substitution of another rule
/ practice less likely and when you do switch,
you bungle it from having focusing your training so much on one skill.
If firms want to avoid learning traps and to become a successful learning organization, they need
to balance exploration and exploitation and beware
of learning traps that can put push them into suboptimal situations.
Applications of Organizational Learning
Now, lets briefly review the theory of organizational learning and then introduce a couple
cases for you to ponder. The first case concerns
the implementation of organizational learning in
schools, and it can be found in the writing of Louis
and Kruse (1998). The second case concerns the
World of Warcraft and how guilds operate in that
context.
Before we begin analyzing these cases, we
should first review the basic features of an organizational learning perspective. When does the theory apply? When does organizational learning happen? Organizational learning occurs in an organization when the participants are continually concerned with improving their practice. They are focused on the core technology how the organization turns an input into an output. As such, it constantly monitors, reflects on, adapts, and remembers practices that work well. In some cases, learning is suboptimal or even false, and this too is of
relevance to the study of organizational learning.
The general perspective of organizational
learning is to view an organization composed of
practices that form the core routines of organization, and to zone in on an organizations intelligence or capacity to alter and improve them. This
enriches the participants identity or role and fur81

thers their commitment to the organization. From

this perspective, the organizational elements are as
follows:
1.

The technology or means by which org learning occurs is via internal adaptation, where
actors alter routines to fit local realities (what
Brown and Duguid called knowledge).

2.

The participants are members within the organization performing the routines and enacting practices.

3.

Their goal is to resolve application problems

to improve their practice so that it better accomplishes defined goals and identities.

4.

The social structure entails mostly informal,

lateral relations, frequent communication, negotiation and dialogue. Identities and roles
are key and closely linked or coupled with
practices. Participants are involved in both a
community of practice entailing local bonding
ties and peer pressures, as well as networks of
practice that span out to other communities
and facilitate knowledge transfer.

5.

And the environment is a source of interorganizational knowledge, tricks, and transfers.

What is the dominant pattern of inference, or

the mechanism of inducing action? Action is the
result of local actors searching, improvising, collaborating, translating and sharing. Through an organizational learning approach, change and improvement occurs because the individuals and the
groups inside the organization are able to acquire,
analyze, understand and plan around information
(or knowledge) that arises in their practice and
the wider environment. They continually adapt
and learn.
And finally, the theory affords some managerial implications. To garner a learning organization, the manager should consider ways of encouraging dialogue, continual improvement of core
practices, and improvisation. They should find

ways to create greater communication within the

firm so ideas are passed and shared, and they
should find ways to create bridges to outside
groups so they can access distinctive forms of
knowledge. They should also find means of creating organizational memory of what works so it is
retained.
Case: Agassiz and Okanagon

Now that we have reviewed the basic features

of an organizational learning perspective, we can
begin to discuss some applications of the theory to
real world cases. The first case is mostly described for us in the reading (Louis and Kruse
1998), while the second I will summarize, but
leave mostly for you to consider on your own.
The first case affords a description of schools
that implement organizational learning. In this
reading, the authors describe two exemplary
schools doing better than expected. They say both
schools are learning schools where the faculty reflect and study their practice in an effort to continually learn and improve.
According to these
authors, learning schools share an inventory of
prior knowledge about the school, its curriculum,
instructional methodology, and students. Learning
schools know themselves and take the time to develop a shared vision and vocabulary with which
to discuss issues of teaching and learning.
The first school Louis and Kruse discuss is
Agassiz Elementary. In Agassiz Elementary, the
faculty learn from each other and engage in a dialogue about their instructional practice and how
they can improve it. The principal (as manager)
tries to stimulate and encourage such dialogue.
She acts as a facilitator of knowledge more than a
director. The teachers also seek to learn from each
other, and engage in dense webs of frequent conversation over practice -- a COP of sorts. The
teachers frequently interact in weekly grade level
meetings, monthly meetings for Kindergarten to
3rd grade and 4-6th grade teachers so they can think
more broadly; and 30 minutes a month of teacher
observation. Hence, teachers have frequent, close
relations over practice. This results in a good deal
of peer pressure to improve instructional practice
82

Organizational Elements
Technology
Advisory Council of teachers and parents, Curriculum Committee, Reading
(what brings
Recovery, grade level teams (6) and faculty study groups (2) to coordinate
about org
curriculum, flexible staffing, teacher involvement in hiring, team teaching.
learning)
Restructuring Roundup for conferences ! all develops teacher interaction and
within-school networks of practices.
Participants
Teachers, Principal, Parents
Goals
Increase student learning and improve teaching
Social Structure
36 teachers (moderate size) organized in a semi-horizontal fashion (see
committees above). Relations are collaborative teachers are part of a team;
they have some decision-making power and leadership roles. Mrs. Cole, the
principal, is an intellectual leader. While listening to everyone, she still makes
some decisions autonomously.
Environment
Context of Desegregation and White Flight. District reforms include openenrollment magnet schools, school-based management, relief from state
curricula, personnel and testing regulations. 650 students.

Agassiz School Organizational Elements

and a culture valuing constant improvement (so
much so that the teachers even pay to attend conferences and join groups that meet on weekends
and evening). The school also holds a conference
on professional development that teachers from
other schools can attend. This not only brings in
money, it compels the Agassiz teachers to assume
the identity of knowledge producers and expert
educators.
If we take the case and render it into our organizational elements we will see where the
authors place the greatest emphasis (see the table
on the next page). The technology in this instance
is the tools used to bring about organizational
learning. In many ways, these are all social structural treatments: e.g., a variety of meetings, the decentralization of authority and greater input from
teachers, etc. All of the means of engaging in
learning are relational and cultural. The participants of this case are members of the school staff
and some are parent. Students are not really mentioned. The goal of Agassiz is to increase learning
and improve teaching and they hope to do this

by reflecting on their practice. The social structure is small, intimate, and the relationships are collaborative. The principal (leadership) facilitates
and mentors more than imposing her will. Moreover, her focus is on practice. Last, what we know
about the environment is related in the setup of the
school and its history. But the case makes little of
it. Instead, the case zones in on practice, social relations, and rituals fostering reflection and improvement on it. Some of these relations extend
into the environment, but only in order to draw in
or send out knowledge on instructional practice.
Now if we focus on management, we see that
the school instated several routines and institutional arrangements to foster such a learning community. These in turn are feasible managerial
strategies to use in other settings.
The second school Louis and Kruse discuss is
Okanagon Middle School. Okanagon Middle
School is much larger than the elementary school
and with just as disadvantaged a population. Okanagon school is divided into 9 families (also called
small schools); students and faculty of core sub-

83

Organizational Elements
Technology
Interdisciplinary Units - teachers collaborate on the units and work to fit the
(what brings
curriculum to their students learning needs. School divided into ten
about org
Families Core teacher family (science, history, etc.) and Discovery family
learning)
(language, band, special ed). School-wide Evaluation Committee. Community
Council for curriculum with teachers and principal ! all develops teacher
interaction and within-school networks of practice.
Participants
Teachers, Principal
Goals
Educational equity and opportunity for poor urban children.
Social Structure
84 teachers (large) organized in a horizontal fashion. Relations are
collaborative teachers have important decision-making power. Teacher
discretion over staffing, schedules, some resources, and even aspects of the
curriculum. Mr. Stone, the principal, is important but not as autonomous as in
Agassiz.
Environment
Okanagon Community School initially closed due to poor achievement.
Reopened as Okanagon Center for Advanced Academic Studies, a magnet
school focused on performing arts. 1,500 students.

Okanagon Middle School Organizational Elements

jects are assigned to families and work within

them. These families have wide discretion over
what they want to work on and improve. Each
family also has a leader with an expanded teacher
role that includes administration and mentoring of
other teachers. All the family leaders get together
as a community council with the principal to ponder the schools direction more generally.
The school has a strong culture. It proudly announces it has a dream to level the playing field
for its students, and render the content of curricula
more relevant to their students. The school is very
concerned with statewide assessments and the faculty have established a variety of powerful school
wide committees that perform self-assessments
and pressure the faculty to perform well on such
exams. But the school also has its own standard of
sorts the Okanagan Standard where students
are called upon to perform community service and
conduct research projects. External ties are important at Okanagon, and trips to external conferences
are common, but they rotate across faculty rather
than centering on any single faculty member.

Hence, information is found elsewhere and presented to the rest of the faculty.
There is some issue of coordination across the
families and some debate as to where organizational learning should focus and what standards to
pay special attention to. The school holds a yearly
retreat and the families do find some topics of
agreement. As a result, there has been a push for
greater school-wide coordination. This has created some tension, but it seems to be helping.
If we look to the organizational elements
again, we can begin to see how Okanagon is similar to and different from Agassiz (see table above).
The technology as before - are the tools used to
bring about organizational learning. These again
are all social structural reforms: the school is divided into smaller family units, they form a variety
of different committees and councils that encourage frequent interaction and assessment over practice and achievement. All of Okanagans means of
engaging in learning are relational and cultural just
as it was for Agassiz. The participants of this case
are members of the school staff. Students and par-

84

ents are not really mentioned. The goal of Okanagon is slightly different in that it has a equality and
social justice concern more than Agassiz did. The
social structure is large, but dividing into smaller
units. Relationships are collaborative and the
teachers / families have great influence, whereas
the principal at Agassiz had greater say. Last,
what we know about the environment is related in
the setup of the school and its history. But the
case makes little of it. Instead, just as with Agassiz, the case on Okanagon zones in on practice and
the social relations, culture and rituals fostering reflection and improvement on it. Some of these relations extend into the environment, but only as a
means to drawing in or sending out knowledge on
instructional practice.
Looking at its management, we see that the
school instated several routines and institutional
arrangements to foster such a learning community.
They encourage constant improvement and features of both community of practice (COP) and networks of practice (NOP).
So what do we see in general at these reputed
Learning Schools? The schools frequently seek
out internal and external bases of knowledge in
both local peers and experts beyond the setting.
They both have processes in place that help transfer individual knowledge and expertise (e.g., Agassizs prof development showcase). Both schools
create knowledge via self-appraisal and selfassessment (Okanagons focus on state testing and
the creation of their own standards is evidence of
this.). Both settings search for expertise and
knowledge beyond the school and they seek to disseminate their own beyond their own walls.
Hence, teachers read and attend external groups
and then report back, demonstrating, running seminars, etc in their own school on those topics. Last,
there is systematic learning via structures that facilitate constant contact (Agassizs grade level
meetings, Faculty Study Committees, Restructuring Roundup; or at Okanagons many committees,
Curriculum Committee, Evaluation Committee,
Portfolio Committee).
In the end these schools are cases for how
features of a COP and NOP can be formed, heightening identities and worker commitment, and gar-

nering improvements in organizational performance. It also sounds like a lot of work! - these
teachers are showing up on weekends and staying
late into the evening (4-7pm) so as to improve
their practice. Louis and Kruse argue they arent
experiencing burn out, but is this sustainable?
Will they eventually burn out? Or is this a model
that will sustain commitment and fulfill identities?
Now that we have some sense of an application to
real organizations like schools, what about its relevance to an organization online, in a less traditional case of an organization, like say, the World
of Warcraft?

Figure. World of Warcraft Game.

(Source:
http://commons.wikimedia.org/wiki/File:World_of_Warcraft_-_Mists_of_Pandaria_Box
_Art.jpg)

World of Warcraft

Some of you may have no idea what I am talking about here, so let me explain. The World of
Warcraft (WOW) is a massive multiplayer online
role-playing game (MMORPG) that was created
by Blizzard Entertainment. It is currently the
world's most-subscribed MMORPG (9.1 Million).
85

The game itself is extremely intricate with many

options and rules. Players can pick races, professions, etc. I believe there is even a currency that
players pay real money for. The main goal is to
interact, go on quests, acquire wealth, power and
experience, and so on. Many people play this, and
they play it often. They log many hours a week in
addition to their day-jobs.
One of the main objectives in WOW is to complete quests. Many of these quests are difficult to
accomplish. The monsters are too strong for a
small band to overcome, or the problem is too intricate to solve without a large collaborative effort.
Hence, characters often form guilds groups of
100 (small) to 200 (large), and they are like communities. In the communities the players chat, coordinate quest efforts, etc. They also develop identities as players and as a team.
Here is a screen shot of what it looks like
when you are a lonely Orc in WOW. And here is
an emblem of one of the guilds on WOW.

http://commons.wikimedia.org/wiki/File:b/be/Ware_guild_original_
logo.png

John Seeley Brown does a nice job of discussing the WOW and why it is a sort of learning
organization. You can view his video directly
here:
http://www.youtube.com/watch?v=BhuOzBS_OM
Brown argues that the guilds in the World of Warcraft resemble a community of practice and network or practice, and that they are such a learning
organization that they are able to confront a multitude of very complex problems, to coordinate their
efforts, and to collectively learn and remember
what worked well so they can train new personnel
to go out and collaborative solve the same and
new problems on their own. Rather than recount
Browns argument, take a look yourself and see
what you think!

Figure. Screen shot of Orc in World of Warcraft

http://commons.wikimedia.org/wiki/File:Male_orc.jpg

Figure. Guild emblem from the World of Warcraft

86

87

Unitary actor or team that

confronts a problem, assesses
objectives (goals) with regard to
it, identifies options, the
consequences of said options,
and then chooses option that
minimizes costs.
Variant: Bounded rationality
and satisficing. Recognize
imperfect info, ambiguity, and
select first satisfactory option
(good enough).

Summary or Basic
Argument

Not salient except as

influencing consequences of
options.

Action = Maximization of
means to ends.

Know alternatives and their

consequences for the shared
goal, and select wisely. Improve
information and analysis.
Management by consequences.

Dominant Pattern
of Inference

Management
Strategies

Know SOPs, what problems they

go with (matching), and who cues
them. Improve rules and matching
with problems. Management by
rules.

Action = output close to prior

output (path dependence), cueing
of SOPs appropriate to problem.

Actors in hierarchical
organizational positions. Cue
sequential routines that accomplish
task or solve problem by routines
available (supply issue).
NA

Formal roles, hierarchical.

Environment

Objectives compliance to SOPs,

match with problem parts.

Goals are defined in regard to

problem.

Goals
(what probs to
resolve)
Social
Structure

Organizational positions

Unified team or actor

Matching identity and SOPs

(solutions) / programs / repertoires
to problem.

Dividing up problem,
coordinating / activating
organizational actors who have
special capacities / SOPs for parts
of problem, conducting sequential
attention to objectives (localized
searches until problems resolved).
Action guided by processes /
available routines.

Exists when the decision is guided

by a logic of appropriateness
matching problem to actors with
procedures for handling it (routineprocess focus).

Organizational Process (OP) /

Limited Problem Solver (LPS)

Participants

Key Organizational Elements

Technology
Maximization of options
(how solutions
(solutions).
get decided)

Exists when there is a unified

actor with consistent
preferences, lots of information,
and clear goals (and time
calculate).

When does it
apply?

Rational Actor (RA)

Summary Table of Five Theories to Date:

Bargain with players (log-roll,

horse-trade, hinder oppositions
coalition formation, etc). Learn
others interests / weaknesses so
you know how to manipulate and
win. Direct management of
relations via bargaining.

Action = result of political

bargaining.

Deadlines and wider array of

stakeholders.

Coalitions enemy/friend

Parochial priorities, goals/interests,

stakes / stands.

Players in positions

Bargaining, or playing the game

(within its rules), or political
maneuvering.

Focus on the players occupying

various positions; their parochial
interests (their conceptions of
problems and solutions); their
resources (expertise, money,
people) and stakes in game; and
bargaining processes between them
that establish agreements /
coalitions.

Exists when there are multiple

actors with inconsistent preferences
and identities, and none of whom
can go it alone without assistance
of others.

Coalitions /
Bureaucratic Politics (BP)

Time when your solution is raised (to

coincide with right participants and
cycle of problems) to maximize energy;
abandon entangled initiatives; know
how to overload system for policies you
detest; and generate choice opportunities
that work to your interests
(access/timing). Indirect managing of
situations.

Action / decision = result of streams

collision in choice arena.

Deadlines and other choice arenas (e.g.,

decision in current arena may be means
of access to another choice arena)

Confluence of multiple streams, such

that solution is connected to problems
and enough actor-energy to see it
through.
Participant stream shaped by political /
career cycles & unplanned departures.
Problems stream determined by public
opinion, prominence / vocalness of
problems in firm, etc.
Access rules segmented, hierarchical,
or democratic.

Focus on choice arenas (when choice

opportunities / windows arise); the
distinct and decoupled streams of
problems, solutions, and participants;
and their access rules to the arena
(whether structural or timed).

Exists when solutions are unclear,

participants turn over, and
preferences/identities are inconsistent.

Organized Anarchies /
Garbage Can (GC)

Find ways to create lateral ties among

workers so knowledge is passed /
transferred more readily / quickly (if
possible, quickly), create means to
organizational memory of what works.
Create applied, social learning experiences
with means to retaining and transferring
expertise. Want communication, collective
improvisation, practice and knowledge

Action = result of local actors collaborative

search (trial & error / transfer) and adapting
rule to situation.

Informal, lateral relations, communication,

negotiation, & collective improv. Actor
identities (demand) important. Network of
practice (professional identity / reach) &
community of practice (cohesive group).
Source of inter-organizational knowledge /
tricks / transfers.

Members of organization doing work /

SOPs
Application problems pattern recognition
not there (no fit).

Internal adaptation, or where actors alter

routines for the better and fit reality
(knowledge).

Acknowledges routines, but focuses on

practices within them that enable their
continual adaptation and change to fit
reality i.e., practices reflecting
organizational intelligence.

Exists when there are clear feedback loops,

adaptations, memory, and support of actorexpertise / adaptations of rules to local
reality.

Organizational Learning (OL) /

Knowledge-Practice Model

Inner City Schools. Chapter 2 (pp. 17-46)

References:

in Organizational Learning in Schools. Tokyo: Swets & Zeitlinger.

Argote, Linda. 1999. Organizational Learning: Creating, Retaining and Transferring

Levitt, Barbara and James G. March.

Knowledge. Boston: Kluwer. (Concretely

1988. "Organizational Learning. Annual

addresses many org learning themes).

Review of Sociology 14: 319-340.

Brown, John Seely and Paul Duguid.

March, James G. 1991. Exploration and

2000. Practice Makes Process, and

Exploitation in Organizational Learning.

Learning in Theory and Practice. Chap-

Organization Science, 2, 1: 71-87.

ters 4-5 (pp. 91-146 [and endnotes appended]) in The Social Life of Information.
Boston, MA: Harvard Business School
Press.
Brown, John Seely and Paul Duguid.
1991. Organizational Learning and
Communities-of-Practice: Toward a Unified View of Working, Learning, and Innovation. Pp. 58-82 in Organizational Learning. Eds. Michael Cohen and Lee Sproull

March, James G. 1994. A Primer on Decision Making: How Decisions Happen. NY:
The Free Press. Chapter 6, pp. 221-272.
March, James G., Lee Sproull, and Michael
Tamuz. 1991. Learning From Samples of
One or Fewer. Pp. 1-19 in Organizational
Learning. Eds. Michael Cohen and Lee
Sproull (also listed in Organization Science, 2(1), Feb. 1991.). London: Sage.

(also listed in Organization Science, 2(1),

Orr, Juliann. 1996. Talking about Ma-

Feb. 1991.). London: Sage.

chines: An Ethnography of a Modern Job

Leithwood, Kenneth and Karen S. Louis.

1998. Organizational Learning in

(Collection on Technology and Work). ILR

Press.

Schools: An Introduction. Chapter 1 (pp.

Suchman, Lucy. 2007. Human-Machine

1-8) in in Organizational Learning in

Reconfigurations: plans and situated ac-

Schools. Tokyo: Swets & Zeitlinger.

tions, 2nd edition. New York: Cambridge

Louis, Karen Seashore and Sharon D.

University Press.

Kruse. 1998. Creating Community in Reform: Images of Organizational Learning in

88

6
Organizational Culture

(Source: http://farm3.staticflickr.com/2339/1807614622_61a0cdc0f1_z_d.jpg)

Organizational Culture
I often ask students in my class to perform
an exercise where they got together in small
groups and developed designs for a learning
school. Each group came up with different interesting designs. Their main efforts focused on creating opportunities for discussion about the core
technology of instructional practices -- e.g., having
small schools within schools, as well as gradelevel meetings, departmental meetings, and so on;
opportunities to transfer ideas -- e.g., speaker series, training sessions, rotating teachers through
assignments, mentoring programs; and means of
establishing organizational memory data collection, storage, and analysis, rule formation, mentoring, and so on.
All of their design suggestions established
interactional settings and routines through which
faculty could discuss and study their practice.
They forged a system wherein they could continually self-assess their performance and make sure
their core technology worked well. That said, a
problem became readily apparent. Their designs
assumed teachers would work extra hours, that
there were resources to fund their training, and
where resources were lacking, it was assumed that
the stakeholders would all come together and pick
up the slack. It became clear that a key assumption of the organizational learning approach was
that everyone shares the same values, is willing to
work extra hours, and is on the same page reformwise. In effect, it assumed participants all share
and buy into the same organizational culture. But
what is an organizational culture and how do you
study it? Thats what we will cover in this chapter.
Exploring Organizational Culture
Most of you recognize organizational cultures when you see them. Take Google for example theyre here in Silicon Valley and employ
many thousands of people, some of whom are in
this class from year to year. The company has a
clear logo, and it is emblazoned on all their
shwag, from pens, to shirts, to cars, to even lava

lamps. And those of us in the area have heard

about the availability of excellent free food, play
areas in the work space (like ping pong or bowling), the large campus, park-like environment, and
even Google bikes readily available to Google employees to use and share as they move from building to building. We also know about the long
hours the employees work, the great benefits, the
casual atmosphere, and the seemingly endless commitment many of them show for their firm. In
short, it has a culture and we can see it.
Within an organizational culture, actors
make sense of their existence according to identities and norms, and these are often constructs afforded by the organization they are in. Think of
the culture at firms like Apple or Facebook all
have an identity and norms surrounding their performance of it. As such, the motive in an organizational culture is the expression and fulfillment of
an identity a strong intrinsic motivator! An organizational culture entails normative (valued) and
cognitive (implicit) aspects of organizational social structures. These are deep structural facets
that guide interaction. If we could only control
this and engineer it, we would have zealous workers! We would have the sort of worker buy in that
organizational learning seems to need! However,
we will look carefully at Gideon Kundas book, Engineering Culture, to put into question the organizational culture ideal. Kunda offers a more nuanced
account of organizational culture: in making the
organizational culture the focus of engineering,
one renders it something that controls and repressesa means of capturing souls!
What is an organizational culture?
What does it mean to engineer an organizations culture? For managers, organizational culture is a gloss for an extensive definition of membership in the corporation that includes rules for
behavior, thoughts and feeling. These add up to be
a well-defined and shared notion of the member
role. Culture is seen as the vehicle by which we
can influence the behavior and experience of others something to be engineered via making pres-

90

entations, sending messages, running bootcamps,

writing papers, giving talks, etc. The culture is a
mechanism of control! You cant make them do
anything; they have to want to (Kunda: 7). Engineering culture is the ability to elicit, channel, and
direct the creative energies and activities of employees. By engineering organizational culture,
we create a membership role in the firm that employees embrace as their own identity and self!
Lets look more closely at the concept of organizational culture and define it. Organizational
culture is generally viewed as the shared rules governing cognitive and affective aspects of membership in an organization and the means whereby
they are shaped and expressed. The traces of an
organizational culture are the shared meanings, assumptions, norms and values governing work behavior; they are the symbolic, textual, and narrative structures in which norms and values are encoded; and it is found in the structural causes and
consequences of cultural forms and their relation
to organizational effectiveness (Martin and Meyerson 1988).
Engineering an organizational culture is a
managerial strategy. Organizational culture is a
means to normative control: it is an attempt to
elicit and direct the required efforts of members by
controlling the underlying experiences, thoughts,
and feelings that guide their actions. Under normative control, members are driven by internal commitment, strong identification with company
goals, and intrinsic satisfaction from work. In
short, it is the employees self that is claimed in
the name of corporate interest! (Kunda: 11)
Lets spend a little more time discussing observable features of organizational cultures. I
think this is important because all too often, the
discussion of culture can quickly seem abstract. I
want to make sure you see grounded, real features
you can point to in organizations when you consider what its culture is and how to re-engineer it.
In particular, I am going to draw heavily on Martin
and Meyersons work on organizational culture,
since they afford the level of concretization I prefer (1988; Martin et al 2004).
So how do you study organizational culture?
What are the elements of a culture? One thing we

can focus on are practices. If you recall, practices

are a central concern of the organizational learning
approach. The difference here is that we do not
presuppose persons in an organization buy into
and align with the practices being enacted like organizational learning does. With an organizational
culture perspective, we look carefully at how participants relate to these practices and express themselves through them.

Figure - Script for Dancing (Dance Notation)

(Source - http://commons.wikimedia.org/wiki/File:Feuillet_notation.jpg)

In many instances, the traces of cultural practices can be found in formal scripts or rules of conduct. When we reflect on societal cultures we
have certain things in mind, like a code of etiquette, or a procedure or script for dancing. We
may also notice informal customs that emerge and
are not planned, such as customs of style. Below
is a diagram showing changes in skirt fashion
showing hemlines have risen, thereby showing
changes in style.

91

Figure - Skirt Fashion

(Source http://commons.wikimedia.org/wiki/File:Hemline_%28skirt_height%29_overview_chart_
1805-2005.svg)

What are the parallel practices within organizations? Within organizations, these practices can
be formal policies, rules, roles, and procedures
like job descriptions, pay distributions, performance assessments, and so on. Examples of formal
policies can also be found in organizational charts
(here the rules are about positions and their relation). But formal policies can also be standard operating procedures (e.g., like rules for promotion,
or as here, rules for processing prisoners). Or
even manuals for operating software or codes of
conduct in an organization. All are formal policies.

Figure - Formal Organizational Chart

(Source http://commons.wikimedia.org/wiki/FileOrganizational_chart_of_Headquarters%2C_D
epartment_of_the_Army.gif

Figure - Manuals
(Source http://commons.wikimedia.org/wiki/File:KC85-HBS.jpg/800px-KC85-HBS.jpg)

Organizational practices can also be informal

customs like norms of communication, customs
of style and conduct, how conflict is managed, and
habits of interaction. For example, workplaces
may have different customs of when to talk, how
to address each other (Bill or Mr. Jones), how
to argue (kindly or hotly), and so on. Some workplace customs even concern dress codes and fashion styles. For example, in Silicon Valley, there is
a trend for executives to wear colorful socks. Or
to wear those shoes that look like gorilla feet.
Such behavioral and stylistic norms emerge and
are not planned.
Other cultural elements are artifacts these
manifest in multiple forms. For example, societal
cultures often have certain symbols like religious
relics and art, historical buildings, or the images
placed on regional postcards. They also reference
certain tools, like the ancient artifacts and technologies the region invented or utilized in important
historical events (teepees, arrowheads, ionic columns, bronze spears, cannons, etc). The parallel
symbols in organizations are their logos, and the
parallel tools are the computers and magnetic resonance imaging machines used within the workplace. Each signifies the organization and what it
does (technology or medicine).
Cultures also entail rituals like dances,
games, activities, Organizations also have rituals
but they tend to be different activities and encoun-

92

ters. For example, organizations often involve

meetings and presentations of work. These can be
of varying forms and styles, reflecting a particular
form of collegiality (whether strict and hierarchical or loose and friendly).
Cultures frequently entail stories people tell,
and even language or jargon (dialects) that acts as
code differentiating them. Organizations too have
stories we all know the story told of Facebook
and its founding, as told by the movie: The Social
Network. We also know the story of other founders, like Waren Buffet of Berkshire Hathaway.
But this also extends within the firm, to heroic
teachers and pariah deadbeats. Firms even have
their own jargon. Google, for example, has a series of terms they use on their campus to refer to
various types of employees:

rangement that reflects distinctions of culture and

meaning.

Noogler (new google employee)

Loogler (employee in legal department)
Gaygler (lesbian, gay, bisexual and transgender google employee)
Spoogler (spouse of googler)
Zooglers (employees in Zurich)
Xooglers (ex-Google employees)
The use of special acronyms too are common. I
cannot begin to tell you how many acronyms Stanford uses for its research centers on campus, many
of which I cannot deconstruct into their compositional terms (SLAC, ICE, SHIPS, FSI, E-IPER,
etc, etc).
Martin and Meyerson also argue that cultures
are qualified by physical arrangements, such as
architecture and placement of objects. One can
readily comprehend this when comparing say a Cathedral to a Quaker meeting house both are Christian religions but of very different architectural
styles. Even the seating is very different. In the
Cathedral, seating is hierarchically arranged while
in the Quaker house it is arranged for dialogue.
Within an organization, we can see similar variation in physical arrangement. The differences in
office building versus campus layout; the difference between closed impersonal cubicles and open
desks. Again, we see differences in physical ar-

Figure - Physical Arrangements

( S o u r c e - h t t p : / / c o m m o n s . w i k i m e d i a . o r g / w i k i / F i l e : R e d b o x _ O ff i c e . j p g ;
http://commons.wikimedia.org/wiki/File:Office-Cubicals-5205.jpg)

Last, Martin and Meyerson argue that cultures differ in the content themes they highlight.
Content themes are abstractions used to organize
interpretations of an organizations practices and
artifacts. One can see these content themes or cultural abstractions being used in the 2012 American
Presidential Election. We hear two candidates
voicing ideological themes that highlight their different values and beliefs (they express normative
arguments vote for life and deregulation versus voting for choice and the middle class.
They express ideational themes that concern interpretations about the meaning of events. For example, that current unemployment is a result of the
prior presidents economic policies or this presidents; that economic reports are a sign of improvement or a sign of continued problems; and so on.

93

Figure - Content themes of last US presidential election

(sources - http://commons.wikimedia.org/wiki/File:Romney_Skidmore.png/; http://commons.wikimedia.org/wiki/File:Jeri_and_Ann_Romney.jpg;
http://commons.wikimedia.org/wiki/File:Bush_Inauguration08.jpg/; http://commons.wikimedia.org/wiki/File:Obama-harding.png;
http://commons.wikimedia.org/wiki/File:Michelle_Obama_official_portrait_headshot.jpg; http://commons.wikimedia.org/wiki/File:RWB-donkey.png)

Cultural Content
Themes

At each political partys convention, we saw how these content

themes are externally presented to the public (public face). But it was
also interesting to see both candidates try to portray content themes as internally held as well. For example, Ann Romney knows the real Mitt is
funny; or that Michelle Obama knows the real Barack Obama is the same
guy he was 4 years ago. And these portrayals are made in spite of, or
counter to, what most of us see in the public image.
Within a firm, this internal perspective on content themes is usually
the inside view on say a Google: i.e., what it is really like to live and
work there (what we see backstage and in private, and that can be reinforcing or undermining). These cultural elements -- whether practices,
roles, procedures, rituals, stories, jargon, symbols, tools, physical arrangements or even content themes all begin to form a mosaic of a culture.
The culture becomes a system of these meanings and affords us a larger,
holistic sense of what that organizations culture is.
Martin and Meyerson portray organizational cultures as amassing
in certain types of paradigms or styles: they call them integrated, differentiated and ambiguous. The most common assumption is that organizational cultures are Integrated, recognizable and uniform. This requires
reinforcing elements where ideology and practices and themes all align
(so consistent elements are mentioned, organization-wide consensus exists, and members deny there is ambiguity). Management and public relations often espouse a uniform view but it is seldom present for long.
The integrated culture hides conflict and tensions (repressed existence).
94

Integrated

Fragmented

Ambiguity

Figure. Metaphorical Images for Types of Organizational Culture

(Sources clockwise starting with the integrated figure - http://commons.wikimedia.org/wiki/File:4/4a/Uniformity%2C_plantation_on_Black_Hill%2C_Forest_of_Alyth_-_geograph.org.uk_-_1433738.jpg;
http://commons.wikimedia.org/wiki/File:Galapagos_archipelago_250m.jpg; http://commons.wikimedia.org/wiki/File:Confusion_of_Tongues.png)

The different types of

organizational
culture: integrated,
fragmented, and
ambiguous.

That said, it is possible there are entrepreneurial firms and schools where
all the employees are committed to a common vision and purpose/ideology
they may actually have an integrated organizational culture.
Martin and Meyerson (1988) explore whether a single firm, OZCO,
has an integrated effort to develop egalitarianism. For there to be an integrated culture of egalitarianism, they would need to identify a series of cultural elements that reinforce and support this claim. Hence, they find that
the firm publicly claims to be egalitarian, has formal and informal practices in place to encourage it, has various stories, rituals, jargon and physical arrangements, that all seem to reinforce and support the existence of
egalitarianism.
A second perspective of organizational culture is that of differentiation or fragmentation. Here, one can regard an organizational culture like
an archipelago or as having different groups or camps with their own perspective and culture. Rather than a uniform culture, there is a differentiated one.
Turning to Martin and Meyerson again, we see them look for instances where egalitarianism is not seen uniformly and there are questions.
For example, egalitarianism as an affirmative action ideology does not fit
the perks and hiring practices being used. In many ways the differentiated
system is conflicted and has countering efforts, or at least efforts pulling
the organizational culture in different directions.
Schools are great examples of such a decoupled system: the administration of a school tends to show the external environment a schools test

95

scores and extracurricular activities but it does not

talk to teachers about educational process (the internal perspective is not the same as the external
one looking in). As a result, different subcultures
can emerge, and this creates consensus within subcultures but not across them. What results is inconsistency, and channeled ambiguity. Is differentiation a more accurate view of organizational cultures in most firms? Do these different cultures
exist in conflict or harmony?
Differentiation
might work better for a highly differentiated context like a multinational firm, or a steeply hierarchical organization.
Last, we have the view that organizational
cultures can be ambiguous, unclear and confusing.
In my mind, I see this sort of organization as one
that gives mixed signals (like this sign is the
road closed or open for wide loads?); or that the
organization can be seen in several ways (e.g., gestalts); or everyone is speaking different languages
and it is hard to make sense of things (e.g., The
Tower of Babel is perhaps a silly allegory here, but
it might help you grasp the concept).
In an ambiguous organizational culture the
elements are unclear and confused. If we return to
Martin and Meyersons example of OZCO we
find that the ideology of egalitarianism is confusing to some, there are unclear procedures and confusion on how to implement things. As such, ambiguity equals lack of clarity, lack of consensus, and
confusion over what things mean and how to do
them.
By comparing the cultural paradigms, we get
a better sense of how they differ. Each has certain
defining characteristics that distinguish them. For
example, an organizational culture practiced by integration only mentions consistent elements, exhibits consensus across the organization, and denies
ambiguity. In contrast, cultural differentiation entails some inconsistency, exhibits consensus within
subcultures of the firm (not between), and channels ambiguity (denying it for their own subculture, but seeing it in others). Last, a culture of ambiguity lacks clarity, has issue-specific consensus,
and frequent confusion. Ambiguity here is acknowledged. One can envision these cultures as

seeming to be a hologram, an archipelago of cultures, or a jungle.

Figure - Ambiguous Sign

(Source http://commons.wikimedia.org/wiki/File:Ambiguous_Sign_-_geograph.org.uk_-_1391048
.jpg)

It would be easy at this point to think that an

integrated culture is most desirable, but perhaps
not for everything. After all, many cults and instances of group-think do not end well! And in
some ways, I get the impression that organizational learning presupposes an integrated organizational culture. For it to work, it often seems to require an organizational culture that is much like a
cult. However, in other instances of organizational learning they espouse the need for improvisation and this might require an organizational culture that lacks clarity.
96

In contrast, there are reasons to believe an

ambiguous or differentiated culture may be more
useful. An ambiguous and differentiated culture
can create inconsistencies and confusion, but they
can also afford variation and be a hotbed for innovation. An organization situated in a dramatically
changing environment may do well to be differentiated or ambiguous so it can more readily adapt and
survive. Also, I get the impression that an ambiguous organizational culture is the type characterized
in the theory of organized anarchy and the fragmented culture is the type characterized in coalition theory. In many ways, organized anarchy suggests manager should embrace ambiguity as it is
there where creativity can arise. Similarly, coalition theory suggests managers should embrace differentiation and bargain ones way to success
within such a context.
My point here is that different paradigms of
organizational culture may be more or less useful
to a firm and its situation. It is not clear an integrated form will be the one most desired in every
context. Also, I think many of you can see that the
organizational theories we discussed earlier may
have a proclivity toward one form of organizational culture over another. This may also confuse
you a bit and make you wonder how organizational culture differs from organizational learning.
If you recall, organizational learning seemed intent
on developing a particular set of interactions and
practices that led to a self-aware and learning organization. It took certain surface structures and
sought to implement them so they would change
deeply held beliefs and understandings (deep structure) and this would result in organizational participants who continually improve their practice.
By contrast, the organizational culture approach is agnostic as to what kind of culture is
best it may all depend on the context. Moreover,
it begins with deep structures, like content themes,
rituals, symbols, etc, and sees how they influence
surface interactions, work relations and company
performance. So in some ways, organizational culture fleshes out what practices are to a richer extent and identifies how they form larger gestalts or
systems of meaning that guide behavior. These
cultural systems come in many forms, only some

of which may be a learning culture! Others may be

an egalitarian culture; or a self-fulfillment culture,
etc. Moreover, it is feasible that ambiguous and
differentiated forms may be advisable under certain circumstances.
Gideon Kundas Engineering Culture
Next, we will discuss Gideon Kundas text,
Engineering Culture (1992). I do not know
which company the pseudonym Tech refers to.
In fact, there are so many high tech firms in Silicon Valley that resemble Tech, that most of us will
recognize this kind of culture (and that is Kundas
point). Here we have a case study of an organizational culture formed in a high-tech company that
seems to influence control and commitment to the
corporation. Kunda approaches organizational culture in many of the same ways Martin and Meyerson describe. But his primary focus is on interactions, and his main tool is ethnographic observation.
Organizational Culture as Ideology
Kundas focus is on the context of normative
transactions: managerial conceptions of culture,
how it is enacted, and the responses of members .
As such, Kunda views organizational culture as an
ideology. He sees cultural enactment in terms of
rituals enacting the ideology and instilling it. And
he focuses on how members negotiate their need
for distance and embracement of the culture and
its rituals.
Kunda regards organizational culture as a
means to normative control controlling the
hearts and minds of employees! The management
is seen as defining organizational ideology
(Kunda: 52). For members, the company perspective on the culture is familiar, systematic, comprehensive, thought-out, well-articulated, and associated with the companys interest. This depiction
of organizational culture as ideology is consistent
with anthropological conceptions of ideology, such
as that of Clifford Geertz all ideologies are schematic images of social order publicly offered in the

97

name of those with a claim to authority as maps of

problematic social reality and matrices for the creation of collective conscience (Geertz 1973:220;
Kunda:52).
So what authority gets across in an organizational culture? Whose image of social order is offered and practiced?The inscription of the organizational identity falls into three distinct categories
each of which derives its authority from a different
source. First there is managerial authority which
derives its authority and influence from the documented views of senior managers, the company
philosophy, taped speeches of the CEO, company
mission statements all framed in terms of morals
and ideals! A second form of authority is expert
authority. This type of authority emanates from
technical papers, reports, and memos that internal
experts write. The third form is one of objective
authority. This type of authority is comes from selective representation of materials produced by outside observers of Tech, such as news clipping, TV
ads, etc. All of these forms of authority combine
to create a company perspective and ideology.
Their influence is additive and compounding.
Managerial authority is obviously ritualized
by senior management, and reflects their views of
Tech. These managers focus on the attributes of
the collective (Tech, us) as a way of lending the
members a moral significance as well shared goals
and history. The managers present their views
through their speeches, interviews, and editorials,
and these give personalized and animated views of
Tech ideology, complementing and fleshing it out.
They build a we sense by referencing the past,
their mission, and shared values, as well as identities of self and others. Membership in the community is presumed to define ones social existence
and personal experience. You not only assume a
role, but incorporate it and become it, making it a
part of your self or so that is how the senior management performs it.
The image is that there is no conflict between individual and company goals (an integrated paradigm). The organization claims to
give employees a place to grow and develop, a
moral order to participate in, and simultaneously
sustains the company and affords members a

meaningful identity. Individualism is a way to

serve the collective interest, and heavy selfinvestment in the company affords personal returns and freedom (greater autonomy and authority). Personal meaning is derived from participation in the collective.
A good place to see this is in the company
documents on goals and missions. They are all
catch phrases and abstract ideals (e.g., mom and
apple pie). They entail things no one would disagree with and only want they would want to emulate. They characterize their members as creative,
hard working, good people.
It is not just Tech who does this. Take a completely different company of Levi jeans and their
mission statement.
People love our clothes and trust
our company. We will market the
most appealing and widely worn
casual clothing in the world. We
will clothe the world.
(Source - http://commons.wikimedia.org/wiki/File:7/75/Levi%27s_logo.svg)

As you can see, it is related in moral and normative terms with mentions of a strong relationships
with customers, mentions of trust, product quality
and universality. What is not to like about it!?
Expert authority is mostly enacted by internal experts. Such experts focus more on the requirements and attributes of a members role. As
insiders, experts give an aura of independence,
practicality, and scientific credibility. A good example of this expert view and identity portrayal
can be seen in the Kundas account of a native anthropologists study of Tech culture (Ellen Cohen).
Her register of speech is often open, pragmatic,
and critical/helpful so seemingly balanced. Her
moral tone is not evident and the ideological faade is acknowledged some. This view is consistent with a managerial perspective but it is less
ideal and more real. The expert even acknowledges downsides and her prescriptions are pragmatic. Role performance is more based on personal success and self-help. That said, the expert
is still viewed as partisan by employees.

98

The creation of objective authority comes

from outside the firm from academics, consultants,
and journalists. Organizations typically decide
which of these perspectives to relate and share,
and they are mostly positive. These accounts tend
to be edited, selective reviews of the company
from outside that are used to reinforce things further. Academic pieces seem to offer an objective
view that the company members (mostly professionals) are oriented toward the firm and its culture (science). Popular books tend to get closer to
the managers ideal but from an external basis.
They relate the zeal / culture to enhanced performance claims (common sense). Last, journalism is
the most widely seen form of media clippings
are posted often focused on CEO giving a rational
actor imagery.
Many similar themes are addressed, but negative / critical pieces are not
shown (journalistic observations).
All three of these views compound to form
one integrated view of Tech! membership in
Tech entails heavy involvement, strong bonding to
company, and zeal leading to collapse in boundary between self and organization. This accomplishment is seen as leading to economic success,
and is accomplished by designing an environment
based on individual autonomy, informality, minimal status distinctions, and seeming disorganization (Kunda:88).
The company culture and ideology is enacted
and instilled in members via presentation rituals
of an organizational self. Presentation rituals occur everywhere in the participants everyday lives
the performance of such rituals...
is a framing device: members acting as
agents of the corporate interest attempt to
establish a shared definition of the situation within which reality claims derived
from the organizational ideology are experienced as valid (Kunda:154).
These rituals are used as vehicles for the exertion
of symbolic power that defines reality.
I know that may seem like a lot of jargon to
some of you, but mull it over a bit. What I mean
here is that every time a Tech employee or man-

ager does a presentation, voices a slogan, or interacts in a meeting, they act as an employee (not as a
father or mother) and as an agent of the firm.
Even people listening in the audience play their
role complement, expecting professional behavior
and a style of interaction that makes the everyday
reality of living in Tech seem different from elsewhere but seemingly valid and natural.
We can see Tech rituals everywhere. We just
have to look. If you recall Martin and Meyersons
focus on cultural elements, you will see many of
the same elements discussed in Kunda. Ritual
presentations of self are most often observed in
persons behavioral displays. At many organizations, these are time-bound interactions specific to
a particular audience and setting. In these interactions, we see people present and attempt to establish a positive definition of their self. They wrangle and maneuver so as to do a good job and to
come off in certain ways. We see these displays
most frequently in presentations, question-&answer sessions, and meetings (notably, all are decision arenas).
They are also in the mundane, private, everyday chatter at lunch, in back offices, and at the water cooler. We can also see presentations of self in
artifactual displays when we walk by work spaces
or observe participants dress. These are standing
exhibits of self meant for passerbys and bystanders. At Tech these exhibits are found at desks
where they display personal mementos, Tech stuff
and humorous jokes about the company.
One can see distinct types of artifactual display at Stanford depending on which department
you walk through. For example, walk down the
hallway of the law school and the computer science departments. At the law school their offices
resemble a lawyers office with cherry wood, Lshaped desks, neat shelves, and so on. In addition,
they dress relatively formal in comparison to the
rest of campus. By contrast, in the computer science department, the faculty offices are casual,
toys and equipment are strewn about, and the professors dress in t-shirts and sneakers (or flipflops). A very different notion of organizational
self exists in those two parts of campus and one
can readily infer it from mere standing exhibits!

99

We can capture and record behavioral and artifactual displays in a variety of ways through interviews we get personal accounts of self, through
observation and recording we get a record of talk,
interpersonal behavior, and exhibits. Through active note-taking and involvement, we can even
form understandings of these encounters as if we
are participants (as opposed to foreigners). All of
these devices help us compile evidence on how ritual interactions shape the workers organizational
self.
Upon observing many such interpersonal rituals and speaking with Tech employees, Kunda
comes to observe a persistent pattern or style to
these interactions. Tech rituals have at least two
features:

1. Tech rituals are characterized by a decentralization of power. In everyday rituals,

power arises in the shifting environment of
different speakers, reputations, projects,
teams, and so on. And these seem to entail
many speakers, changing projects, and shifting reputations so power seems decentralized.

2. Tech ideology is one of openness, informality, individual initiative and real feelings.
Hence, symbolic power is often exerted subtly. It is revealed in brief episodes of social
drama, like question & answer sessions in
talks where some individuals seem to establish authority (and if you recall that can be
of several forms managerial, expert, and
external).
Some people just come off
smarter, they project a self and statements
others identify with, and in character-jousts
like disagreements, or debates they tend to
win.

At Tech, these mini-dramas of control are an

ever-present part of presentation-rituals. The dramas follow a predictable pattern there is a challenge, rising tension, and then actors acting in the
corporate interest use various techniques to suppress and redefine dissent, silence the deviants,
and gain support. In short, it is through micro-

rituals in meetings, talks, presentations, and the minor disagreements and gaffes, that persons come to
exert norms of behavior and guide presentations of
self so they reflect and reinforce Tech culture.
Individual Reactions to Organizational Culture
Thus far, we have described how Tech culture is a normative culture developed and imposed
as a means of normative control. The company engineers the culture to acquire greater worker commitment and to increase worker efficiency. And
this is accomplished by having members enact a
variety of behavioral displays or interpersonal rituals, where standards and identities are assessed
and redirected in ideological ways. In this manner, the firm hopes to go deep into the persons psyche to have them embrace their organizational
self as their virtual one and from this all sorts of
company gains will result. But how do Tech employees react to these presentation rituals of self
and the seeping in of an organizational culture and
identity? Are they fine with it? Do they dearly
value the organizational self they portray? Or do
they feel like a tool, or like they are just playing
a part? Do they resist and play an ambivalent self?
How do they respond to Tech culture?
Kunda writes that employees respond in
several ways. The most common outcome is the
expression of role-embracement here, this embracement is expressed whole-heartedly in talks by
top level management; it is reserved and tentative
in training workshops; and pragmatic / conflictual
in work group meetings. Participants who embrace their role (like managers) may experience
some emotional dissonance. In those instances
their perception of an acted role and the experience of an authentic self, become hard to disentangle. They find it hard to have an identity distinct
from the one they have at work.
A second reaction is to engage in role distance. Here, the Tech employee suspends their
role-embracement in the process of performing behavioral displays. We have all seen individuals
suspect their formal role and organizational self.
When I teach, I may drop my teacher role at the
beginning or end of class, during transitions and

100

timeouts, and so on. These brackets intersperse

my presentations and are opportunities for rolereversals: I can talk to students as a peer, joke,
mention my kids, and so on.
At Tech, it is not so much a roleembracement as role-distancing, however. This is
when members assume a reflective and openly
self-conscious stance. They comment on their condition and the ritual performance itself. When this
occurs members temporarily detach themselves
from their performance of the member role, comment on it, and share with others the awareness of
the theatrical nature of the proceedings. They put
colorful labels on behavioral scenarios. They say
they are setting up, that they do not want to engage in pissing contests, backstabbing, crucifying, hanging by their shoestrings, or engage
in hidden agendas, etc.
The act of distancing oneself from your
presentation and manager role does not invert the
hierarchy like my jokes might, but rather they confront the meaning of authenticity (who is real
and who is not) and inclusion (who is in and who
is out) that is being enacted. This is actually
more important than many realize. By enacting
role-distance and taking a self-aware stance on
your talk and role, you show you are a person distinct from it. I am just a guy playing my part like
you are! By doing this, we connect and there is a
communion among the self-aware and talented actors who comment on their roles and performances
(Kunda: 158).
Some Tech employees are aware of all this
and show a great deal of social skill and elegance.
That is, they have the controlled ability to shift
stances and frames. This ability to shift stances is
key because members evaluate each other on their
ability to express both embracement and distancing and knowing when to stop (Kunda:158).
Kunda calls this a contrived self because
participants enact rituals with an explicit awareness of the dramatic mechanisms that underlie the
process of framing reality, and an open acknowledgement of the manufactured nature of cultural
categories and symbols, including those that are
central to the ritual performance itself. And this is
where it gets interesting! The self-consciousness

that can be seen as a fatal flaw is now itself ritualized! This creates a potentially unstable balance
between role-distance and embracement that constantly calls into question the authenticity of experiences associated with the member role for persons targeted by normative control.
Presentation rituals are vehicles of enacting,
enforcing and reinforcing the sanctioned display of
member roles and are thus a mechanism mediating
normative demands and responses. The mediating
role of rituals is not simple though! They can juxtapose a variety of themes and stances: for example,
they can juxtapose ideology and common sense;
notions of obligation and choice; seriousness and
humor; affirmation and denial; internal and external viewpoints; participation and withdrawal.
Switching between embracement and distance
forms a web of normative pressures.
In the end, you have to wonder if a strong
organizational culture leaves room for individual
freedom of expression. What is real and prescribed here about your self? Even the contrived
self one that you accomplish by social skill and
by switching between embracement and distance
is something the organization prescribes and rewards. Think of all the managers who effectively
do this performance and dance. They are not that
unique! It is sort of like the rebellious kid. Sure
they rebel, but they rebel much like all the other
kids and are not that unique.
Not all members are invested equally in
Tech. Some members are marginal, like temps
and what Kunda calls wage 2 class earners. They
are not subject to same role demands and organizational ideology, so they are exempt to a degree because demands on their self are reduced. At same
time, it makes some of them feel left out and they
develop an estranged view.

101

Figure - Images of a Wage 4 and Wage 2 Worker

(Source - http://commons.wikimedia.org/wiki/File:Robert_schonberger.jpg;
http://commons.wikimedia.org/wiki/File:thumb/Marissa_Mayer.jpg/)

Other members are fully invested. These are

typically higher status members of the firm, or
what Kunda calls wage 4 class employees. These
employees must meet the enormous demands of
their member role and face a fundamental dilemma. By seeking acceptance and higher status,
these actors expose themselves to greater demands
of the organizational ideology. They must sell
their soul! The price of power is submission! Low
status participants merely submit their behavior
but not their attitude. The high status participant
gives up their behavior, thoughts and feelings!
This can lead to a cynical view as these members form a contrived self.
Balancing Demands of Organizational Culture
To this point you may be thinking the following: if I join a firm with a developed organizational culture and I want to get to the top, then I
may find myself being brainwashed! Luckily, all is
not lost. Even if you play the part of a contrived
self and are cynical about your self and the firm,
you can do some things that leave room for your
authentic self. The organization need not take
over all of who you are!
We all have multiple identities and selves. I
am one thing at home with my kids and another at
work with my students. In both instances, I may
show role-distance and reveal my character or
something about me as a person independent of
those identities. The organizational self for Tech
managers is one that arises from balancing accep-

tance and rejection of the organizational ideology

and the member role it prescribes (Kunda:161).
You can prevent the contrived self from being your authentic self in several ways. First, you
can manage time. Tech work takes lots of time
and energy. This blurs the distinction between
work and non-work. As a response, people create
boundaries around their time and their relationships that develop at work. Non-work time is sacred, protected, and kept separate. Workers can
also define their authentic self by what they want
to become, or in spheres outside work.
Second, you can manage your response to
the organizational self. Many workers regard
over-involvement as a problem. They believe that
having a fair exchange with the company is desirable, and anything else is undignified
(Kunda:177). In fact, role-distance is often condoned.
Cognitive distancing and disputing popular
ideological formulations is viewed as a good idea.
You need to be autonomous enough to know what
is going on in the company and dignified enough
to express such knowledge. Otherwise, you will
be seen as a zealot or a tool. Employees do this
by being (i) cynical and complaining a lot; by performing (ii) detached theoretical observations (using a lens like a scientist or researcher), and by
(iii) adopting a common sense perspective (they
effectively try to view the organization from an alternative frame of reference).
Workers can also distance themselves emotionally with respect to their feelings. This can
happen by (i) denial. Here, they claim their motives for membership are purely instrumental, like
for money, and deny an emotional attachment.
Emotional distance can also arise from (ii) depersonalization. In these instances, they distance
themselves from emotions experienced at work.
They say they are have a thick skin, or talk
about their emotions abstractly as pain or warm
fuzzies etc, and they do not take things personally. Last, they can regard their emotions in terms
of (iii) dramatization. Here they view emotional
expression as strategically driven. It is used to accomplish goals, and therefore suspect of authenticity.

102

In sum, actors engage in two efforts at selfpreservation: (1) they attempt to control and stake
boundaries to their other selves by managing time
and separating work from non-work; (2) and they
seek to control their cognitive and affective responses at work when they are enacting their organizational self.
So lets sum up Kundas argument. According to Kunda, organizational cultures are a means
to normative control or an ideology. The ideology
is enacted and instilled in members via presentation rituals. These rituals are like layers of control
plied on. Lower status workers are under utilitarian control (they want pay!), but higher ranking
workers are under cultural and utilitarian control.
They sell their selves to the company! Now this
might be desirable or not. And most every true
member of an organization performs some roledistancing. In so doing, they free up other features
of their self independent of the company. But
even so, the higher one goes, the more roleembracement is needed and the distancing becomes part of a contrived self and an act.
But ask yourself something - is Kunda viewing organizational culture as a cup half empty
when maybe we can see it as a cup half full? If
I do not embrace my organizational self, then I
must be embracing another self in other spheres of
my life. For example, one might embrace being a
youth league soccer coach. Why is that organizational self more sacred? What if my organizational self at Stanford also serves some good? Is it
ok then? Or perhaps Kunda is saying any roleembracement has this quality of becoming more
and more of our virtual self? And that with any
role we fully embrace, we eventually assume a
self-referential perspective on it. But then this
process is merely descriptive of our being in an organizational world and how we manage our selves
more generally in todays society.
I do not have all the answers. I just know
Kunda hit on something profound. We want to create an integrated organizational culture and for employees to embrace it in many of the organizations
we hope to found and manage. And yet we have
this precarious relation with our self when participating in such an organization.

103

104

Acknowledges routines, but focuses on practices within

them that enable their continual adaptation and change
to fit reality i.e., practices reflecting organizational
intelligence.

Summary or Basic Argument

Internal adaptation, or where actors alter routines for the

better and fit reality (knowledge).
Members of organization doing work / SOPs
Application problems pattern recognition not there (no
fit).
Informal, lateral relations, communication, negotiation,
& collective improv. Actor identities (demand)
important. Network of practice (professional identity /
reach) & community of practice (cohesive group).
Source of inter-organizational knowledge / tricks /
transfers.

Action = result of local actors collaborative search (trial

& error / transfer) and adapting rule to situation.

Find ways to create lateral ties among workers so

knowledge is passed / transferred more readily /
quickly (if possible, quickly), create means to
organizational memory of what works. Create applied,
social learning experiences with means to retaining and
transferring expertise. Want communication, collective
improvisation, practice and knowledge sharing to arise.

Technology
(how solutions get decided)

Participants

Goals
(what probs to resolve)
Social Structure

Environment

Dominant Pattern of Inference

Management Strategies

Key Organizational Elements

Exists when there are clear feedback loops, adaptations,

memory, and support of actor-expertise / adaptations of
rules to local reality.

When does it apply?

Organizational Learning (OL)

Summary Table of Resource Dependence Theory (RDT)

Find ways to confer ideology and lead others to identify with

it (using a variety of practices and artifacts), but dont make
it so explicit / fanatical that cynicism emerges. Give room
for autonomy and self-expression so distancing is
unnecessary, and encourage members to generate a culture
of their own (~org learning culture NE to Tech culture which
is top-down engineered).

Action = result of deep structure or culture that is generated

in the organization, but which is mediated by the members
relation to it.

Many elements of culture have origins from outside, and

they are transported in, then translated to the local culture.

Create intrinsic motivation (sense of fulfillment), and

remove differentiation / cynicism in most cases.
Deep structure composes the elements of culture themes
(beliefs & norms), their expression via practices (rituals,
etc), and their manifestation or expression in artifacts
(reports, mission statements, etc).

Matching, sense-making / meaning-making, or where actors

seek to express beliefs, norms, and values via a variety of
practices and externalize them in artifacts depicting shared
understandings / notions of appropriateness.
Actors within the organization, and those salient to meaningmaking.

Actors seek expression and fulfillment of identity, and

organizational culture is the medium for such
expression/sense-making.

When the cognitive and normative aspects of social structure

are of concern and seem to guide organizational decisions
(sense-making) and outcomes.

Organizational Culture

References
Geertz, Clifford. 1973. The Interpretation of Cultures. Basic
Kunda, Gideon. 1992. Engineering Culture: Control and Commitment in a High-Tech Corporation.
Philadelphia, PA: Temple University Press.
Martin, Joanne. 1992. Cultures in Organizations:
Three Perspectives. New York: Oxford University
Press.
Martin, Joanne and Debra Meyerson. 1988. Organizational Cultures and the Denial, Channeling
and Acknowledgment of Ambiguity. Chapter 6
(pp. 93-125) in Managing Ambiguity and Change,
L. Pondy, R. Boland, and H. Thomas (Eds).
Martin, Joanne, Peter J. Frost, and Olivia A.
ONeill. 2004. Organizational Culture: Beyond
Struggles for Intellectual Dominance. To appear
in S. Clegg, C. Hardy, W. Nord, and T. Lawrence, (Eds.) Handbook of Organization Studies,
second edition, London: Sage Publications.
Van Maanen, John and Gideon Kunda. 1991.
"The Smile Factory: Work at Disneyland." Pp. 5876 in Reframing Organizational Culture (eds. Peter Frost et al). Sage.

105

7
Resource Dependency
Theory

Environment
Tech
Core
Firm
(Source: http://commons.wikimedia.org/wiki/File:Circular_buffer.svg/)

Resource Dependency Theory

We now begin focusing on theories that are
primarily concerned with the environment and
how they influence organizational effectiveness
and survival: resource dependence theory, network
organization, and neoinstitutional theory. These
theories and conceptual frameworks are relatively
recent contributions to organizational research,
most emerging in the literature in the 1980s onward. All of these theories provide what Scott
calls an open systems perspective on organizations (Scott 2003). Each of these theories will argue that there is no single best way to organize a
corporation or to make decisions. The optimal
course of action is always contingent (or dependent) upon the external situation of the firm. As
such, the best way to organize a firm depends on
the nature of the environment to which the organization relates. The theory we will discuss in this
chapter is Resource Dependence Theory, and it
views an organization in terms of its resource dependencies with other firms in the environment.
Resource Dependency Theory Compared
Given we have already covered a series of
different theories in the course, it may help to contrast them with the theory of resource dependence.
So let me review a few and compare them: we will
look at coalition formation, organizational learning
and organizational culture, and we will discuss
how they differ from the resource dependence perspective. In the earlier chapters, we learned about
coalition formation and what it entailed. We
learned that coalitions arise when multiple actors
have inconsistent identities and preferences, and
none of them can go it alone without the assistance
of others. We read about players having their
own interests and resources, and how they had to
negotiate (or exchange and bargain) until they
reached an agreement by which coordinated action
could follow. We also learned that coalitions
could be managed and formed through various
processes of exchange and bargaining like horsetrading and log-rolling. These exchanges were all
pair-wise, or dyadic, and they aggregated within a

group to form a shared goal and agreement. The

time frame on these exchanges and agreements
were narrow as the coalition agreement was often fleeting.
Resource dependence theory is similar to coalition theory in that it concerns exchange and efforts to produce agreements. However, it differs
from coalition theory in at least two important
ways. First, it shifts the unit of analysis from coalitions of persons to inter-organizational relationships of dependence. Here, the concern is with a
focal organization and its multiple resource dependencies with other organizations in the environment. Second, while coalition theory focuses on
narrow windows of time specific to each transaction, resource dependence theory concerns extended forms of exchange, or exchange relations.
An organization can form a wide variety of buffering or bridging maneuvers used to overcome persistent dependencies in the environment. For example, you will learn that when a company merges
with another, it is often a means of absorbing dependencies and acquiring a degree of autonomy in
the environment.
These dependence relations can also be asymmetric in fact, managers of resource dependence
actively seek ways to render other firms dependent
on them, but not vice versa. So with resource dependence theory, we have an egocentric view of an
organization trying to acquire the best exchange
relations it can in an environment of many potential partners. In prior weeks we also discussed organizational learning. If you recall, organizational learning focused on how organizational participants adapted their practices within the firm as
they engaged in the process of doing their work.
This was facilitated by efforts to encode best practices into organizational memory and by communicating about practice in local communities of practice and by communicating outwardly in networks
of practice beyond the organization.
Managers try to develop employee concern
with improving practice and by forging social relations and interactions that facilitate knowledge experimentation and transfer. Most of the emphasis
lies in local adaptations of routines and as such,
the argument is that internal application (learning

107

by doing) is the main means to understanding and

expertise.
Resource dependence theory has some similarities with organizational learning. Like organizational learning, resource dependence theory focuses on the technological core of an organization.
However rather than describe the internal process
of practice improvement and knowledge transfer,
it describes how the technological core of an organization is buffered from the environment. Resource dependence theory describes how the organization (as a sort of unitary actor) bridges with
firms in the environment so as to garner autonomy
and control. Hence, concern is placed on becoming effective in an external environment and by establishing certain SOPs for resource exchanges
with other firms. So the focus shifts from a mostly
inward view to a mostly outward one.
In the last chapter we discussed organizational culture, and there the goal was to create an
ideology or culture that members identify with personally, and managers used all sorts of strategies
(rituals) to make that happen. Now of course, its
possible that different paradigms of organizational
culture (integrated, fragmented or ambiguous) will
apply best to your firms goals or context but the
general argument is somewhat similar to that of organizational learning: adaptation is internal to the
organization and not focused on external relations
outside. Whereas for organizational learning, the
effort was to generate relations and practices, here
the effort is to engineer deeper social structures of
cognition and norms. Here, managers worry about
internal contingencies, like layering on a culture
too thick and having organizational members reacting in resistant ways. For example, you recall
Kundas worry about generating cynics. Managers
have to balance the effort to prescribe a culture
with allowing participants room for their selves.
Otherwise, the participants relation to the culture
will undermine its effect.
As such, organizational culture is inherently
concerned with the process of sense-making and
ritual performance. Standard operating procedures
are viewed as practices, and deeper, broader sets
of practices than perhaps organizational learning
relates. By contrast, Resource dependence theory

is not concerned with sense-making but with the

selection of SOPs that manage the firms resource
dependencies in the environment. In a way, resource dependence theory is a step back toward
the organizational process model. It brings our
theories back up to the surface of ostensive rules
and routines, and away from deeper forms of
sense-making. Managers form and select SOPs
that concern relations in the environment; and they
seek relations that create favorable exchanges or
favorable consequences. So resource dependence
theory is also a shift back toward a logic of consequence in certain regards.
We can also discuss the prior theories in this
textbook more generally as natural systems, as
compared to the open organizational system being
characterized in resource dependence theory. A
good example of this can be found in how prior
theories described organizational uncertainty. It
was something that arose within the firm, from inconsistent preferences, identities, unclear rules,
routines, practices, and so on. Resource dependence theory is also concerned with organizational
uncertainty, but it sees uncertainty as residing in
the firms external relations of interdependence.
When external dependence relations are not managed and coordinated well, they create uncertain
conditions (if not unfavorable conditions) for the
firms survival. Prior theories also regard dependence and uncertainty differently from resource dependence theory. For example, for coalition theory, dependence is not a problem but something
sought after to make the coalition hold. And uncertainty or ambiguity is often reason for why a coalition stays together.
By contrast, in resource dependence theory,
the firm tries to accomplish autonomy and certainty, and it does this by freeing itself from dependence on other firms and by forging contracts.
Hence, whereas uncertainty and dependence are an
asset to coalition formation, they seem to prevent
firms from acquiring an advantageous resource position. In sum, the shift from immediate local exchange conditions within a firm, to externally sustained exchanges in the environment, seem to have
different consequences and implications for each
of our theories.

108

An Overview of Contingency Theory and Resource Dependency Theory

Lets now briefly discuss the history and

core features of the theory of resource dependence.
Resource dependence theory, in part, grew out of
contingency theory. Therefore it helps to understand the core features of that theory before going
further. Contingency theory was a class of organizational theory from the 1950s through the
1970s that argued a firms optimal course of action was contingent upon the internal and external
situation it found itself in. As such, contingency
theory offered a natural and open system view of a
firm. Perhaps the most complete characterization
of contingency theory can be found in Thompsons
work (1967). He describes how firms need to
buffer and protect their technical core from all
sorts of internal and external disturbances that can
disrupt its functioning. He affords several prescriptions on how to minimize these contingent problems:
For example, managers need to seal off their
technol\ogical core and buffer it from internal and external influences.
Managers can prevent and reduce environmental uncertainty by distinguishing both the
input-acquisition functions (such as supply)
and output-disposal functions (such as sales)
from the technical core.
Internal strategies of the firm might include
stockpiling and smoothing, or internalizing
uncertainty through growth (thereby absorbing uncertainty).
External Strategies include maintaining alternatives and minimizing dependence. Some
specific aspects of this include cooptation,
contracting, and coalescing (like joint ventures).
Resource dependence theory builds off contingency theory and greatly elaborates on maneuvers
firms can use to manage disturbances in the external
e n v i r o n m e n t .
Resource dependence theory was founded by
Jeffrey Pfeffer and Gerald Salancik. According to

Pfeffer and Salancick, organizations modify their

boundary so as to manage disturbances in the external environment. The firms central goal is effectiveness in a context or environment (e.g., survival). This is different from Organizational Learning where internal efficiency and improvement is
the focus.
Resource dependence theory is primarily
focused on relations with the external environment, rather than on ones within the firm. As
such, resource dependence theory views organizational conditions in a particular way. It presumes
there is environmental determinism. This means
an organizations behavior can be explained by
looking at its context, such as external constraints
and controls. It assumes an organizations specific
goals are contingent on dependence relations keeping it alive (i.e., the relationships that secure its
necessary resources). Within this context, the
firms general goals are to find greater certainty
and autonomy. From this it follows that organizations respond to resource dependencies in at least
two ways: they comply and adapt to dependencies
or they avoid & manage them.
What are the core features of resource dependence theory? One of the most important features of the theory concerns the resources involved
and how they establish dependencies. To identify
resource dependencies, it helps to ask - What are
the key resources in an environment? Who controls the resources in question? Resources come
in a variety of forms, they are valued differently
depending on their importance and availability,
and they differ in terms of who has discretion and
control over them.
There are various types of resources firms
depend on, such as physical materials. These
might be actual materials the organization builds a
product from. But firms may also depend on technical resources like information or knowledge as
well. And last, they may depend on social resources, like prestige and reputation, that enable
them to survive. All these resources can vary in
value. On the one hand, their value can differ by
the importance of the resource. Is it in demand - is
it valued? Does the firm need the resource to survive? Is there a critical resource? For example,

109

Physical)

Technical)

Social)

!
!
!
!
!
!
!
!

!
!
!
!
!
!
!
!

!
!
!
!
!
!
!
!

h#p://upload.wikimedia.org/wikipedia/commons/
thumb/0/07/Brake_shoe_materials.jpg/800px@
Brake_shoe_materials.jpg!

h#p://upload.wikimedia.org/wikipedia/commons/
5/54/InformaEon.png!

h#p://upload.wikimedia.org/wikipedia/commons/
thumb/a/a4/
Tokyo_University_Entrance_Exam_Results_5.JPG/
800px@
Tokyo_University_Entrance_Exam_Results_5.JPG!

Types of Resources
What does Stanford need to survive? Does it need students? Does it require a physical location, books, teachers, students, money, heating,
food? What can it live without? Is there demand for safety, healthy food,
expert teachers, and awards? For example, what is the greatest demand
of Stanford alumni it is sustaining high SAT scores, winning national
championships (prestige)? What resources are considered most and least
important?
On the other hand, does the availability, or supply of the resource
influence its value? Is the resource scarce? Do only some of the other
organizations have it? How concentrated is the resource? Are there alternatives to the resource? Can another kind of resource be substituted for
it? Who else has it? Lets consider Stanford again what does it offer
that is unique and that no other can provide? Discretion over a resource
also defines relations of resource dependence. Discretion is defined in
two ways: First, who controls the resource? Can the exchange partner dictate how you use the resource? Is the resource regulated by the government (changing districting to increase resource / student pool)? Is your
firm dependent on the supplier (materials and funds) or consumer (students / families)? Second, what controls dependencies? (laws) Are there
copyrights or contract licenses (curriculum)?
In sum, resource dependence varies from a variety of factors: there
are different types of resources, and they can vary in value due to their importance and availability. And then certain actors and institutions can control discretion over them. Now clearly, important and rare resources are
of greater value. Moreover, actors and institutions that have the greatest
discretion over them (and least amount of dependence) will be the most
110

autonomous and capable for forging certain relations with other firms in the environment.

Figure. Resource Value (Demand and Supply)

(Source - http://commons.wikimedia.org/wiki/File:Supply_and_demand_curves.svg/)

Managing Resource Dependence

Once you understand the value attributed to
a resource and the game of resource dependence
which is to be autonomous and establish beneficial
resource relations with the environment (on important ones you control!) - a variety managerial
strategies follow. Some of the first managerial
strategies echo contingency theory: protect the
technological core from the environment via buffering strategies like coding, stockpiling, downsizing (or what some call, removing slack), advertising (which showcases strengths), an so on.
None of these buffering strategies change the core
task and technology of a firm. They are more concerned with putting SOPs in place to manage the
organizations boundary. Lets take each strategy
one at a time:
1. First firms can perform coding Coding
occurs when an organization classifies inputs
before inserting them into the technical core.
Such preprocessing facilitates proper routing

and, if necessary, proper exclusion. For example, many schools track and stream students. This classifies inputs (students) into
homogeneous ability groups so as to buffer
instruction from uncertainty (when you have
students of wide ability, your technology or
curriculum in this case is variably received
and has uncertain effects.
2. Second, firms can buffer their core tasks by
stockpiling. Organizations can collect and
hold raw materials or products, thereby controlling the rate at which inputs are inserted
into the technical core or outputs are released
to the environment. It is easy to imagine this
for raw materials like wood needed for furniture manufacturing. But another example of
this can be found in universities. A good portion of a universitys budget is dependent on
grants, but granting agencies can change the
amount of funding they make available, and
some years faculty fail to secure those funds
(so you have cycles of feast and famine!).
This is partly why universities are increasingly concerned with securing endowments
and gift funds. Universities with large endowments can stockpile funds and dip into
them during difficult times and maintain the
same number of students in their programs.
3. A third strategy entails leveling, or smoothing. Leveling is an attempt by the organization to reduce fluctuations in input or output.
Whereas stockpiling is a passive response,
leveling entails a more active attempt to
reach out into the environment so as to motivate suppliers of inputs or to stimulate demand for its outputs. Here, an example
might be a district (or even a university
again) advertising its strengths so that enrollments and housing values stay high. By creating demand, they sustain inputs in a recession.
4. A fourth maneuver entails forecasting If environmental fluctuations cannot be handled
by stockpiling or by leveling, organizations

111

may have to anticipate changes and attempt

to adapt to them. For example, a university
may foresee that their school will lack funds
and look to identify sources for private funding say, if a Republican candidate is expected to become president and slash the
budget of the National Science Foundation
or National Institutes of Health, then universities might develop relations with private
foundations and industry partnerships as a
means to buffer research and student training
from these resource constraints.
5. The final buffering strategy involves adjusting scale. Here, the firm changes the scale
of its technical core in response to information provided by forecasting, or for other reasons. A good example of this occurs when
firms downsize programs, or when school
districts get rid of performing arts and foreign languages but retain a focus on math
and science. It is a drastic move, but it does
not involve changing the nature of the technical core, rather just its size.
In addition to buffering the technological core
from the environment, an organization can protect
itself via bridging strategies. The goal of bridging strategies is to shape dependence relations in
the environment. One can do this by negotiating
with other firms, by selectively exchanging certain
resources with them, by pooling resources across
them (or partially absorbing other firms), or by performing mergers and totally absorbing other firms.
These are all increasingly greater efforts at bridging. Lets look at examples of each in turn.
The most minor bridging efforts, or rather prebridging efforts, arise in negotiation. The least
costly means is to negotiate with other firms and
evoke normative coordination here behavior is
regulated by common informal expectations that
reduce uncertainty. In Pfeffer and Salanciks work
(1978:147-151) they give a nice story of a teacher
union relating their demands to a school board: the
union gives a list of 6 demands, the first five of
which concern the quality of education (smaller
class sizes, more preparation time, etc.), and the

last concerns their salary. The school board approves 1-5, but not 6, since they regard the last
one to be a private demand cloaked in socially legitimate trappings.
The norm evoked here is one of informal expectations about trust and honesty. The managements job is to note where normative constraints
affect dependence relations, noting whether they
are beneficial, and if not, to seek ways to change
them via persuasion. And thats what happened
when the school board rejected the teachers 6th demand.
Unfortunately, normative coordination
does not always work, and free-riding and opportunism can burn an organization (e.g., one assumes that teachers will not strike during school
year, but that does not always happen). Hence, additional bridging efforts are typically sought.
A second pre-bridging tactic is to bargain.
Here the manager uses a family of tactics to ward
off impending dependence relations. The firms negotiate and exchange in an attempt to prevent the
resource relation from becoming imbalanced. We
saw this type of bargaining occur in the week on
coalitions, so we can gloss over it here.
More serious forms of bridging involve exchange, or the mutual giving up of autonomy for
an exchange of resources. Firms can do this
through contracting. There, the firm attempts to
reduce uncertainty by coordinating their future behavior in limited, specific ways. They define the
rules of inter-organizational contact and exchange.
Negotiated contracts are an excellent way to acquire greater certainty in environmental relations.
For example, it does not hurt to have routine negotiation with teacher unions about their contracts so
as to avoid strikes.
Another, form of exchange can arise via the
creation of interlocking directorates. Here, members of competing organizations are given a position within the central organization that oversees
them (e.g., a board of directors). By being on each
others boards, the firms trade away their sovereignty in exchange for some mutual support. By
giving external members a role, the organization
accomplishes the partial co-optation of an external
organizations interests as their own, but also gives
up some of its control. A focal organization may

112

then become more effective in an environment because they have coopted external members that
might have control over resources central to its
functioning.
For example, I had a student write up a case on
Stanfords Committee of Undergraduate Education
when it was formed several years back (Pope
2006). The committee was trying to reform the undergraduate curriculum and it encountered a good
deal of resistance from the environment and stakeholders. In response, the committee was organized via a Noahs Ark model where they secured
representatives from all the environmental stakeholder organizations: 1 undergraduate from the student council, 1 graduate students from the student
council, etc. This opened up representation but
also co-opted their dissent in the process. The administration gave up some control for greater effectiveness in the environment of vocal stakeholders.
Another form of resource exchange can arise
in hierarchical contracts. These are contracts developed to manage dependencies via conditional
clauses evoking hierarchical mechanisms to handle disputes. It is a contract that preserves and defines the rights of parties in case some problem of
contingency arises. For example, it can be a
clause for subcontracts ensuring full pay if the subcontractor does not come through. These are more
complete and detailed exchanges or developed contracts.
A more extensive means of bridging with
other firms can entail the pooling of resources
across them. One means of accomplishing this is
to engage in a joint venture. Here, two or more organizations create a new organization in order to
pursue a common purpose. For example, two private schools pool their resources to create a daycare center to serve teachers and their children,
thereby reducing the uncertainty of teacher attrition / retention.
Firms can also enter strategic alliances as a
means of pooling resources. These are agreements
between two or more organizations to pursue joint
objectives through the coordination of activities or
sharing of resources. For example, Berkeley and
Stanford have a courtesy program where students
can take courses at one anothers university.

Last, firms can join associations and cartels.

Cartels entail more pooling and loss of autonomy,
but they are also rare. Cartels like OPEC go above
and beyond informal norms, and have actual
means of sanctioning members for not following
their decrees they effectively act as a block of organizations. Notably, cartels are illegal in the
United States.

Producer

Supplier

Supplier

Figure - Horizontal and Vertical Mergers

(Source - http://commons.wikimedia.org/wiki/File:Horizontal_%28PSF%29.png/)

Now firms also can perform a complete pooling of resources or total absorption through
mergers. This can arise in several forms. First,
the firms can perform a vertical merger. Here, the
firm extends control over exchanges vital to its operation. Hence, a high school would merge with a
middle school, or a manufacturer / producer would
buy out a supplier to get control and to create certainty of supply. Firms can also perform horizontal mergers. They can accomplish this by taking
over their competition, thereby reducing uncertainty and increasing organizational power in their
exchange relations. An example of this would be
when one high school takes over another one so
that it can benefit from economies of scale and
pool resources. This actually arises sometimes in
rural areas of the United States where a township
high school is created as a merger of several
smaller rural high schools. Last, one can engage
in diversification. This is a Method for decreasing

113

dependence by acquiring entirely different types of

businesses. For example, I once observed a high
school that took over a nonprofit dedicated to art.
It became a museum and a school! I also saw a
school that expanded into a private day care as
well. Therefore, a firm can merge all its resources
in several ways to bring itself greater autonomy
and control over resources in the environment.
That is a lot of managerial strategies! They
move from simple negotiation, to exchanges, to
pooling and partial absorption, to complete mergers. I think it helps at this point to take a step back
and ask what are the general managerial strategies one can take away from the resource dependence approach? I think there are two basic prescriptions. The first general strategy is to avoid resource dependence on other firms. This can be
done by using buffering strategies like stockpiling,
and engaging in long-term contracts that buffer
your output. You can also try and change the legal
rules and set regulations so as to manage competitive markets. It also makes sense to diversify, and
to find substitutable exchanges (backups). A second strategy is to break your firms dependence on
other firms (and to possibly create their dependence on you!). Here you can use secrecy, restrict
information, begin an anti-trust suit, co-opt the controlling firm, acquire control over the input of the
controlling organization (via something like a vertical merger), and set up rules of regulation.
Forms of Dependence and Theory Limitations
So you have two general approaches, and a
variety of particular managerial strategies you can
use to work your firms resource relations in the
environment. Can we predict certain forms of dependence will arise if some of these strategies are
used over others? Scholars like Richard Scott
think so (2003:118-119). He predicts that certain
managerial strategies will result in certain resource
dependence relations. For example, some firms
tend to assume a symbiotic interdependence this
occurs when two or more kinds of organizations
exchange different resources. This can give rise to
power differences if the resources exchanged are
not of equal importance and value (A!"B). An

example of this might be subcontracting where

money is exchanged for expertise. Such symbiotic
dependence (from moderate to extreme see Scott
page 212) corresponds with normative coordination, contracts and their clauses (hierarchical contracts); as well as joint ventures and vertical mergers.
Another form of dependence is commensalistic or competitive. This occurs when two or more
organizations compete for the resources of a third
party (A"C!B). This is often resolved by differentiation (one specializes and becomes a supplier,
so there is division of labor and interdependence).
An example of this might arise when multiple consulting firms compete for the same contract. According to Scott, competitive dependence arises
under normative coordination, co-optation and the
forming of interlocking boards of directors, trade
associations, joint ventures, and horizontal mergers (where competitors merge).
Just like all the other theories we review in
the course, resource dependence theory is not a perfect theory and it has certain shortcomings. Resource dependence theory assumes all organizations are more or less similar. They acquire resources in an uncertain world and are staffed by
boundedly rational managers who seek to optimize
both their own and the organizations interests.
But is that accurate? Do some organizations live
outside the issues of resource dependence? (e.g.,
Rich ones?)
What does certainty and uncertainty mean
for resource dependence theory? Is dependence
on social resources and knowledge less clear than
dependence on money and materials? Resource
dependence theory is purely resource and
exchange-based, and it assumes there is clarity of
value and importance. Unfortunately, the value of
a resource is often unclear until well after the fact.
Moreover, meaning-making / sense-making are
lost on resource dependence theory. All tasks are
related to efficiency and effectiveness. But what
happened to culture and mission? Normative coordination is not thoroughly described in resource
dependence theory and we find stronger characterizations in theories of organizational culture.

114

Last, all of these dependencies are described

in pair-wise fashion. What about the larger network? Can the larger network pattern define opportunities and constraints? Can the network define norms and pressures better than relations of
dependence?
Case: The Near Merger of Northwestern and
the University of Chicago
In what follows, I will describe a case where
two universities tried to perform a merger and
failed. In the process of relating the case, we will
review the main features of resource dependence
theory and see how they apply to the case. In this
manner, I hope you will get a better sense for how
to apply the theory and recognize its strengths and
limitations. The case we will relate was written by
a historian, Sarah Barnes. She gives a nice account of the failed 1933 merger between the University of Chicago and Northwestern University
or what might have been the worlds first superuniversity. The merger effort arose during the
great depression when both universities were undergoing financial difficulties. In many ways, the
two universities were competitors in the city of
Chicago for students, recognition, funding, and so
on. Northwestern was situated on the north side of
Chicago in a bucolic area overviewing Lake Michigan. It was a large undergraduate institution that
mostly recruited locally and it placed a strong emphasis on applied programs like Journalism and
Medicine. Like I said, Northwestern is in a bucolic setting. Very pretty, and the students seem to
have lots of fun, even today.
In contrast, on the south side of Chicago in
an urban neighborhood was the University of Chicago, a large graduate institution with an elite, national reputation and a strong emphasis on the pursuit of truth and theory. Now the University of
Chicago is where I studied, and my recollection of
the place (even in the 1990s) was that it still was a
leading graduate school, and a very serious intellectual place; a wonderful place to learn, but often
very somber. Back in 1933, the contrast between
these institutions was quite stark. Northwestern

had tax exempt status, Chicago did not; Northwestern had a safe neighborhood, Chicago did not; Chicago had prestige and reputation, Northwestern
did not; Chicago was innovative, Northwestern
was not. Chicago was international, Northwestern
was local. Chicago was theoretical, Northwestern
was applied. Together, the two could benefit from
each others strengths and lose their weaknesses.
In addition, the merger has some financial
benefits. Merging would save each university
$1.7M in annual upkeep and better economies of
scale. And both institutions seemed to recognize
this, and the negotiation moved swimmingly along
until they broadened representation on Northwesterns review panel so as to begin vetting the
merger with larger constituents. Alumni and resistant groups were now included, and they saw
things as too rushed. When the two schools got
down to details, the merger began to unravel: Chicago wanted to keep its college and undergraduate
program, so that would still compete with Northwestern Universitys program; and neither wanted
to lose their medical school or education school.
The merger also fell apart because a key proponent on the Northwestern University board of
trustees died. Also, when the Northwestern University review panel expanded, damaging news
and gossip leaked into the press upsetting alumni.
The gossip was to the effect that the merger was
already decided, that it was a takeover, that it
was a last ditch effort by Hutchins to save his
failed presidency; that NW would lose its
identity, etc. The discourse was partisan and both
social support and public focus on the mutual gain
fell asunder.
So what can resource dependence theory tell
us about this case? Lets briefly review resource
dependence theory again. When does resource dependence theory apply? It has relevance to a case
when there are focal organizations interested in decreasing competition, increasing autonomy, increasing power, and (possibly) increasing efficiency. The main mode of organizing action is to
scan the environment for resource opportunities
and threats, attempt to strike favorable bargains so
as to minimize dependence and maximize auton-

115

Figure. Images of Northwestern and University of Chicago

(Source - Clockwise starting at Northwestern University); http://commons.wikimedia.org/wiki/File:Wikipedia_Education_Program_Northwestern_University_logo.svg;
http://commons.wikimedia.org/wiki/File:University_of_Chicago_biting_chimera.jpg; http://commons.wikimedia.org/wiki/File:University_of_Chicago_antique_postcard.jpg;
http://commons.wikimedia.org/wiki/File:Hutchinson_Hall%2C_University_of_Chicago.jpg/; http://commons.wikimedia.org/wiki/File:Northwestern_University%2C_ivy.JPG)

Case: Merger of
Northwestern
University and the
University of
Chicago.

omy / certainty. So far, this seems to capture Northwestern University

and the University of Chicago some.
Resource dependence theory also characterizes organizational elements in certain ways. The technology (or what brings about changes in
dependence), is focused on external adaptations in order to increase autonomy and/or decrease dependence (see management). The participants are
focal organizations and the organizations that have resource interdependence with. The goal of a firm is survival through external adaptation, and
this is accomplished by establishing certain relations that place the focal
organization in control or with greater autonomy. The social structure focuses on inter-organizational relations, and the effort is to manage standard operating procedures and to perform bargaining / politics on these
relations. And finally, the environment is central. The focus is on exchange partners and external relations more than internal dynamics. All
of these characterizations of organizational elements also seem applicable
to our case.
Last, a manager guided by resource dependence theory would perform some buffering strategies, like the stockpiling resources, leveling
with advertisements, forecasting where their needs will be, and adjusting

116

the scale of their core technology. They will also

bridge with other firms to bring security to the organization within a competitive environment. The
bridging strategies they will use can entail negotiation and long-term contracts, partial absorption
and sharing of resources, such as in forming joint
ventures and alliances, or total absorptions via
company mergers. Here too we see aspects of
merger, co-optation, adjustment of scale in the
case.
So lets apply resource dependence theory to
the case. The Chicago-Northwestern merger is
clearly a case where the environment is uncertain
and problematic for both universities. It is also a
case where the two organizations explore a horizontal merger (or in secrecy, an effort to co-opt)
between competitors. In theory, the merger would
entail some buffering, where each university downsizes their core technology losing their worst programs but keeping their best, But then they combine in a complementary form that would have an
improved economy of scale. The merger would
have made for an unparalleled super-university".
One can only image what it could have been by
combining the strengths of the two campuses!
So why didnt it happen? Does resource dependence theory give us an explanation? Lets apply our concepts and elaborate the theorys application to each school and see. For Northwestern, the
proposed technology of organizational change was
the merger and how it might increase economies
of scale (making the university more efficient and
autonomous) and minimize some of the problems
related to the great depression. The larger size of
the combined university could also lead to increased diversification, thereby increasing autonomy.
The participants mentioned on the Northwestern side were Walter Scott, the board of trustees,
outside consultants, faculty, alumni, and the press.
The goal of Northwestern was to become an ideal
university, where they increased their positioning
and quality of education via improving their applied programs in professional and community
practice. The social structure was such that undergraduate programs were more valued than graduate programs. The network of supporters centered

on president Scott and a key Board of Trustee

member, but then opponents emerged in the
School of Education, the Medical School, and
their College of Liberal Arts all of whom wanted
to get a fair deal from the merger. Other stakeholders enter later due to press leaks. These are
alumni, and they are mostly opposed. The deeper
belief structure at Northwestern sought to promote
service, social pragmatism, and utilitarianism.
These were in stark contrast with the University of
Chicago, as you will see next. The environment in
which Northwestern found itself was one of a major economic downturn hurting all aspects of the
university.
Lets next look at the University of Chicago.
Chicago, like Northwestern, saw the merger as a
means of increasing economies of scale, making
the university more efficient and autonomous, and
minimizing some of the problems related to the
great depression. The larger size of the combined
university would also lead to increased diversification, thereby increasing the universitys autonomy
in the environment. The participants at Chicago
were Robert Hutchins, Chicagos board of trustees, an outside consultant, faculty, alumni, and the
press. Chicagos goal, like Northwesterns, was to
develop an ideal university. But here the goal was
to increase the sustainability of the university by
increasing the quality of education via the pursuit
of theory and truth. Moreover, Chicago had an
elite focus instead of an applied one. Chicagos
mission was also more developed and frequently
voiced by the charismatic president Hutchins.
The social structure and values of Chicago
were quite different from Northwestern. Hutchins
was a strong leader, so it was a more centralized
decision structure. In addition, Chicago valued
graduate training far more than undergraduate
training. Last, its professed beliefs rested in the
pursuit of theory and truth, and an idealization of
the Great Books. Northwestern, on the other
hand, viewed Chicago as an Easterner funded
(Rockefeller) school led by elitist, idealist
(Hutchins). The fact that its goals clearly aligned
with its practices and beliefs, served to form a relatively integrated university culture at Chicago.
The environment for Chicago was much like that

117

118

Goal is organizational survival through external

adaptation (for certainty and autonomy).

Formal roles, standard operating procedures, interorganizational bargaining / politics.

Goals

Social Structure

Buffering: protecting technical core from

environmental threats (coding, stockpiling, leveling,
forecasting and adjusting scale).

Management Strategies

Bridging: security of entire organization with relation

to the environment. Total absorption via merger
(vertical, horizontal, and diversification), partial
absorption (cooptation [vertical or horizontal],
interlocks, joint ventures, strategic alliances,
associations)

Key component of the perspective. Exchange

partners and external relations more salient than
internal dynamics; Bridging more relevant than
buffering.

Environment

(note: coalition approach emphasizes individuals and

interests. Here, the organization is the main actor
and exchanges are with other organizations.)

Focal organization and other organizations with

resource interdependence.

External adaptations in order to increase autonomy

and/or decrease dependence (see management).

Resource Dependence Theory (RDT)

Participants

changes in dependence)

(what brings about

Technology

Key Elements

When does it apply?

Horizontal merger

Influence of the media/press,

Major economic downturn hurting all aspects of the

university. Supposedly, economic pressures were
greater for Northwestern than for Chicago.

Cognitive / Normative: Service, Dewey pragmatism,

utilitarianism.

Horizontal merger

Influence of the media/press

Major economic downturn hurting all aspects of the

university.

Cognitive / Normative: Theory/Truth, Idealism and

Great Books

Trustees, Deans, Alumni, Students.

Coalitions/Opponents: School of Education, Medical

School, and College of Liberal Arts.

Undergraduate and graduate programs.

Evanston Campus, Downtown Campus, Trustees,

Deans, Alumni, Students.

Increase the sustainability of the university, educate

and train citizens

Robert Hutchins

Merger would increase economies of scale (making

the university more efficient and autonomous) and
minimize some of the problems related to economic
cycles. The larger size also leads to increased
diversification thereby increasing autonomy

University of Chicago

Undergraduate and graduate programs.

Increase the sustainability of the university, educate

and train citizens.

Walter Scott

Merger would increase economies of scale (making

the university more efficient and autonomous) and
minimize some of the problems related to economic
cycles. The larger size also leads to increased
diversification thereby increasing autonomy.

Northwestern University

Summary/Basic Argument -- The resource dependence perspective suggests that organizations seek to avoid dependence and uncertainty.

Resource Dependence, Chicago-Northwestern Merger (1933):

for Northwestern. Chicago was experiencing major economic woes that hurt all aspects of the university. However, these economic pressures were
supposedly greater for Northwestern than for Chicago.
In sum, simply identifying the organizational
elements and how they are characterized in the
case reveals how the two organizations differed.
They had very different social structures and
goals, and Chicago was slightly less financially dependent on the environment than NW.
Lets consider the different managerial concerns these two schools had. For Northwestern
there was much to be gained from the horizontal
merger, but also some to lose. There were certain
resources it wanted to retain. It wanted to retain
the tax break it got (a buffer) as part of its charter.
It was willing to lose its graduate programs if it
could retain its applied professional schools and
undergraduate program. In exchange it would get
an elite graduate program and international prestige. In addition, it would co-opt its regional competition for students, faculty, funding, etc.
Chicago on the other hand wanted the benefit of Northwesterns tax break. But it was not
very willing to lose its professional programs and
undergraduate college.
It wanted to keep its
school of education, medical school and college
Chicago was working for an edge! In some regards, Chicago saw the merger as an opportunity
to move its less desirable programs off-site (applied programs). Last, it saw this as a chance to
co-opt its competition and form a world-leading
super-university.
Resource dependence theory would approach
this case with a focus on the different levels of dependence. And it would cite those levels as a reason for Chicagos more aggressive approach and
the merger failure. It would note that Chicago
tried to change the rules of the merger toward a
more asymmetric contract and that Northwestern
saw this as a violation of normative coordination.
Other theories seem to help with the details of this
case. The internal workings of each schools deciding bodies are better characterized by coalition theory. There we can see how the build-up to a contract and merger required a good deal of political

wrangling. Also, the death of a key player at

Northwestern seems central, at least as presented
by Barnes. Moreover, coalition theory can help
make sense of all the camps for and against the
merger at either school. So coalition theory may
help explain how the internal mobilization efforts
fell apart while resource dependence theory helps
explain why the two universities approached the
merger differently and incompatibly for a merger.
However, mergers are often somewhat asymmetric, so the issue becomes how asymmetry got
in the way. My sense is that neither resource dependence theory nor coalition theory are welltuned to the deeper cultural differences in the
two universities that likely played a huge role.
The distinctive, highly valued cultures of each university made it imperative for the merger to proceed in an equal form or long term contract even
though Chicago may have had the greater resource
advantage. Moreover, Chicago had a more pronounced and integrated intellectual culture at that
time, and it may have made Hutchins and the Chicago camp over-value their notion of a university
and make them approach the merger more as a
takeover than a pooled effort. In this manner, the
two sides never came to an agreement.

119

120

Acknowledges routines, but focuses on practices within

them that enable their continual adaptation and change
to fit reality i.e., practices reflecting organizational
intelligence.

Summary or
Basic
Argument

Action = result of local actors collaborative search (trial

& error / transfer) and adapting rule to situation.

Find ways to create lateral ties among workers so

knowledge is passed / transferred more readily /
quickly (if possible, quickly), create means to
organizational memory of what works. Create applied,
social learning experiences with means to retaining and
transferring expertise. Want communication, collective
improvisation, practice and knowledge sharing to arise.

Dominant
Pattern of
Inference

Management
Strategies

Informal, lateral relations, communication, negotiation,

& collective improv. Actor identities (demand)
important. Network of practice (professional identity /
reach) & community of practice (cohesive group).

Find ways to confer ideology and lead others to identify with

it (using a variety of practices and artifacts), but dont make
it so explicit / fanatical that cynicism emerges. Give room
for autonomy and self-expression so distancing is
unnecessary, and encourage members to generate a culture
of their own (~org learning culture NE to Tech culture which
is top-down engineered).

Action = result of deep structure or culture that is generated

in the organization, but which is mediated by the members
relation to it.

Many elements of culture have origins from outside, and

they are transported in, then translated to the local culture.

Application problems pattern recognition not there (no

fit).

Goals
(what probs to
resolve)
Social
Structure

Source of inter-organizational knowledge / tricks /

transfers.

Deep structure composes the elements of culture themes

(beliefs & norms), their expression via practices (rituals,
etc), and their manifestation or expression in artifacts
(reports, mission statements, etc).

Members of organization doing work / SOPs

Participants

Environment

Create intrinsic motivation (sense of fulfillment), and

remove differentiation / cynicism in most cases.

Internal adaptation, or where actors alter routines for the

better and fit reality (knowledge).

Matching, sense-making / meaning-making, or where actors

seek to express beliefs, norms, and values via a variety of
practices and externalize them in artifacts depicting shared
understandings / notions of appropriateness.
Actors within the organization, and those salient to meaningmaking.

Actors seek expression and fulfillment of identity, and

organizational culture is the medium for such
expression/sense-making.

When the cognitive and normative aspects of social structure

are of concern and seem to guide organizational decisions
(sense-making) and outcomes.

Organizational Culture

Technology
(how solutions
get decided)

Key Organizational Elements

Exists when there are clear feedback loops, adaptations,

memory, and support of actor-expertise / adaptations of
rules to local reality.

When does it
apply?

Organizational Learning (OL)

Summar y Table of Resource Dependence Theory (RDT)

Buffering: protecting technical core from environmental threats

(coding, stockpiling, leveling, forecasting and adjusting scale).
Bridging: security of entire organization with relation to the
environment. Total absorption via merger (vertical, horizontal, and
diversification), partial absorption (cooptation [vertical or horizontal],
interlocks, joint ventures, strategic alliances, associations)

Action = scan environment for resource opportunities and threats,

attempt to strike favorable bargains so as to minimize dependence and
maximize autonomy / certainty.

Key component of the perspective. Exchange partners and external

relations more salient than internal dynamics;
Bridging more relevant than buffering.

Formal roles, standard operating procedures, inter-organizational

bargaining / politics.
(note: coalition approach emphasizes individuals and interests. Here,
the organization is the main actor and exchanges are with other
organizations.)

Goal is organizational survival through external adaptation (certainty

and autonomy).

Focal organization and other organizations with resource

interdependence,

External adaptations in order to increase autonomy and/or decrease

dependence (see management). Comply / adapt, avoid / manage.

Focal organization with input/output concerns that cannot be resolved

without considering the environment.
For the most part, organizations are considered unitary actors (some of
the struggles/internal divisions are minimized) in order to highlight the
interactions with suppliers and clients.

Exists when there is a focal actor interested in decreasing dependence,

increasing autonomy, increasing power, and (possibly) increasing
efficiency.
Preferences and goals are unclear except in relation to dependence.

Resource Dependence Theory (RDT)

References

Barnes, Sarah V. 1999. A Lost Opportunity in

4:453-473 (Resource dependence view of how

power is established within universities).

American Education? The Proposal to Merge

Scott, Richard. 2003 (5th ed). Resource De-

the University of Chicago and Northwestern

pendence (pp. 118-119) and Managing Task

University. American Journal of Education,

Environments (pp. 197-212) of Organiza-

Vol. 107, No. 4:289-320.

tions: Rational, Natural and Open Systems, 5th

Davis, Gerald F. and Walter W. Powell. 1992.

Edition, Englewood Cliffs, NJ: Prentice-Hall.

Selection from Organization-Environment Re-

Thompson, James D. 2003 (1967). Organiza-

lations (pp. 315-326). In Handbook of Indus-

tions in action: Social science bases of adminis-

trial and Organizational Psychology, Vol 3 (2nd

trative theory. New Brunswick, NJL Transac-

ed.). Eds. Marvin D. Dunnette and Leaetta M

tion Books.

Hough. Palo Alto, CA: Consulting Psychologists.

Pfeffer, Jeffrey and Gerald Salancik, 1974. "Organizational decision making as a political process: The case of a University budget," Administrative Science Quarterly, 19: 135-51.
Pfeffer, Jeffrey and Gerry Salancik. 2003
(1978). The External Control of Organizations:
A Resource Dependence Perspective. Harper
& Row.
Pope, Chris and Daniel McFarland The Commission on Undergraduate Education. Stanford
School of Education Case, 2006-01.
Salancik, Gerald R. and Jeffrey Pfeffer. 1974.
The Bases and Use of Power in Organizational
Decision-Making: The Case of a University.
Administrative Science Quarterly 19,

121

8
Network Forms of
Organization

Social'Network'Perspec1ve'on'Organiza1ons
In this chapter, I will describe how organizations researchers look at social networks within
organizations. In addition, I will describe how
some theorists contend there is a network form of
organization that is distinct from hierarchical organizations and markets. So this week, I will relate two perspectives: a purely analytic one that describes networks within organizations, and a theoretical one concerning a prescribed form of interorganizational association that can result in better
outputs.
Lets start with the social network perspective. The social network perspective embraces the
notion of social embeddedness as related by Mark
Granovetter (1985). He argues that, on the one
hand, economics and market accounts of behavior
are under-socialized: actors behave as if their actions are unfettered by social contexts. On the
other hand, sociologists and institutional accounts
of behavior are over-socialized, and relate social
actions as socially deterministic and devoid of
choice. He posits the notion of embeddedness as a
middle ground, where social action is embedded in
transactional networks. This embedding applies
even to economic transactions in markets. It is
within those structures that we decide and act in
intentional ways.

Figure - Global Trade Network

(Source - .http://commons.wikimedia.org/wiki/File:b/b9/TheProductSpace.png)

This image is a global trade network and the

related products or proximate products. Notably,
it is not random. Some products are far more related to others, suggesting the transactions are socially patterned and embedded in a structure. This
structure is often the focus on the network analyst
they seek to understand context of association in
which actors are embedded, and from that, they
learn how it constrains and enables actor decisions. The social network analyst studies embeddedness by focusing on the social structure
within the firm, and/or the web of relations across
many firms in the environment. These contexts of
association differ in pattern and represent different
contexts for action. Where an organizational actor
is located can also differ, affording the manager
and firm different opportunities and constraints on
their action.

Figure - Network and Structure

(Source - http://commons.wikimedia.org/wiki/File:ProductSpaceMST.png,
http://commons.wikimedia.org/wiki/File:ProductSpaceFDSL.png )

Here we have a network that looks like a

spanning tree and it has a lot of reach. You will
note that the actors do not have many interconnections and that there are key bottlenecks or cutpoints in the network that could undermine the
transfer of resources. If we took certain players
out the network would fall apart. In the second network, we see some dense hubs of interconnected
actors and there we could pull several players out
without concern for the networks persistence. We
also see actors on the periphery who seem to be
well out of the way of core transactions. Each of
these positions represents a different context for
action and information. Where you are located
matters. Opportunities and constraints change for
an actor or firm depending on their location. So-

123

cial network approaches confront questions about

network form and positioning.
Networks of association not only influence
social action, they are the result of decisions.
Hence, social structures and networked environments are dynamic and evolving. They change!
Social network scholars have only just begun to
develop tools to help us understand how to engineer and develop different social structures within
and between firms.
Here we see a schema of a set of ties that
grow more interconnected and eventually form
quasi independent groups of association.
They change!
! Stage 1

! Stage 2

! Stage 3

Figure - Growth of Ties

(Source http://commons.wikimedia.org/wiki/File:Network_self-organization_stages.png)

In this chapter, I will present to you some basic concepts used by social network scholars studying organizations (for review, see Kilduff and Tsai;
Borgatti). In so doing, I hope you come to see it
as a distinct empirical approach that you can possibly apply in the organizations you participate
within.
A network analyst typically asks and answers
a series of questions when they study a firm. First
they ask about the unit of analysis and the boundary of a network. They try to define what the network is and where it begins and ends. Many network analysts will focus on the individuals within
a firm and how they associate. However, it is not
always possible to study everyone in a firm. For
example, when studying schools, network analysts
may only study the teachers and students within a

classroom or school. As such, they recognize natural boundaries to the core work environment, but
they will forgo studying support staff, counselors
and parents. The same occurs when the analyst
studies a firm. They may focus only on managers
and employees who are in the building or a particular division, and they will ignore the clients, support staff, etc. In these cases of exclusion, the analyst is defining a nominal boundary to the organization, and it is important they consider whether that
boundary is sensible given the phenomena they
want to study. For example, a focus on teachers at
the exclusion of students may be an acceptable
boundary to define, if the concern is with how
managers exchange information on instruction and
pedagogy. If you think managers learn most of
their pedagogy from students telling them what
other teachers do, then it could be a problem for
your study and the conclusions you draw.
Analysts also look at larger units of analysis
like sets of firms and their relations. Here they
study a field of organizations and the transactions
they mutually recognize as most relevant to their
firms functioning. For example, in Powells 2005
paper in the American Journal of Sociology, he
looked at the field of biotechnology and included
in his sample universities, biotech companies,
banks, venture capital firms, and government granting agencies that were all co-partnering on patents,
sharing commerce relations, transferring experts,
and so on. As such, his work tries to capture the
boundary around an entire organizational field as
seen by those participants. That is often hard to
do, so many analysts will focus on one type of
firm and their core transactions and ignore others
for example, there have been ample studies of venture capital firms and their networks of copartnering on inventions.
Next the analysts ask (often implicitly)
what is the unit of time? When does the network
begin and end? At issue here is the temporal
boundary to the social structure in question. Most
confront the issue of temporal boundaries by considering which relationships are most important to
the organization. Are they transactions on a discussion forum; are they exchanges done each quarter
of the year; are they yearly contracts that have rela-

124

Figure. Aggregations of classroom interactions by unit of time (Source: Bender-de Moll and McFarland)

tive permanence? And then they have to ask during which time period will they study the firm?
Will they look at exchanges occurring only in the
first week of a new technology being rolled out,
the span of a project (e.g., this whole course), or
over multiple years?
The unit of time issue matters since it has implications for how transactions aggregate into different patterns. Take the simple case of a classroom. If we look at interactions by the minute,
well see sequences of dyadic or pair-wise exchanges. If we look at interactions aggregated
over 5 minutes, we see the network structure
change from one activity (of say lecture) to another (of say groupwork). And if we aggregate all
transactions across 35 minutes, we start to see the
general interactions of the classroom as a central
tendency of association. Hence, your unit of time
presumes different notions of social structure
from the structure of momentary interactions, to

activities and practices, to social norms. As an analyst, you need to ask yourself if you are trying to
understand the structure of micro-routines, larger
tasks, or group norms, and from that you can decide the unit of time to adopt. Once a social network analyst has a sense of where and when important relations occur and when and where the network begins and ends, they can start to study the
quality of relations and analyze the network.
There are many kinds of transactions that occur within and between firms. A good analyst will
go to the heart of the matter. They will listen to
the clients concerns and seek out the most relevant types of transaction. These relations are typically observed behaviors or they are ones the subjects perceive. Observed relations can be identified in company records, observations and even
reports. Perceived relations are typically reported
by the subjects and require surveys or interviews.
Clearly, the focus on perceived relations implies

125

that worker perceptions of their relations matter

for organizational behavior. The focus on observed behavior assumes that one type of association may presage another and influence outcomes.
The quality of a tie need not only be defined
by the content of interactions or the type of association. It can also be defined by its form, or
strength. Strong ties are generally considered to
be influential ties. When your close friends pressure you, it has greater influence on you than
when a total stranger pressures you. Strong ties
entail frequent interaction, positive liking or affect
(typically mutual!), and they have a history to
them. In network research, we characterize strong
ties as having bonding capital or a cohesive pull
on people that guides their action, and we frequently refer to these relations as ones of friendship, mutual support, and liking. In contrast, weak
ties are more infrequent, casual interactions between acquaintances. Whereas strong ties have
bonding capital and bring local persons into
greater contact, weak ties have bridging capital,
and typically bring distant persons into contact
across groups.
In certain regards, these two types of association reflect the ties had in a COP and a NOPs that
Organizational Learning theory relates. Social network research has identified a consistent set of
mechanisms that form interpersonal relations in
firms and other settings. Most of these findings
concern strong ties but they afford analysts and
managers some sense for how tie creation could be
facilitated, so I want to relate them here.
People naturally form close ties when they
are in proximity to one another (contact opportunity is key!) or when they are similar. In the latter,
case, we associate by homophily, or as the saying
goes birds of a feather flock together. We also
associate by means of exchange and rational
choices concerning them. We engage in reciprocity too. Many years ago anthropologists found
that gift giving induces a sense of obligated reciprocation across societies. Therefore, when you
meet new people, make a kind gesture and it is reciprocated more often than not! Other theories
like status-attainment suggest persons seek out interpersonal advantages and dominance (so weak

ties tend to be rank-ordered). And then we have a

theory of social balance, where persons seek consistency in their relationships more generally: for
example, most of us feel at unease when our
friends befriend our enemy. We have to either
sever our tie, or they have to make my enemy their
enemy if were to stay friends (for review see Dahlander and McFarland 2013).
Last, an under-emphasized mechanism of tie
formation concerns practices and identities. In
most organizational contexts we cue routines, and
they oblige us to assume different interaction roles
(listener, hearer, inquisitor, speaker, coauthor, etc).
The same goes for cued identities we frequently
cue notions of a storied tie and what that entails.
For example, when I make kind gestures of friendship, we assume that entails reciprocation, frequent contact, supportive interactions, and so on.
When I evoke the identity of professor, that cues
other expected role-relations and their interactions.
Hence, all of these micro-mechansisms have been
studied and related to tie formation, and they suggest pathways to tie creation within organizations.
So we have multiple types of networks and a
variety of mechanisms have been found to generate them. What about our positioning within these
networks? Social Network scholars love to study
the positioning of actors in networks because it is
not just the overall form of the network that defines action potentials, but it is also your location
within them that does. After all, it is a very different situation to be a central actor in a centralized
network than a central actor in a fractured one.
When network analysts speak of positions
they often refer to the notion of centrality. Centrality can refer to positions of prominence or to positions of mediation and brokerage (there are other
notions of central positioning, but we will limit it
to these for this course [see Kilduff and Tsai for
review]). In the second network image of this
chapter (Network and Structure), you can see
the prominent persons with many ties and the peripheral ones with few ties. But then there are individuals who are on the path of others and if we
took them out the network would fall apart. These
are key mediators between different parts of the
organization. Identifying these central players can

126

Figure - University Faculty Network (Source - McFarland unpublished research)

be useful to a firm and especially when it comes to

information flows and bottlenecks.
Another type of network position is more
group-based and concerns the location of hubs of
interconnectivity. When looking at networks, analysts frequently try to identify these sets of actors
and clusters of ties since they reflect locations in
the network where conformity and social influence
likely occurs. They are also spaces of redundant
information.
Clusters and groups of interconnected actors are important to the study of firm social structures, because they often influence firm
outputs and behavior. I will show you more of this
in the next lecture.
Thus far, we have described organizations as
webs of transaction and actors as occupying different positions or groups. This can be complicated
further if we consider that social realities are such
that organizational actors are actually embedded in
multiple networks. This is where it both gets more
interesting and more complex. Take the example

of trade associations. In most fields of business,

trade associations assume a position central to information exchange but they are peripheral to the
networks exchanging financial resources.
As
such, they can persuade other organizations via the
use of information as a resource, but they cannot
enforce other firms to conform because they lack
hard financial resources.
Within firms, it is a similar situation. Lets
look at faculty and their relations in a west coast
research university from 1993-2000 (see figure
and key above). This school has around 1500 faculty ranging in status. Some are clinical faculty
engaged in the hospital and patient care; some are
assistant professors; and others are tenured faculty.
Also, these faculty come from different divisions:
some come from the humanities, others social sciences, law and education, natural sciences, earth
sciences, the physical sciences and engineering,
and medicine.

127

Figure - Coauthoring among Faculty

Figure - Co-Advising Students among Faculty

(Source - McFarland unpublished research)

(Source - McFarland unpublished research)

Now lets impose the network of collaborations on these actors. In particular, lets look at
where coauthoring of publications happen. They
happen mostly in medicine, the hard sciences and
engineering.
Notably they collaborate more
within those fields than between them, but there is
a good deal of crossover. By contrast, social scientists and humanists mostly publish alone.

Now lets look at student training and how

faculty work together to provide that. Here we see
faculty who are linked when they co-advise doctoral students. Notice how the network disappears
over medicine and forms over the social sciences
and humanities. And notice how the collaborating
medical faculty are tenured (pink triangles).
So what do we learn from this? We learn
that each field is embedded in multiple types of activities and networks of collaboration, but they assume different positions across them, and this
likely results in greater cumulative advantages to
each field. The medical faculty collaborate on publications and grants; the social science and humanities faculty collaborate over student training they
also publish and write occasional grants, but its
not a point of collaboration. Last, engineers and
scientists collaborate in everything they do.
Moreover, they frequently interact with medicine.
Hence, we see a differentiated collaboration
structure. The implication is that the social sciences and humanities mostly transfer knowledge
via mutual efforts at training students, while medicine does it mostly over research (and probably
postdocs!). And yet other fields are high-speed
thoroughfares for knowledge transfer via collaboration in everything so engineering is very much a
linchpin in this university bringing together high
powered domains of research and training.

Figure - Co-granting among Faculty

(Source - McFarland unpublished research)

Now lets look at grant collaborations. Notice the network shifts and we get a second hub of
dense collaboration within engineering.
But
again, there is clear clustering and some inter-field
collaboration across engineering and medicine.

128

Social Networks and Organizations

This part of the chapter introduces you further to the social network perspective on organizations, but it elaborates how networks influence organizational behavior and outcomes, and it describes ways organizations can create different network patterns and positioning. When we study the
effects of networks on organizational behavior, we
view social relations and actor positioning as an
independent variable shaping outcomes we consider whether people influence one another and diffuse their motivations through their friendships, or
if being in key locations of the network have certain returns or advantages for the worker and the
firm. When we consider network formation, we
look at the network as an outcome or as a dependent variable. Here we want to know the factors
that lead persons to form relations and factors that
lead a network to assume a particular shape (and
perhaps even a pattern you as a manager desire!).
So lets look at these using examples in the
slides that follow. First, lets consider how relations influence behavior or what we call peer influence. The general argument is that the people
we associate with influence us, and they lead us to
act in ways we wouldnt normally act if we were
on our own. In organizations research, these studies often focus on processes of social diffusion and
the adoption of organizational innovations. Some
research studies whether collaborating with productive colleagues increases your productivity (does a
particular mentoring program have solid returns?).
In most of these studies, researchers find
close ties are a great means to diffusing attitudes
and behaviors. At the inter-organizational level,
scholars find that the adoption of organizational
innovations often flows through associations like
interlocking boards of directors. For example, a
string of papers found that the use of poison pills
in corporate takeovers was an organizational innovation spread via interlocking boards of directorates. The poison pill for those of you wondering
was a strategy firms would use to prevent takeovers in the 1980s and 1990s it was a way of
making the firm seem like an expensive, lowprofit gamble, so not worth taking over. Craig

Rawlings and I have studied how faculty productivity diffuses through collaboration networks (Rawlings and McFarland 2011). We found that a university improves its grant record by getting successful
grant-seekers to collaborate with novice grantseekers. Such collaborations improved application
rates, success rates, and the amounts awarded.
The diffusion of expertise was even greater when
these collaborations were repeated more than
once, thereby ensuring novices learned how to get
grants on their own and with others in the university. By contrast, persons who did not collaborate
struggled to win awards.
In other work, the conduit of influence is not
a strong tie, but a weak one. Granovetter has written some seminal work on social networks, and in
particular he has made a strong case for the importance and usefulness of weak ties (1974). In his
research on job seeking, he found that most people
learn about a job and acquire it through weak ties
and indirect ties (friends of friends) rather than
their close friends. He argues this occurs because
weak ties often bridge groups and bring more
unique information. Persons relying on strong ties
and cliques mostly found redundant information.
Hence, the person with weak ties was more able to
access knowledge about jobs.
Strong and weak ties are often characterized
as bonding and bridging forms of social capital
or types of association that bring social advantages. Strong ties and bonding capital generate social control and conformity as well as socialization
or diffusion, while weak ties and bridging capital
often extend a persons reach into pools of useful
information. Strong and weak ties imply the creation of certain network configurations and network
positions. Hence, I want to turn to positional effects next.
A common finding within organizations is
that persons occupying certain positions have an
advantage, like access to recognition and information, and this enables the occupant to be successful
in their career.
The same is said of interorganizational networks. Organizations assuming
prominent and brokerage positions tend to survive,
grow and have greater control and influence on the
field of organizations they are embedded in.

129

David Krackhardt offers us a nice illustrative

example of the effects of network positioning on
firm behavior and outcomes (1992). He describes
the case of a technology firm that is subject to a
unionization effort. According to Krackhardt, the
unionization effort fails because union proponents
do not co-opt the informal leaders of the strong tie
networks. Via social network surveys, Krackhardt
shows that the union recruits persons peripheral to
both advice and friendship networks at the firm.
His analyses also demonstrate that the firms leaders were in a precarious position, as they too were
somewhat peripheral to the firm. As such, the union missed an opportunity when they failed to recruit Chris, everybodys favorite guy. And in
many ways the leadership of the firm (Steve) lucks
out at the unions poorly targeted recruitment efforts.
I find Krackhardts case to be simple and elegant. His methods are simple: he uses interviews,
surveys, and observational records to retell the
story of how a unionization effort failed. He first
describes the organizational chart of who reports
to whom and identifies the collective bargaining
unit the union tries to establish. Then he goes on
to show how the key union proponents are neither
central to the advice network of experts nor to the
friendship network of trusted relations. They neither co-opted the experts nor the popular individuals in attempting to create a bargaining unit. Had
they known to check the network and co-opt Chris
and his close friends, then they might have received the social support they needed to successfully unionize the firm.
Krackhardts case focuses on the effects of
positioning. What about cliques or social groups
and their effects on workers? Long ago in 1939,
Rothlisberger and Dickson studied a bank-wiring
room where workers essentially created circuits.
Rothlisberger and Dickson found that the friendship groups of these workers altered the rates of
their work output and normed them so that they
stayed within a particular output level that worked
for the set of friends. Subsequent scholars have
remarked on how peer groups or clusters of strong
tied individuals can be a strong force in organizations influencing their outcomes.

I see this in my own work on American high

schools and their classrooms. In those settings
youth act with their friends in mind. In most classrooms, youth form friendship groups or cliques
and conform their behavior to them. One example
is a high school English composition class I observed as a graduate student. It was composed of
11th and 12th grade students, equally well
equipped to read and comprehend the course material concerning William Shakespeares written
works. The teacher, Sophia, liked to encourage
dialogue and frequently called on students. Nonetheless, the students formed clusters of association
based on gender, race and age, and these groups
were rank-ordered within grades. As such, there
was a popular core group and a hanger-on group,
and this arose for each grade. Interestingly, the
core 11th grade group and the core 12th grade
group did not compete on the same stages. Instead
they specialized in distinct conversational arenas
and topics the seniors dominated the public stage
of academic discussions and the juniors dominated
the backstage of social discussions.
In the following matrix you can see the
friendship relations during the semester in which I
observed them. I used network analytic tools to
identify their groupings and they break down into
4 clusters. The ties can be read as row-column relations, or from-to relations. Hence, the value of 1
from 16 to 15 and an . from 15 to 16 suggests
that 15 thought she was friends with 16, but 16 did
not reciprocate that sentiment.
Notably, most of the groups are homophilous
by grade, gender, and race so they follow the saying, birds of feather flock together. Moreover,
many have reciprocated ties (more than chance).
While it is not shown, it is also the case that many
friends sit by one another. Therefore, propinquity
is also in effect. In the matrix you can see the
smaller secondary peer groups within each grade
you can see they hang on or look up to the core
group in the off diagonal relations spanning the
cliques. These secondary cliques want to be
friends with the larger core clique, but it is not reciprocated. Hence, there is a degree of rank ordered clustering in each class.

130

Figure - Matrix of Friendships in a High School English Class

(source - McFarland unpublished research)

We can render these relations into a network

image where the y axis is the prominence or popularity of individuals. The shaded circles reflect the
general boundaries of each clique. Notably we see
the two grades as somewhat disconnected and
each having a core clique with a hanger on clique.
In other analyses I tested whether observed interaction patterns conform to these cliques over and
above seating and homophily, and they strongly
do. For ease of interpretation, I will superimpose
the observed behaviors and interactions on these
groups. From doing so we learn a few things.
First, we learn that most of the interaction is di-

rected within the cliques. Second, we notice the

cliques specialize their behavior.
Here I use red to denote where task or academically focused interactions emanate. I render
the red bolder where the rates and density of such
interaction are higher. Clearly the senior core
dominates such interaction (and the core clique in
either grade). Next I use blue to denote where social or non-academic interactions emanate (like
play, joking around, etc). I render the blue bolder
where the rates and density of such interaction are
higher. Here the junior core dominates such interaction (and the core clique in either grade). A
slough of statistics can accompany these images

131

Hierarchy and Clique Structure for English Class

(source - McFarland unpublished research

and further the argument. However, the point for

this lecture is more conceptual and schematic the
structure of the informal network and its cliques
strongly guide behaviors. Moreover, the cliques
arise from a variety of tie formation mechanism of
homophily, reciprocity, status-seeking and even an
effort at specialization (so as to avoid competition). So the sum of it is that it is not just single
relations that influence workers and their firms,
but also network positioning and groups!
But what of network formation? In the beginning, I spoke of how analysts often view networks
as an outcome or as having a desired structure that
managers want to achieve. How do we accom-

plish and engineer different structures of association? In my work on high school adolescent networks, we find they vary in macro-structure from
school to school and classroom to classroom.
Some of these settings entail hierarchical worlds
like the first image, and others are heavily segregated and clustered like the second. Nevertheless,
friendship networks are all shaped by the same
sorts of tie formation mechanisms homophily,
propinquity, reciprocity and hierarchicalization.
This raises a conundrum? If the same micromechanisms apply to every friendship network,
then how is it their patterns vary?

132

The potential answers are interesting. It

turns out that certain types of ties usually correspond with a type of network form or pattern. In
the case of our schools, the ties are friendships and
friendships tend to be reciprocated and local so
they accent clustering. And this closure and reciprocity of ties happens to be the strongest feature
driving high school friendship network formation.
By contrast, if the ties were weak, then it is likely
the structure would entail more spanning trees,
rank ordering and fewer groups. Acquaintance ties
have greater imbalance and looseness to them, and
enable different network patterns to arise.
In my work on high schools, we find that the
friendship networks vary from school to school because the organizational context amplifies and
dampens the salience of certain micromechanisms. What this means is that the composition of participants and the utilized organizational
rules moderate natural bases of association. Take
the case of a large heterogeneous population (of
say multiple equally present races), and one where
contact is by choice (meaning the students are not
sorted into ability groups by the school), we find
the pattern of association is highly segregated by
homophily. In large schools with no organizational sorting and lots of choice, there tend to arise
rank ordered cliques. The only time we see random, dense ties of association are when organizations presort the populations and place members
into small interactive settings. We also see more
random association in highly interactive team settings like classrooms where students rotate
through different task groups.
Once analysts have a good description of a
network, its key influence processes on outcomes,
and the key mechanisms that drive it to assume certain patterns, they can begin to prescribe all sorts
of treatments. I hope many of you infer what
those might be from the prior slides of this chapter. Nonetheless, I can tell you what many firms
will want. They will want to facilitate the creation
of efficient network patterns. For example, many
will want interactive, dense networks of positive,
work-related collaborations (as opposed to sociable ones!). And they will want them to span
groups so good ideas can travel around the firm.

Many companies will also want to forge teams

composed of differently skilled persons who will
rely on each others strengths (as an organic whole
greater than the sum of their parts) to make a product or solve a complex problem.
In addition to facilitating the creation of
ideal network forms, companies will also want the
analyst to perform network correctives or to solve
coordination problems. In schools we have many
instances of this occurring when students assume
positions or form groups that drive behaviors in
negative directions. For example, in many classrooms certain kids dominate and take up all the
teachers time and attention, thereby making
achievement games less equal. To offset this, researchers have suggested positional treatments. To
offset unequal access, the tasks are designed to involve decentralized formats like group work so
more people can talk, and then they call for differentiated roles so everyone has a job to do. This
type of treatment equalizes status and renders participation more active and even. Another problem
in schools concerns group norms and peer influence through cliques. To offset this, scholars suggest propinquity changes such as rotating groups
and seating assignments.
These are clearly simple examples, but they
convey cleanly how its feasible to both facilitate
the emergence of certain network forms, and to redirect them in different directions.

Network Forms of Organization

To this point, I have discussed a social network perspective on organizations. For the rest of
this chapter, I will relate network organization as a
distinct theory of organizing, or how organizations
coordinate networks or partnerships so as to improve the delivery of a service.
In the last chapter, we discussed resource dependence theory and described how its theory concerned a firms power-dependence relations in the
environment. This chapter, a variety of networks
are described in the readings, and if we inspect

133

! Resource Dependence
"

! Network Organization
"

Focal Organization, Dyadic

Exchange

Broader context of
relationships

Figure - RDT and Network Interpretations

them a little closer, we see they are built up by

some of the bridging efforts discussed last week in
resource dependence theory: joint ventures, associations, interlocking boards of directors, strategic
alliances, and partnerships.
For example, in this chapter, several authors
describe project-based networks or issue-networks
where the effort is to organize around a specific
project or to work together to push a single issue.
If you recall, the last chapter on resource dependence theory discussed how this arose in pairs of
firms through joint ventures.
In this chapter, we discuss professional networks and trade associations, as well as interlocking boards of directors that bring firms into greater
communication with one another. In resource dependence theory, these were seen as a means of coopting other firms and sharing resources like information.
In this chapter, we discuss Smith and Wohlstetter description of organizations as forming a
group affiliation, or where sets of organizations

form a family and work together voluntarily

(2001). We also review some of Goldsmith and
Eggers work where they discuss external partnering, such as when governmental agencies contract
out particular tasks to private companies and nonprofits in belief that coordinating providers will enable government agencies to serve citizens more
efficiently (2004). In resource dependence theory
there were pair-wise efforts like this called strategic alliances or agreements, and they were performed to secure and /or prevent advantages or to
pool resources and work together.
It is important to note that the sorts of partnerships described in network forms of organization are not mergers where total absorption of one
firm into another occurs. Instead, they are partial
absorptions and strategic alliances. Nevertheless,
the point here is that many of the readings referenced in this chapter describe networks formed by
bridging efforts that resource dependence theory
discussed. Nonetheless, the focus shifts from pairwise relations to the entire network. At issue for

134

network organization is how to coordinate and

manage organizations that handle different facets
of provision.
If I had to put my finger on one key difference between resource dependence theory and network organization, I would say an important one is
their unit of focus. Resource dependence theory
considers the egocentric view of an organization
and its immediate relations. By contrast, network
organizations consider the global, socio-centric
view of both direct and indirect relationships. Network forms of organization see the network as constraining and enabling action; and that the wider
network of organizations is a source of stability
and change for the focal organization. The reading
by Davis and Powell (1992:334-341) describes the
difference between resource dependence theory
and network organizations best: resource dependence theory views a traffic jam of cars from your
own car, while network organization views the traffic jam from a helicopter. From which perspective
do you get the best understanding of your cars
movement?
If we schematize this (see prior page), we
can see resource dependence theorys view of a
firm in the far image. The direct, egocentric network is in focus. In contrast, network organization
looks at the broader array of indirect ties beyond
the focal firm. And depending on how broadly
one looks into that network, the situation can
change. Here, it is from one of being peripheral to
a group, to one of brokering groups.
One can also find similarities between network organization and coalition theory. If you recall, coalitions were like an interest network or a
temporary alliance. By contrast, network forms of
organizing reflect a persistent structural property
or a particular coordination pattern that is maintained over time. In short, the network form of organization is neither a coalition nor a resource dependence set.
Network Organization even has similarities
with organizational learning theorys description
of communities of practice and networks of practice. Many of you will recall this image. For organizational learning the focus was on practice
and the individual relations between participants

employing these practices. Each cluster resembles

a community wherein discussions of practice occurred. If we stopped there, each group would
come to their own optimal solution, or local optima. The linkages across communities reflect the
network of practice that enables the transfer of solutions across groups, and thereby facilitates their
reaching global optima. As such, organizational
learning focuses on individual actors and their relations within and between organizations.
By contrast, network organization focuses on
organizations as the unit of analysis and discusses
the patterns of interconnections across firms or
the interorganizational network. Hence, if we aggregated the far image we would acquire some
sense of how network organization views the same
situation. Moreover, it considers multiple types of
relations across firms. So network organization
has some similarity to prior theories, but it also differs. Let's look more closely at how organizational theorists have related the details of this theory.
An article by Borgatti and Foster (2003) relates a brief history of network organization. They
argue that at the turn of the 20th century, network
organization was a fashionable description for repetitive exchanges among semi-autonomous organizations relying on trust and embedded social
relations to protect their transactions and to lower
costs. At the time, proponents of network organization argued that markets and hierarchical structures had become inefficient as commerce grew
more global, hyper-competitive, turbulent and technologically dynamic. In their stead rose a network
form of organization that balanced the flexibility
of markets with the predictability and stability of
hierarchy and this brought intelligence, flexibility, and speed of response to organizations adopting the model. Network organizations had more
enduring and diffuse connections than did markets, and they had more reciprocal and egalitarian
arrangements than hierarchies (Scott 2003:282).
As such, they were a sort of middle ground, distinct from market and hierarchical forms of organizing. The network form of organization entailed
interdependent firms that competed successfully
with larger corporations. Network organization

135

Key Factors

Hierarchy

Network

Market

Normative basis

Employment
relationship

Complementary
strengths

Contract property
rights

Means of
communication

Routines

Relational

Prices

Mode of conflict
resolution

Administrative fiat supervision

Norms of reciprocity

Haggling resort to
courts

Degree of flexibility

Low

Medium

High

Amount of
commitment

Medium to high

Medium to high

Low

Tone of climate

Formal, Bureaucratic

Open-ended, mutual
benefits

Precision and/or
suspicion

Actor preferences

Dependent

Interdependent

Independent

Networked
Organizations:
Neither Hierarchy
nor Markets "
"
(Adapted from Powell 1990: 300)

had become possible because many organizations had become increasingly specialized and information technologies (phone, fax, email, teleconferencing, computing, etc.) made it feasible for them to coordinate delivery of services.
Woody Powell was one of the first to describe network organization
as an intermediary form between hierarchies and markets, and he elaborated its distinct logic of exchange (1990). You can find this paper in the
additional readings of the week. Powell argues that network organizations are neither market, nor hierarchy. They entail more enduring and
diffuse connections than markets but more reciprocal and egalitarian arrangements than hierarchies.
Lets look at an adapted table from Powells 1990 article where
these forms of organizing are compared. I have reordered and slightly edited the table for ease of presentation here. Powell goes through a variety
of organizational features and processes and contrasts them for each type.
For example, we all recall what a hierarchical form of organization is: it
is a centralized organizational chart with levels of reporting that winnow
down like a pyramid to the top. In these systems, the normative basis of
association is an employment relation (reporting); the means of communicating entail established routines; conflict is resolved by administrativeoversight; there is little flexibility in procedures; worker commitment to
the firm is medium to high; the tone of the climate is formal and bureaucratic and actor preferences are dependent on the firm and its centralized
actors.
136

At the opposite spectrum is a market form of

organization where associations are guided by formal contracts and transactions; where the means of
communication are prices; where conflict is resolved by bargaining and haggling; where flexibility is high, commitment is low (everyone is out for
themselves without constraint!), and where the
tone is based on precision and suspicion of competition and actor preferences are independent of
each other just as one would expect in a free market. Many scholars in economics view organizational fields in this way, and they have models following these logics.
Between these two poles is the network form
of organization. It has characteristics that distinguish it from market and hierarchical forms of organization. Through network organization, firms
seek out complementary strengths in forming collaborations, they communicate through their network of relationships, and they resolve conflicts
via norms of reciprocity. In network organizations
flexibility is moderate, as the actors are constrained by their pre-existent ties, but they are not
more fully determined as in a hierarchical organization. Participants of network organization experience moderate to high levels of commitment, they
find the climate to be one where mutual benefits
are sought, and where most firm preferences are
interdependent, if not complementary.
Different forms of network organization are
feasible, and each form can address different kinds
of problems. Cross, Lieftka, and Weiss (2005)
have a nice article describing different network
forms of organization and the sorts of problems
they are best suited for. They discuss networks
within and between firms, but I think you can
imagine sets of firms, even small firms, forming
similar patterns of collaboration so as to compete
with larger firms adopting this sort of internal patterning.
The first is a core-periphery structure where
there are dense internal ties and extensions into the
environment for novel information. These networks are like those proposed in the theory of organizational learning. There, a central hub can
process and transfer information and the external
ties reach out into the environment. This type of

network is arguably well-suited to addressing ambiguous problems in need of innovative solutions

and finding ways to implement them in local conditions. One can imagine a network of organizations
arranged in a similar way to address ambiguous
problems.
The second form of network has linked
cliques that afford a modular response to problems. Here, a complex problem can be broken
into components addressed by each unit or cluster,
and they can be dealt with in a sequence. Again, a
set of organizations can arrange their relations in
this manner to do the same.
The final network is a simple chain format,
and there the network organization can deal with
familiar problems with known responses its a
format you might see in a streamlined organizational process model where efficiency is an issue.
Now clearly, the first two forms are more in
line with what scholars mean by network organization. However, it is helpful to consider the fact
that even between the first two there is some variance in form that might influence coordination and
delivery.
Case: US Governmental Agencies Utilizing Network Forms of Organization
In addition to Powells theoretical exposition
on network organization (1990), we have Goldsmith and Eggers applied look at network organizing (2004). Goldsmith and Eggers discuss US government agencies again, but I think you will find
this sort of network organization relevant to many
other parts of the world. And perhaps they are an
especially salient form of organization for markets
at the margin, such as alliance networks on environmental sustainability. Note that when I say
markets at the margin I mean that most firms see
environmental issues as secondary to their main
interest in profit and survival. As such, this is minor interest where other firms like a government
agency or lab is trying to foster a network that facilitates change on this issue in firms.
In the United States, all too often the agencies adopting network forms of organization are

137

Figure. Golden Gate National Recreation Area (GGNRA) "

"

"

"

"

(Source: http://commons.wikimedia.org/wiki/File:Map_of_Golden_Gate_National_Recreation_Area.png; http://commons.wikimedia.org/wiki/

File:Golden_Gate_Bridge_from_Marin_Headlands_1.jpg)

138

ones that have fewer resources and work to become more efficient. Or they are agencies that
find themselves unable to perform the task required or they want to distance themselves from
the work they need performed. As such, the
agency coordinates the web of service providers or
they hire a coordinator.
Many US government agencies have adopted
the network form of organization. Government
agencies contract out more and more tasks to private companies (for and not-for-profit forms) in
the belief that competition among providers will
increase efficiency. Today, public agencies find
themselves working in a world of partnerships and
networks (Goldsmith and Eggers 2004). They
form alliances that include mixtures of agencies,
large and small organizations, and so on. This sort
of network form of organization is seen as a viable
alternative to large-scale corporations and hierarchical public bureaucracies. Goldsmith and Eggers describe how government agencies hire contractors and they hire subcontractors in an effort to
provide a service. In those instances, the agency
integrates and coordinates the web. But in other
instances, the government agency either wants
more distance with the service, or they find a third
party provider can coordinate better. Here, there
forms a slightly different structure.
So government organizations find themselves working in a context of partnerships and alliances. In short, they are engaged in networks of
smaller and larger organizations that span public
and private sectors, and this is widely seen as an
alternative form of organizing in comparison to
large-scale corporations and public bureaucracies.
Many of you can probably relate a variety of
examples of network forms of organizing in your
respective industries and parts of the world. I will
give you a few examples. Goldsmith and Eggers
describe the Golden Gate National Recreation
Area (see image on prior page), and how it heavily
relies on partners to take care of the parks and
their services. Only 18% of park services rely on
forest service employees the rest relies on partners.
We can also consider the Iraq War as an example where firms like Bechtel and Halliburton

were contractors that coordinated a variety of services and worked on the reconstruction effort.
Now I do not know the details of this operation
and how it went. But in terms of network organization, it is an interesting case. One could arguably
believe that there was little local Iraqi trust for the
United States agencies. Is that why contracting
firms were used? And even then, was their much
trust to make the network form of organizing helpful? And again, I have to plead ignorance here,
but I think it interesting to ponder how the network form of organizing was applied and how it
performed in such a context.
A final example for the network form of organizing is the Manhattan Project (see image next
page). Here, over a dozen universities and a network of scientists, engineers, military agencies
and service providers were brought together to create an atomic bomb. To many, the project was
such a success that they now see it as a network
form they want to repeat in other areas of knowledge creation. So for example, I am often asked to
study interdisciplinary and transdisciplinary projects and centers that bring together a heterogeneous set of individuals and ask them to learn from
one another and create new ideas they never would
form if they stayed in their respective disciplinary
silos. The model of organizing is very much a network form. And many see it as a potentially powerful one for organizations engaged in todays
knowledge economy.
Why do network forms of organization like
this come about? In chapter 1 of their text, Goldsmith and Eggers relate a few reasons why governments use network organization, but we can extend it to firms more generally (2004). The first
reason is that a firm lacks the capacity to provide a
service and that it must rely on other firms. For
the government, this means the use of for profit
and non-profit firms as contracts and subcontracts.
The second reason is to provide more integrated
services. Outsourcing alone is not enough. It
merely creates 4 subcontractors and narrow channels to a service that would have existed via 4 government agencies anyway. Network organization
calls upon agencies and subcontractors to join-up
or partner horizontally and vertically so as to pro-

139

Figure. Manhattan Project "

"

"

"

(Source - http://commons.wikimedia.org/wiki/File:Y12_Calutron_Operators.jpg; http://

commons.wikimedia.org/wiki/File:Manhattan_Project_emblem.png)

vide more integrated services: a one stop shopping experience instead of

a disparate 4-stop one. A third reason is the digital revolution and technological capacity. Technology has made external partnering far more viable. Companies can share scheduling information on production, demand, shipments, transactions, etc. All sorts of information can be shared
by digital means and often instantaneously, making a variety of partnerships across small, specialized firms feasible. A final reason for the emergence and expanded relevance of network organization is demand. Citizens today want more options and choices, and they have become less tolerant of mediocre service. In a network organization, there are multiple
contracts and subcontractors, many of whom compete to meet demand
and win consumer choices.
Goldsmith and Eggers contend that the network form of organization has certain advantages for government agencies (2004): The first is
that it enables experimentation by allowing agencies to explore wider sets
of alternatives of service provision. The second is that the government
can focus more on management and delivery by outsourcing tasks to the
best providers and experts. Network organization also increases managerial flexibility. The government agency often finds it can provide services
more quickly and change the nature of service by drawing on multiple resource providers. Last, network organization is a decentralized, fluid
form and the autonomy of allied organizations enables citizens to play a
greater role in decision-making. Network organizations listen and react
more quickly to citizens and consumers than do large hierarchical firms.
140

A core feature of network organization is the

creation and maintenance of partnerships. You can
find this in chapter 5 of Goldsmith and Eggers, but
I think it helps to consider factors of partnership
formation and splitting because it is a central
means of successfully forging and undermining
network organization. On the far side there are factors that form a trusting partnership and sustain it.
On the near side are factors that create competition, distrust and effectively split partnerships
apart. Many of the features reflect common sense.
For networked organizations, partnerships
are healthy and sustained when the organizations
address discrete functions: either the discrete functions an agency needs to address, or the discrete
functions a firm needs to address in order to keep
itself alive. Also, the firms need to cooperate on
matters outside their core business. So one firm
will coordinate and align with another in a division of labor each needs from the other, but not the
same labor they both produce. That is, they do not
want to compete and they need to regard one another as differentiated collaborators, or as rolecomplements in the provision of services. Firms
can sustain healthy collaborations when they are
open and trusting with information and they lack a
history of competition if there is any sense of
competition or lack of trust, the collaboration will
disintegrate. Last, it is important that the firms involved do not regard the information in the current
collaboration as proprietary or to see it as putting
them in a disadvantage with one another later.
Managing Network Forms of Organization
Managing a network of organizations is not
easy. There are a variety of things you can do to
make network organization work better for you
and your clients. First, you need to weigh partners
carefully! You want trusted collaborators that can
assume role-complement with one another. As
such, each pair becomes the ying for the others
yang, and vice versa. You do not want a network composed of directly competing firms or it
will be undermined. As a network manager, you
must think about the larger network structure and
how it can be integrated. How can the larger net-

work of alliances work so as to afford a suite of

services that encourage firms to join up or across,
and that clients will want to utilize. Next, go
deep! Try and align the goals and cultures of these
firms so they value collaboration, trust and openness. If you can forge these beliefs and values,
then its likely their surface relations will manifest
as healthy, complementary partnerships. Developing the capacity for group processing is also important. Here, the network manager needs to listen to
other firms and include them, but somehow still
move the network forward on the interests everyone shares. In addition, the network manager
needs to focus partners on their discrete functions
(so there is no internal competition) and then coordinate their activities. The manager needs to ensure there is no direct competition between companies in the network of provision.
Network organization also responds well to
shared information on performance. Since the
functions are differentiated and action distributed,
it helps to let other parts of the network know
what the others are doing and how their coordination relates to performance. As such, open access
and discussion of performance data is frequently
beneficial to the manager and firms in the network.
All the features related thus far build trust,
but there are a few other things one can do to build
trust and manage relations. An obvious one is to
bring out into open any initial contention. Another
is to create a joint governance structure and shared
decision making that spans partners. As such, everyone has a stake and a responsibility in decisions
so the network holds and proceeds.
Again, network management is not easy. Often, one finds they are constantly managing relations and the network as a whole in ways akin to a
coalition. However, the network is more stable if
done right: where firms regard one another as complements to one another and that in the whole they
form a system of service provision that is superior
to what they can do on their own. Such an arrangement extends well beyond a single decision, and
toward the repetition of many.

141

Summary of Network Forms of Organization

So lets summarize what we know about the
theory of network organization. The first question
to ask is when does it apply? Does it apply if I am
looking at a particular decision among individual
workers? Probably not so well. But it does seem
well suited to studying the wider context of organizational relations and how they influence the organizations behavior and survival.
As for the general summary or argument, I
suppose we can attempt a caricature: Organizations focus on network relations, positions, and the
larger context in developing strategy and deciding
their behavior. Multiple types of networks are feasible (trust, exchange, etc.) and they can guide resultant firm behaviors. If we consider how the organizational elements are typically related we also
get a sense for where the theory focuses and where
its concern rests. For example, the participants in
this case are organizational stakeholders engaged
in the network of organizations or which are potential partners.
The goals of an organization attempting to
create a network organization are to deliver a service via collaborations and outsourcing aspects not
central to its technological core. The technology
by which the network organization forms is linking, coordinating, allying firms in order to deliver
a service. It entails outsourcing, subcontracting
and partnering in order to focus on the core technology. The social structure then consists of communication and coordination relations and the positions and roles therefrom. The network as a whole
and its pattern influences the organizations output
and performance. Undergirding the network is a
norm of trust (not competition) that allows the interdependent organizations to work together. Last,
the environment network organizations extend
well into the environment. In fact, its focus is on
inter-organizational relations in the environment.
So the array of elements and their characterizations are distinctive. If we consider the dominant pattern of inference or means of organizational action, we learn further how it differs from
other theories presented in this course. Organizations trying to accomplish network organization

identify complementary strengths, form alliances,

establish collaborative norms, create opportunities
for open-ended mutual benefits, and outsource secondary tasks all to survive and create a positive
network environment through which firms can
complement one anothers needs and delivers a
service.
And last, how can one manage a network
form of organization? You can manage it by designing the network in a way that selects partners
wisely so their values and efforts complement and
align with one another. You establish frequent, informal, active communication channels with the
involved organizations. You coordinate member
activities by preventing internal competition between collaborators, by creating access to shared
information in the network, by forming a shared
decision-making structure, and by getting participating firms to focus on their distinctive functions
and the coordination across them. Last, you establish a norm of collaboration and reciprocity in the
network. In this manner, you create a distributed
organization in the environment. Instead of housing divisions and functions within a single firm, it
is rendered into a network form in the environment. And this accomplishment requires a distinctive set of managerial approaches and efforts to
make it work.

142

References
Bender-deMoll, Skye and McFarland, Daniel A.
(alphabetical listing). 2006. The Art and Science
of Dynamic Network Visualization. Journal of Social Structure, vol. 7, no. 2.
Borgatti, Stephen P. and P.C. Foster. 2003. A Network Paradigm in Organizational Research: A Review and Typology. Journal of Management
29(6):991-1013.
Cross, R., Liedtka, J. & Weiss, L. (2005). A Practical Guide To Social Networks. Harvard Business
Review 83(3), pp. 124-132.
Dahlander, Linus and Daniel A. McFarland. 2013.
Ties that Last: A Longitudinal Study of Tie Formation and Persistence. Administrative Science
Quarterly 58 (1)69110.
Davis, Gerald F. and Walter W. Powell. 1992.
Selection from Organization-Environment Relations (pp. 334-341). In Handbook of Industrial
and Organizational Psychology, Vol 3 (2nd ed.).
Eds. Marvin D. Dunnette and Leaetta M Hough.
Palo Alto, CA: Consulting Psychologists.

Nohria and Robert Eccles. Boston: Harvard Business School.

Powell, Walter W. 1990. Neither Market nor Hierarchy: Network Forms of Organization. Research in Organizational Behavior 12: 295-336.
Powell, W.W., Douglas R. White, Kenneth W.
Koput & Jason Owen-Smith. 2005. Network Dynamics and Field Evolution: The Growth of Interorganizational Collaboration in the Life Sciences.
American Journal of Sociology.
Rawlings, Craig and Daniel A. McFarland. 2011.
The Ties that Influence: How Social Networks
Channel Faculty Grant Productivity. Social Science Research 40: 1001-17.
Roethlisberger, F. J. and William J. Dickson. 1939.
The Internal Organization of the Group in the
Bank Wiring Observation Room. Management
and the Worker (chapter xxi, pp. 493-510). Cambridge, MA: Harvard University Press.
Smith, Andrew K. and Priscilla Wohlstetter,
2001. Reform Through School Networks: A New
Kind of Authority and Accountability. Educational Policy 15, 4:499-519.

Goldsmith, Stephen and William Eggers. 2004.

Governing by Network: The New Shape of the
Public Sector.
Granovetter, Mark. 1985. "Economic action and
social structure: The problem of embeddedness,"
AJS 91:481-510.
Granovetter, Mark. 1995 (1974). Getting a Job: A
Study of Contacts and Careers. Chicago: University of Chicago Press.
Kilduff, Martin andWenpin Tsai. 2003. Social
Networks and Organizations. SAGE Pub.
Krackhardt, David. 1992. The Strength of
Strong Ties: The Importance of Philos in Organizations. In chapter 8 of Networks and Organizations: Structure, Form, and Action. Eds. Nitin

143

9
Neoinstitutional Theory

(Source: http://commons.wikimedia.org/wiki/File:Young_Afghan_girls_inside_the_classroom_of_Aliabad_School-2012.jpg;
http://commons.wikimedia.org/wiki/File:Newells_classroom.jpg/; http://commons.wikimedia.org/wiki/File:Classroom_in_India.jpg;
http://commons.wikimedia.org/wiki/File:BMS_classrooms.jpg; http://commons.wikimedia.org/wiki/File:Ashs-learning-common-miro.jpg))

Neoinstitutional Theory
In this chapter, we will continue our discussion of organizations as open-systems whose survival depends on their relation with the environment. In particular, we will discuss one of the prevailing organizational theories stemming from sociology, called neoinstitutional theory. In oversimplified terms, one can think of neoinstitutional theory as arguing that an organizations survival depends on its fit with the cultural environment.
That is, a firms success depends on whether it
adopts structures that are deemed rational and legitimate in the external environment; that the firm
mirrors environmental beliefs about what a legitimate organization of that type should look like.
Neoinstitutional theory has always been one of the
harder theories for students to fully grasp, so I
have organized the chapter to be a little repetitive.
I will discuss many of the core concepts twice and
relate them in different ways so you get a better
sense for what this theory conveys.
Introduction to Neoinstitutional Theory

Neoinstitutional theory tries to explain institutional isomorphism, or how the same organizational forms develop, spread, and become legitimated in one sphere of activity after another. The
theory tries to explain how and why spheres of activity, like organizational fields of biotechnology
or education, are composed of organizations that
look more alike than they differ. Lets take the example of the organizational field of education:
why do most schools and classrooms look the
same? I recall talking with one of the founders of
neoinstitutional theory, John Meyer, and he recounted his travels all over the world visiting
schools and classrooms. He described how he had
visited typical American schools; poor subSaharan villages with classes taught outside in
these ground indentations without chairs and tables; how he had seen religious fundamentalist
schools in Saudi Arabia where boys and girls were
taught separately; and even wealthy Western
schools with computer tablets in every hand. In

spite of the different locations and cultural distances traversed, all had enough similarities that
one knew right away what kind of organization it
was, and what scripts were being referenced -they were all school classrooms!
In many regards, all these settings conform
to widely held institutional beliefs about what
schooling entails. These beliefs and conceptions
are cultural-cognitive controls, or deep social
structures in the environment. As Richard Scott
relates, they are sets of beliefs developed in social interaction, provide models, schema, and
guidelines for governing and guiding behavior in
social situations (Scott 2003:119).
Institutional controls are practiced in several
forms (Scott 1995:34-45). An explicit form of institutional control is practiced through regulations
or regulatory institutions. These constrain behavior through rules or laws and behavioral inducements like incentives and punishments. A second,
deeply ingrained institutional control is normative.
Normative controls guide what we should or
should not do, or how we should and should not
appear. In great part, these are informal rules and
guidelines, but they are just as influential on organizational behavior as laws and regulations. Last,
there are institutions that run very deep and these
are cognitive beliefs. Scott argues that compliance
with cognitive institutions occurs in many circumstances because other types of behavior are inconceivable. Cognitive beliefs are naturalized, taken
for granted ways of doing things, such as take for
granted routines and activities (Scott 1995:40-45).
In many instances, these institutions are layered on top of each other in reinforcing ways like
an onion. But in some instances they conflict and
segments in the environment adhere to one set
over another. As such, the cultural environment
can be varied. Organizations typically respond by
building that external complexity into their internal, formal structure, however.
I always find it easiest to distinguish the
three forms of institutions with the example of a
sport. I am going to take the game of soccer and
describe how these three institutional controls can
be layered so as to make the performance of soccer games relatively the same and recognizable.

145

Different contexts of
soccer activity "
(Sources: http://commons.wikimedia.org/wiki/
File:StateLibQld_1_194039_Shot_for_goal_duri
ng_a_soccer_match_in_Brisbane
%2C_ca._1937.jpg/; http://
commons.wikimedia.org/wiki/
File:Uz_vs_Jap_2009-Free_kick_%28before
%29.JPG; http://commons.wikimedia.org/wiki/
File:Israel_v_Brazil_1.jpg/; http://
commons.wikimedia.org/wiki/
File:Defense.gov_photo_essay_100503F-3745E-376.jpg)

What are the regulatory controls of soccer? Those are the rulebooks
and rules of soccer, as well as the referees acting as agents to enforce
them. Penalties are incurred for violating the rules in this case.
What about normative controls? The norms of soccer characterize
our notions of better and worse players, better and worse sportsmanship,
and so on. Norms lead players to act in certain styles within the tacit activities and routines they enact.
What is a cognitive or deeper form of institutional control? Cognitive institutions are taken-for-granted routines. For soccer, these entail
the activity of soccer itself. It is inconceivable that someone would approach the game of soccer using a different activity schema and roles of
say basketball. The enactment of a soccer game is taken for granted and
persons engage in it unquestionably. If they do not, everyone gets upset:
norms and regulations are violated. And we find this cognitive layer
when we go to different contexts of soccer play. Regardless of whether it
is a game in 1937, a playground game, or even one on a beach, all of
them share a family of resemblance in their routine such that we regard
them as soccer.
So multiple institutions can control behavior and render them into
scripted forms that are deemed legitimate and ideal. For any organization
their actions might be driven by taken for granted routines and activities,
norms and expectations of best practices and players, and explicit surface
regulations catching violations. John Meyer, Brian Rowan, Paul DiMaggio and Walter Powell were some of the first neoinstitutional theorists,
146

and we rely heavily on their ideas in our summary

of the neoinstitutional approach (Meyer and
Rowan 1977, 1978; Dimaggio and Powell 1983).
They wrote about how organizations look alike because there are processes leading firms to adopt
many of the same institutional controls. In particular, they stressed the important role of particular
rationalizing agents who generate these institutional controls or ritual classifications. These
agents were governmental units, professional
groups and associations, universities, and even
public opinion. The classifications they propose
are Scotts aforementioned cultural-cognitive categories, normative beliefs, and regulatory policies
and laws. The basic idea is that scientists and professionals increasingly work at the world-system
level holding international conferences, issuing
statements, providing recipes and policies for reforming and rationalizing one sphere of activity
after another (from health standards, to human
rights, to education).
Going back to my conversation with Meyer,
I remember him saying that in spite of the immense cultural differences of nations from being
socially liberal Western societies, to orthodox Muslim ones, to remote cultures in poor regions of developing countries -- they all seemed intent on
adopting topics taught in Western schools and the
progressive forms of pedagogy espoused by their
educational professionals. All the schools had
similar subjects; they used many of the same instructional formats; and they all seemed intent on
improving themselves by emulating pedagogy
deemed legitimate by various professional associations, non-profits, and other rationalizing agents.
Meyer argued this was happening because rationalizing agents proposed the classifications and typifications, and that they were regarded as rational
and legitimate even if their returns to efficiency
were not fully established in each case (or even at
all).
In sum, from the neoinstitutional vantage
point, organizational survival and success are contingent on integrating institutional beliefs (or ritual
classifications) from the environment that are believed to be signals of legitimacy. In most cases,

institutions are legitimated when they are widely

held and believed to be rational.
Comparing Resource Dependence Theory, Network Theory, and Neoinstitutional Theory

Now that you have an initial sense for neoinstitutional theory, let us contrast it with theories
discussed previously in the course. In particular, I
think it helps to compare neoinstitutional theory to
prior open system views and to prior cultural arguments since they are the most relevant. As you recall, resource dependence theory offers strategies
thought to be effective in exchange environments.
In contrast, neo-institutionalism offers strategies
thought to be effective in environments replete
with institutionalized beliefs about organizations
and their appearances. There is a shift here in how
firms view and respond to their environment: from
a logic of consequence (resource dependence theory) to a logic of appropriateness (neoinstitutional
theory). Neoinstitutional theory argues that organizations survive and succeed in their surrounding
environment by not only accomplishing economic
fitness and efficiency, but from accomplishing a
social and cultural fit with the environment.
You saw me remark in a past chapter about
Disney and its various theme parks. Disneys efforts actually reflect neoinstitutional arguments
about cultural fit to some extent. Disneyland in
the USA has a particular feel, a particular food
menu, and other features that are not easily
plopped down in another cultural context. In the
case of Euro-Disney, the company needed to take
into account the beliefs of the local environment,
and adjust its for-profit model and American
theme-park script to local views. The end result is
a different version and feel of the Disney theme
park: so less junk-food, less shopping mall and
main street appearances, and more old-world
charm.
Lets compare and contrast facets of resource
dependence theory with neoisntitutional theory so
you have a better sense of their differences. A lot
of the concepts I am going to introduce may seem

147

Unit of
Analysis
Change

Resource
dependence theory
and neoinstitutional
theory compared.

Resource Dependence
Organizations with resource
dependencies

Neoinstitutional
Organizational fields

Coordination of resources
Greater homogeneity of field
(greater interdependence and as rational myths spread
stability over time)

foreign at first, but do not worry, I will come back to them again in the
chapter so you get a richer understanding.
If we consider each theory, we can view them on a variety of dimensions (see table above). The first is their unit of analysis. Resource dependence theory is primarily focused on resource dependence relations
that an organization has with other firms. Neoinstitutional theory is concerned with entire organizational fields, or domains of activity where the
firms are aware of one another as relevant to that domain. Both theories
focus on the environment, but they target slightly different things in it: resource relations for one and cultural matching for the other. They see
change differently as well. Resource dependence theory argues there is a
movement toward greater coordination of resources, or greater interdependence and stability over time, while neoinstitutional theory sees a progression toward greater homogenization as legitimate classification
schemes spread. These changes are promulgated by different processes.
In resource dependence theory, the managers try to minimize their own
firms dependence on others while they increase the dependence others
have on them. By contrast, neoinstitutional theory generates change via
institutional isomorphism where each organizations effort to survive
and secure resources leads them to fall in line with external cultural pressures and rationalized myths on what a legitimate firm should look like or
what an ideal product should be.
Each theory also offers a distinctive view of an organizations structure: one that is characterized by dependence relations, and one whose for148

mal structures and classifications are radically decoupled from the technical core (this is the concept
of loose coupling, and we will discuss it at length
later). Last the theories espouse distinctive organizational needs. Resource dependence theory says
firms need resources and autonomy for survival;
neoinstitutional theory says firms need environmental legitimacy so as to secure resources and survive.
Theories from other chapters can also be contrasted with neoinstitutional theory. Research on
networks falls somewhere in between resource dependence theory and neoinstitutional theory, with
some scholars working to align network research
with neoinstitutionalism. The diffusion of particular formal structures, appearances, reforms and/or
practices through networks is a way these two literatures interrelate. However, cultural scripts,
norms, and ideas are not easy to locate in networks.
In some ways neoinstitutional theory
aligns with the notion of standard operating procedures and organizational culture, but neoinstitutional theory places much more emphasis on
taken-for-granted norms or ways of doing business instead of formalized rules and codebooks
for behavior. Moreover, neoinstitutional theory abstracts away from focal organizations to the field
level, taking organizational culture to a macro
level.
Features of neoinstitutionalism

Now that we have a general sense for neoinstitutional theory and how it compares to previous
theories in the course, we can begin to delve more
deeply into its core concepts. At this point in the
chapter, I will draw on a couple additional primary
sources. In particular, I want to discuss the basic
ideas presented in the 1977 piece by Meyer and
Rowan, "Institutionalized Organizations: Formal
Structure as Myth and Ceremony", and the 1983
paper by Dimaggio and Powell, The Iron Cage
Revisited. As we go along, Im going to draw out
examples on schools as an application since both
Meyer and Rowans 1978 paper and Mary Metz
1989 paper both do a terrific job of providing us

concrete examples of how neoinstitutional theory

applies in those settings.
The seminal article for neoinstitutional theory is Meyer and Rowans 1977 paper. In that paper, the general argument is that, independent of
the drive for efficiency, organizations ceremonially incorporate institutions into their formal structure that are believed to be rational. By doing this,
organizations gain legitimacy and secure social resources from the environment. The institutions
they incorporate are things like regulations and procedures, classifications, rules, and practices. Note
that I said that these institutions are incorporated
because they are believed to be rational. The
adopted practices and formal structures are called
rational myths, or legitimated institutions we
adopt on the assumption they are rational, but we
do not investigate whether they really improve efficiency or not. They are taken for granted as such.
They are believed to be rational.
These institutions are built into society as
typifications and ritual classifications. For example, we believe educational institutions are more
legitimate when they have buildings, classrooms
with chairs, lectures, student-teacher roles, mathematics and other core subject matter, credentials,
and so on. These roles, classifications and rules
are ceremonially applied much like we enact a
script and adopt appearances in various ritual ceremonies. The ceremonial adoption of such appearances is done on the basis of belief. They are
myths because we believe they are legitimate
forms to use (we take them for granted as natural)
and we view them as rational myths because we
think they help the organization function better
without actually investigating their relation to efficacy.
In order to survive in modern societal environments, organizations must be regarded as legitimate, and this legitimacy is accomplished by maintaining ceremonial conformity.
Organizations
look their part in an ongoing script or play for that
type of organization. Hence, their formal structures are organized to reflect the rational myths located in the external environment. This conformity leads organizational fields to have organizations that look more alike than they look different.

149

The key point here is that organizations

adopt institutional rules as rationalized myths.
They are rationalized because they are impersonal
prescriptions identified in a rule-like way as the
appropriate means to pursue various goals. They
are myths because we adopt them on faith or in a
taken-for-granted way. We believe they are rational constructs, but we seldom look deeply at
whether they are efficient or if other constructs
would work better. We are limited problem solvers who adopt rationalized myths. We use shorthand logics encoded in the environment. The efficiency and efficacy of standard operating procedures and organizational structures is presumed on
the basis of their wide adoption and / or the endorsement by professionals like Stanford academics. As such, the sources of legitimation vary from
public opinion, ideologies, regulatory structures,
certification and accreditation bodies, professional
norms, credentials and government requirements.
Where do rationalized myths come from?
What are their origins? They arise in a context of
dense, complex networks (e.g., in the context of
modernization); and they arise in an effort to make
rational decisions where there is much ambiguity and uncertainty. Rational myths and the reliance on rationalizing agents are a short-hand
means to deciding. They diffuse through networks, and they are passed because the practices
are believed to be rationally effective. Rationalized myths are used because leaders within each
organization want their firm to have legitimacy in
the wider environment: Whereas, resource dependence theory saw this arising from the creation of
resource demand (so the manager built greater external dependence on their organization), neoinstitutional theory seeks to create demand by mirroring institutional rationalized myths in society. By
looking like the real deal, or like an exemplar,
they garner attention and resources from the environment.
This notion of a rationalized myth can extend to organizational products. Take car advertising as an example. When creating the idea of
what a good car is, advertisers project appearances
of the firm and its products as if they exhibit externally legitimate, rational myths. For example, con-

sider a Jaguar. It is a nice looking car, but what

makes you think its legitimate? Is it the cars performance? Sure - perhaps. A neoinstitutional ad
would trumpet various awards regardless of what
they are for (safety or speed?). Such awards are
rationalizing agents, and claims to any award are
great. A good example of this can be seen with
movie reviews in newspapers. Many of us recall
looking up movie listings in the paper and seeing
reviews. In many cases, the critics giving a new
movie good reviews are unknown, so they give the
movie the appearance of legitimacy when it likely
is not.

Figure. Jaguar and a Car Award

(Source http://commons.wikimedia.org/wiki/File:Volt_MT_COTY_WAS_2011_835.JPG/;
http://commons.wikimedia.org/wiki/File:Jaguar_XKR_-_Flickr_-_The_Car_Spy_%2819
%29.jpg/)

Case: Real Schools

In many regards, the article on Real
Schools by Mary Metz gives you a clearer example of rationalized myths using the case of educational organizations(1989). In her paper she describes how educational organizations symbolically code their structures to resemble beliefs
about real school that are held in the institutional
environment. She thinks this is why American
high schools all look the same on the surface in
spite of being really different internally. They
look the same and plod along in spite of having differences in content and output. Metz describes
symbolic coding as arising when organizations
adopt a common script. The script is like that of a

150

play, and educational organizations play the part of

a real school.
These organizations engage in rituals or ceremonial performances by looking their part in the
play when interacting with the environment:
Hence real schools have buildings, classrooms, desks and chairs, etc. They have agegraded students, undifferentiated teacher roles, department chairs, principals, and various other staff.
They have differentiated course subjects, whose
scope and sequence are recognizable to colleges
and employers. They have familiar technologies
like lessons, many of the same tasks (lecture, recitation, seatwork, etc); textbooks, computers and
blackboards. They have coded time into school
days, school weeks, quarters / semesters and
school years. They use many of the same symbols
of ranking and completion like grades, test scores,
and credentials, many of which are used as ritual
classifications by external organizations.
All of these features are typifications that we
recognize and expect a school to have. We take
these features for granted, and we place confidence in them as being normal and rational without much inspection of their efficacy! In short, educational organizations put on a play, or the appearance of real school, in spite of some kids failing
in reality. The script serves symbolic purposes
more than technical ones. The same can be said of
universities and their development. Over the last
100 years or more, universities have grown increasingly common, and their forms increasingly isomorphic. New universities quickly adopt courses,
subject matters, departments, credentialed employees, and other ceremonial features of leading universities as rationalized myths of what a good university should be.
University structures have
grown increasingly complex over time as they try
to appeal to different segments of the institutional
environment. Consider what new universities, like
Qatar University, look like. Do they adopt dramatic shifts in ceremonial features or do they mirror exemplary universities?
How are rationalized myths sustained if they
arent efficient or optimal (Logic of confidence)?
The formal structure of many organizations is
adopted like a sacred ritual. Rituals are like mar-

riage rites, where people adopt a range of appearances and go through a series of scripted actions
so they resemble husband and wife, or they transform into such an embodiment. When we say an
organization reflects ritual classifications we mean
it displays appearances so as to embody a ratified
organizational identity considered legitimate in the
environment. To maintain the ritual and the plausibility of legitimacy, the organization presumes a
chain of confidences and adopts an assortment of
face-saving efforts to preserve this myth.
Here are a few of the face-saving efforts
used to preserve these myths. The first is avoidance. Avoidance is maximized when units are segmented so interaction across units is minimized.
In this manner one unit cannot see into another
and question their contents. The second is discretion. Discretion is maximized when inspection is
minimized and participants are cloaked in professional, credentialed authority. By placing trust in
teachers, we give them discretion and let their profession act as rationalizing agents. Last is the assumption of integrity. Here, organizations often
engage in ritual performances and their appearances have integrity, and this sentiment allows
them to overlook problems and label them as
anomalies.
In education, there exist a sequence of confidences that are never fully inspected: The state has
confidence in the district; the district in school; the
school in the teacher; and the teacher deserves confidence due to their degree and the programs accreditation. The accrediting agency doesnt inspect the teaching and skill of the graduate, but
has confidence in the college administrators, faculty, and the courses offered. These people in turn
have confidence in the teachers training them to
label certain courses as history without carefully
inspecting them. So it is a system of confidences
(Meyer and Rowan 1978:207-8).
Loose coupling

The sequence of confidences is greatly sustained by a structural adaptation called loose coupling. Organizations may all come to look alike

151

in terms of their formal or ceremonial aspects, but

that does not mean their actual internal practice
and activity are the same. Many organizations decouple their formal structure from technical activities and outcomes. But why?
Decoupling occurs in schools for several reasons: First, decoupling protects the formal structure from uncertainties of the technical core (buffering). There is much uncertainty in how curricula are delivered and received, and measuring their
effect is difficult. All too often inspection creates
doubts about the legitimacy of instruction, and we
lack clear working alternatives. Second, decoupling enables the organization to adapt to inconsistent and conflicting institutionalized rules (flexibility). The plurality of environmental pressures can
put conflicted demands on an organization. By differentiation and isolation, the firm can forgo coordination and avoid incompatibilities and inconsistencies. So for example, segmenting special education apart, the use of tracking and streaming, and
the formation of departments, are all means of
making an educational organization appear rational, but also of segmenting content inspection
(e.g., think of what accreditation efforts entail mostly counting of surface features and the presence of labels). Third, decoupling enables participants to avoid inspection, and this avoidance is a
display of trust and confidence. As such, decoupling contributes to the logic of confidence and increases the commitment of internal participants (responsibility is pushed onto teachers and teacher
professionalism). Last, a great deal of the value in
education has little to do with the efficiency of instructional activities. It is not so much about learning per expended dollar, as much as value residing
in the ceremonial enactments of schooling that are
regarded relatively equivocal: buildings, teachers,
books, topics, accreditation, classrooms, desks,
and so on are all valued. By decoupling formal
structures and categories from core practices and
activities, uncertainty about the effectiveness of ritual categories is reduced (Ibid:206).
Why does loose coupling arise in the US education system? And what other systems might it
occur in? Decentralization tends to co-occur with
decoupling. The US education system is decentral-

ized and relies on resources from local populations

(e.g., school boards, counties, mayors, etc.). In
contrast, other countries have a centralized structure with examinations and a clear inspection system that ensures conformity in activity. In the US,
exams are privatized and not universal. A national
system would define almost all of the kids from
some communities as successful or as failures.
This is dangerous for a system that depends on legitimating itself in and obtaining resources from
local populations (Ibid:205). Local control deprofessionalizes administrators but professionalizes the teachers. Hence, American schools have
weak administrators who struggle to drive through
educational reforms.
In sum, neoinstitutional theory argues that
organizations succeed in the environment by engaging in symbolic coding, or the adoption of rationalized myths about structures that rely on a
logic of confidence. Then they decouple their formal structure from the actual internal activities and
performance. This affords them greater flexibility
and buffers the technical core and internal workings of instruction from the likely conflicted concerns of the external environment. Decoupling
and the logic of confidence enable managers and
employees to do their work without close inspection.
Then why adopt the formal rules and structures when observation or inspection are not all
that relevant? Is the adherence to rational myths
helpful in some way? Organizations need legitimacy in their environment to survive.
Independent of material needs, organizations need to
look like a real organization and at least appear to
behave like a real organization. The creation of
and adherence to prevailing rational myths provides organizations with many resources.
Again, lets take the case of schools to flesh
this out: Credentials, classifications and categories
of schooling constitute a language that facilitates
exchange across organizations and with the environment. Funds are frequently allocated in a categorical fashion e.g., vocational education, special education, elementary school, high school, etc.
Having them in place makes such transfer feasible.
The system of ritual classifications can be ex-

152

Figure - Organizational Field

(Source - http://commons.wikimedia.org/wiki/File:
4/4eAGC_key_technology_alliances.JPGAGC_key_technology_alliances.JPG)

ploited in order to gain prestige (e.g., you can hire

prestigious faculty, incorporate innovative programs and then see your ranking and resources increase). Organizations rely on the ritual classifications to provide internal order. Actors derive identities from the socially derived categories of education.
In many ways, this is a managerial proscription. If you attempt the above, you will conform
to the institutional environment and reap rewards
from it. By incorporating externally defined teachers, curricula, and students into their formal structure, schools stay legitimate and get the necessary
funds and participant involvement so they can operate. In short, the rewards for adherence are the
increased ability to mobilize social resources for
organizational purposes.
Organiza(onal+Fields+and+Isomorphism

The second theory paper I want to discuss is

that of Dimaggio and Powells Iron Cage Revisited (1983). What is great about this article is

that it shows how neoinstitutional

theory relates to both resource dependence theory
and population ecology (which is a final theory we
will cover in chapter 10). In addition, the articles
describe a variety of bridging tactics leading organizations to resemble one another in form.
Dimagio and Powells big question is why
do so many organizations look the same? Why is
there a progression from a diverse set of organizational forms to a homogeneous set? Dimaggio and
Powell focus on organizational fields and how organizations within them grow isomorphic. Notably, this is the same question Metzs case on real
schools asks -- how come schools look so similar?
Lets first define the concept of an organizational field since it describes the bin in which the
process of organizational homogenization arises.
Organizational fields are composed of organizations that, in the aggregate, constitute a recognized
area of institutional life; [for example] key suppliers, resource and product consumers, regulatory
agencies, and other organizations that produce
similar services or products (Dimaggio and Powell
1983:148). Here is an example of what might be
regarded to be an institutional field of technology.

153

How does a field like this form? The process

of field definition, or structuring, consists of 4
parts: First, there is an increase in interaction
among the members; second, greater interorganizational patterns of hierarchy and coalitions
among them; third, an increase in information load
to contend with; and fourth the development of mutual awareness among the members.
Within these fields, how does organizational
homogenization arise? The process is one where
one unit of a population comes to resemble others.
Neoinstitutional theorists call this isomorphism.
In usual parlance, isomorphism can be expressed
in various ways. Visually you can think of it as
mirroring or when buildings assume the same
form or appearance. But it also has more of a
mathematical or even geometrical expression. Notice the 1, 2, 3, 4 have the same pattern of association as 5, 6, 7, 8. Same for a-d and g-j. They are
structurally equivalent sets and substitutable.

Figure. Graphs and Isomorphism

(Source - http://commons.wikimedia.org/wiki/File:Graph_isomorphism_b.svg/;
http://commons.wikimedia.org/wiki/File:Graph_isomorphism_a.svg/)

For neoinstitutional theory, these appearances can

decouple from function, so perhaps the next image
helps even more. It is of an orchid whose flower
mimics a bee. By showing appearances of one
sort, it attracts resources pollen. Now, none of
these are perfect similes, but hopefully they give
you a better sense of what is meant by isomorphism.

Figure. Orchid and Mimetic Isomophism

(Source http://commons.wikimedia.org/wiki/File:Bee_Orchid_%28Ophrys_apifera%29_-_geograp
h.org.uk_-_1362149.jpg)

Dimaggio and Powell describe multiple processes by which isomorphism arises. The first process is one they call competitive isomorphism. In
these instances, certain forms of organizing do not
survive because they are sub-optimal and because
decision makers learn appropriate responses and
adapt their organizations so they survive. Dimaggio and Powell suggest this occurs in fields where
open competition exists. We will discuss this form
more next week when we discuss population ecology as our final organizational theory in the
course.
The second form is institutional isomorphism. This is the core process within neoinstitutional theory: here organizations do not just compete for resources and customers but for political
power and institutional legitimacy. The concept of
institutional isomorphism is useful for understanding the politics and ceremony that pervade modern
organizational life (Ibid:150).
Powell and Dimaggio describe three forms
of institutional isomorphism:
1. Coercive forms of institutional isomorphism
most closely resemble those observed in dependence relations as discussed in resource
dependence theory. Here a dependent firm is
subject to political influence. This coercive
influence results from both informal and formal pressures exerted by other organizations
upon which a focal organization is dependent, and by societal cultural expectations

154

within which the organization functions

(Ibid:150). The firm is coerced to conform
and this leads them to follow and adopt organizational forms of the organizations they
depend upon.
2. Mimetic institutional isomorphism is different from coercive isomorphism. Mimetic isomorphism is a standard response to uncertainty and ambiguity.
When ambiguity
arises, organizations model themselves on
other organizations, and particularly those
perceived as legitimate / successful. Hence,
the mimetic process here is one driven by the
focal organization and their effort to secure
resources. Hence, many universities do as
Stanford does, not from certainty about efficiency, but because in a context of ambiguity
and uncertainty, they can rise through the
ranks by looking more like a leading institution.
3. Normative forms are different yet again from
both coercion and mimesis. Here, isomorphism is associated with professionalism.
Professionalism is defined as the collective
struggle of members of an occupation to define the conditions and methods of their
work, to control the production of producers,
and to establish a cognitive base and legitimation for their occupational autonomy
(Ibid:152). Rather than direct coercion or
imitation, the firms in these instances try to
fit in and mirror professional norms from
which they draw legitimacy.
Two aspects are key here: the emphasis on
formal education credentials, and the development
of professional networks via associations. These
create pools of individuals who are relatively the
same and substitutable. As such, professionalization enables normative forms of isomorphism and
renders firms relatively similar in terms of who
they hire, what tools they use, and so on.
The theoretical features of neo-institutional
theory can be summed up to this point. The theory
argues that firms buffer themselves from the envi-

ronment by symbolic coding of their formal structure. As such, new universities adopt many of the
same subjects and departments that established universities have. In this manner, their formal structure fits ceremonial classifications, and the constructs are supported by a logic of confidence that
extends throughout society. The labels are assumed rational because rationalizing agents support them: e.g., confidence in elite universities and
their accreditation. Further buffering the core activities of the firms is the process of loose coupling, where the formal structures and codings of
the firm are distinct and unrelated to the actual
work activity. By segmenting them apart, the firm
exudes rational competence and cultural fit, but
does not allow them to be inspected with relation
to actual activity. This decoupling enables the
firm to run on trust and not have to confront the
potentially unsolvable issues of what works best
and why.
Firms also bridge in the environment, but here
it is mostly done through networks. Dimaggio and
Powell argue that these networks of association
lead to isomorphism via several routes. The first
entails political pressure as we learned about in resource dependence theory. The second entails mimetic behavior where firms look to exemplars and
peers so as to imitate what seems to work well or
is legitimate (i.e., trendy). Last, firms respond to
pressures of professional networks, like professional norms and standards on how to assess and
consider their firms performance. All of these
bridging efforts render the firm more institutionalized and legitimate in the cultural environment in
which it is found. And this in turn draws in social
resources and continues the firms survival.
Management and Critique of the Neoinstitutional Approach
In the remainder of this chapter, I will focus
more on management, discuss some cases of strategic manipulation of institutional environments,
and then critique the neoinstitutional approach.

155

Management of Cognitive Structures

Given the above, how do we manage a firm

using neoinstitutional theory? Oliver describes a
series of strategic responses organizations can take
to their institutional environments (1991): such as
acquiescing, compromising, avoiding, defying, and
manipulating. Implicit in the approach is some
sense of how integrated, differentiated and ambiguous the surrounding institutional environment is
that affords some sense as to which strategic response may be most successful. Acquiescing is
the most common one described in the literature of
neoinstitutional theory.
There, a firm merely
adopts and aligns with the institutional environment as if it is natural to do so.
This makes a lot of sense if there is consensus in the environment, or if the firm appeals to
certain niche institutions and beliefs. Compromising is different and entails balancing differentiated
demands and negotiating with institutional representatives. This typically occurs in conflicted and
differentiated environments where one must play
one perspective off another. Avoidance can be accomplished via buffering strategies like loose coupling since it prevents careful inspection. In a
way, this move is akin to disguising the firm and
using smoke and mirror tactics to distract. Decoupling is used when institutional rules conflict with
technical requirements (ritual features are not appropriate for outcomes), or when the institutional
environments are themselves in conflict. In these
instances, decoupling helps a firm an open system firm avoid confronting internal or external
inconsistencies. By contrast, coupling and alignment across ritual classifications and technical output can occur when organizations are centralized
and rewarded for technical performance. Or when
the institutional environment is focused on certain
issues and dependence is highest there. So for example, the use of accountability in schooling now,
creates a more focused coupling.
Oliver also mentions how firms can defy or
resist their institutional environments by adopting
norms and interests different from the surrounding
environment and the imposing regulations on it.
In most cases, the firms doing this lose. Last, or-

ganizations can co-opt and manipulate institutional environments in an effort to improve bargaining power. This is often done by developing
symbolic linkages with sources of power so
many of the bridging efforts of isomorphism apply
here of coercion, mimesis, and norming.
In sum, the manager must find ways to align
the institutional environment, or to find ways to
help the organization wind its way through conflicting institutions in the environment. To do this,
they often conform and adjust their ritual classifications and outward appearances, and they buffer
their technical practice via decoupling. Or they
manipulate the situation by playing to the myths in
the environment. Therefore, managers can hire
planners and economists to waste time ratifying
plans already made; or hire human relations professionals to deflect blame from conflicts; etc. I
think we can go further in this regard and discuss
how marketing and advertising are used to receive
endorsements and support from the environment.
Later, I want to turn next to framing and framing
wars as a case for this.
Case Interlude: Framing Wars
Framing better captures strategic aspects of
cultural mirroring and fit since it is all about cultural alignment efforts. Recently, there were a series of scholarly and media articles concerning the
framing wars in politics (Lakoff 2011; Bai 2005)
and debates about intelligent design (Wilgoren
2005; Anonymous 2005). Both framing wars describe how organizations and their leaders manipulate narratives and meanings so as to better align
with the national consciousness or even segments
of the environment. The beauty of framing is that
it captures both aspects of this tension strategy
and cognition.
Back in 2004, we had a presidential election
between John Kerry and George Bush. They both
had stances on a variety of issues, but wanted to
legitimate themselves with voters and secure the
popular election. Bush won and linguistic experts
like George Lakoff argue that he won because his
strategists framed positions in a way that resonated
more with the voters (Lakoff 2011; Bai 2005). Or

156

Figure. Framing Wars in Politics and Religion (Kerry, Lakoff, Bush)" "
(Source - http://commons.wikimedia.org/wiki/File:John_Kerry_headshot_with_US_flag.jpg/; http://commons.wikimedia.org/wiki/File:Pop%21Tech_2008__George_Lakoff.jpg/;http://commons.wikimedia.org/wiki/File:GeorgeWBush.jpg/)

as we might say, the Republicans framed their positions in a way that resonated with deeply ingrained
rational myths. Lakoff makes it clear that the issue here is not about finding the right words
these conceptions or grand metaphors do not
just suddenly emerge, and the right words will not
suddenly change the national taken-for-granted understandings of the world around us. They must
resonate with beliefs and arguments many of us
hold.
As such Bush related he was against partialbirth abortion (not intact dilation extraction of a
fetus). Note how the label highlights the contradiction between birth and abortion. Bush also said
he was for exploring for energy, not oil drilling and fracking which are seen as potentially
damaging. He was also for tax relief, not tax
cuts for the rich which is seen as unfair. So by
reframing their policies in ways that make them
appeal to and resonate with deeply held beliefs
even if inaccurate -- seems to work. Now the
Democrats have gotten wise to this and have their
own set of framings like tax cuts for the rich,
Wallstreet Bailouts, etc.

In the United States we also have seemingly

perennial debates about what to teach in our
schools. Given the nation is partially formed on
puritanical immigrants, there are subsections of
the population that are fundamentalist Christians
and they sometimes take a literal view of the Bible
and regard evolutionary theory as lacking and an
affront to their beliefs. They would much rather
see Creationism be taught in schools and for students to learn that the Bible says the universe is
3000 years old.
What is interesting about this controversy is
how framing plays a part. Here again, religious
conservatives have successfully found ways to
frame their arguments so they resonate more fully
in the environment. In fact, the most common refrain in the debates about intelligent design is
teach the controversy. Given education and science rest on teaching and exploring evidence
through debates, the pressure to teach the controversy seems appealing on its surface. It is just
that evolutionary scholars, biologists and most educated people think there is no controversy to begin with. So the use of wording to resonate with
157

commonly held rational myths serves to undermine the efforts of rational agents like professors,
universities, natural science fields and medical professionals. What I am trying to suggest is that we
can use rhetoric to manipulate opinions and to secure social resources from the environment we
just need to find interesting ways to appeal to rational myths!

ics mean that neoinstitutional theory mostly identifies weakness in other theories instead of revealing
direct evidence of its own claims. However, this is
easier in theory than in practice.
Demonstrating the diffusion of cognitive
scripts and conceptual frames (grand metaphors) is
much easier to do ex post and through proxies than
through ex ante prediction or the direct measurement of institutional variables. The theory has intuitive appeal and we can identify cases where diffusion and isomorphism occur, but it is hard to distinguish processes of normative and mimetic isomorphism and to identify the features being homogenized. Neoinstitutional theory is one of the
most vibrant theories of organization, however,
and many scholars are working hard at developing
it further empirically.
Case Challenges for Neoinstitutionalism?
There are certain trends in educational institutions that seem to counter neoinstitutional arguments. So lets consider them for a moment.

Figure. Symbols for Evolution and Creationism

(Source - http://commons.wikimedia.org/wiki/File:Darwin_fish_ROF.svg/;
http://commons.wikimedia.org/wiki/File:Ichthus2.svg/)

Critique+of+Neoins(tu(onal+Theory

Neoinstitutional theory, like all theories, is

not perfect and it is prone to critique. Most readers regard neoinstitutional theory as describing
over-socialized and passive human actors. As
such, power and parochial interests get slighted by
neoinstitutional theory. What matters is the external environment and mirroring rational myths.
Many argue that neoinstitutional theory has gone
too far in the direction of cognition and shared understandings of modernity, thereby trivializing politics and power. The framing literature provides a
potential means forward and around these critiques
but it is currently empirically underdeveloped.
Some criticize the neoinstitutional approach as affording mostly negative evidence. By this, crit-

No Child Left Behind - Recoupling

An interesting paradox in education right
now involves No Child Left Behind (NCLB) and
similar educational policies that strongly rely on
inspection and accountability. On the one hand,
NCLB increases inspection so much that it challenges the notion of loose coupling in education.
It forces teachers to conform, and it places greater
power and responsibility in administrators. At the
same time, the reform creates greater pressure for
conformity to a particular myth of the Real
School and the appropriate elements a school
should reflect. The ritual categories are more focused, innovation in methodology is minimized,
and conformity is imposed by a testing regime.
Hence, test scores developed by rationalizing
agents like testing services and academics become
standard-bearers. The test scores indicate whether
one school is successful and another is not. Teachers then teach to the test, and in some cases even
cheat to sustain appearances of adhering to the rationalized myth. But is there a clear sense of what

158

works better? What kind of learning is more efficient and desirable? Yes there is a sense of this
but it is conveyed through the lens of a test.
Interestingly, the Bryk article discussed very
early in this quarter could be interpreted as strategic management based on recoupling. Changes in
the regulatory environment led to strategic responses like those described by Bryk and we saw
the system conform, but it was not clear it had become more efficient and successful than before.
Moreover, teachers felt deprofesionalized and their
motivation began to wane. So it is not clear
whether recoupling and centralization render organizations any less reliant on rationalized myths.
They just seem to focus them on certain standard
bearers more than others. Ambiguity and uncertainty remain and only one facet of the institutional environment (the currently dominant one) is
linked.
Massive Open Online Courses - MOOCs
Another conundrum for neoinstitutional theory is the creation of massive open online courses
(or MOOCs) and what they mean for the organization of universities. Could MOOCs threaten the
rationalized myths upon which the modern university is constructed? Does Coursera challenge the
common script and neoinstitutional conceptions of
organizational fields like education? How? Why
might MOOCs lack environmental legitimacy?
How do MOOCs challenge myths of schooling
and question the legitimacy of higher education institutions?
What will MOOCs do to community colleges, University of Phoenix, and to actual classroom experiences? What if superstar teachers can
effectively convey material and students learn it
almost as well as in person, but at a fraction of the
cost? What if Coursera and other platforms offer a
degree and it is just as effective and valued as an
actual credentialed, university degree?
What
might happen?
The whole societal apparatus
seeks credentials as standardized language by
which exchanges can be made across institutions.
With Coursera, success is not a scarce commodity!
People who cannot get into Stanford can potentially get the same credential and accomplish the

same course. Does it mean the credential is illegitimate? The market would be flooded with people
who have the same skills since Coursera has room
for far more people. As a result, they would give
employers little room to hire some and not others.
Who would do the janitorial work if everyone is
overqualified? Who would stand out as being able
to do complex tasks? What if the credential is associated with tons of variance? Is it less legitimate? MOOCs raise a lot of questions about legitimacy for one of our most central societal institutions and the rationalized myths it rests upon.
There has been much written about MOOCs
online, and there is much to be recommended.
However, one recent article seems to do a little
more research than most, and I can recommend it
to students in this course (Grossman 2013).
Rather than delve into this topic too deeply, I
would rather leave it to the forums of the class to
debate MOOCs and what they represent for neoinstitional understandings of organizatons.

Summary New Institutional Theory

The basic question neoinstitutional theory

asks is why do organizations within a field adopt
the same (or similar) formal structures? This similarity arises because organizations want to be legitimate in their environment, and to do this they
structure themselves to reflect prevailing rational
myths on what that type of organization should
look like. Because of this, organizations within a
domain often come to resemble one another in appearance, and in spite of their being distinct in
terms of performance and actual activity. While
many courses are labeled algebra class, we
know the actual instruction, content covered, and
learning that occurs across them can radically differ.
So what gives? Organizations must be legitimate in their environment in order to receive a variety of social resources. The formal structure of an
organization incorporates an environmental theory
of the organizations activity. This theory must
give the appearance of being rational and func-

159

tional. The prevailing environmental theories and

categories are taken-for-granted understandings of
organizing: Organizational actors must therefore
take into account both what they are doing, and
the appearances of what they are doing (Meyer
and Rowan 1978:109). In order to accommodate
appearance and reality, appearances are decoupled
from actual activity. We must find planners and
economists to waste their time ratifying plans already made; they must hire human relations professional to deflect blame from conflicts; etc, etc.

160

161

Actors seek expression and fulfillment of identity,

and organizational culture is the medium for such
expression/sense-making.

Summary or
Basic
Argument

Goal is organizational survival through external

adaptation (for certainty and autonomy).
Formal roles, standard operating procedures, interorganizational bargaining / politics.
(note: coalition approach emphasizes individuals and
interests. Here, the organization is the main actor and
exchanges are with other organizations.)

Create intrinsic motivation (sense of fulfillment),

and remove differentiation / cynicism in most cases.

Deep structure composes the elements of culture

themes (beliefs & norms), their expression via
practices (rituals, etc), and their manifestation or
expression in artifacts (reports, mission statements,
etc).

Many elements of culture have origins from outside,

and they are transported in, then translated to the
local culture.

Action = result of deep structure or culture that is

generated in the organization, but which is mediated
by the members relation to it.

Find ways to confer ideology and lead others to

identify with it (using a variety of practices and
artifacts), but dont make it so explicit / fanatical
that cynicism emerges. Give room for autonomy
and self-expression so distancing is unnecessary,
and encourage members to generate a culture of
their own (~org learning culture NE to Tech culture
which is top-down engineered).

Environment

Dominant
Pattern of
Inference

Management
Strategies

Buffering: protecting technical core from environmental

threats (coding, stockpiling, leveling, forecasting and
adjusting scale).
Bridging: security of entire organization with relation to
the environment. Total absorption via merger (vertical,
horizontal, and diversification), partial absorption
(cooptation [vertical or horizontal], interlocks, joint
ventures, strategic alliances, associations)

Action = Scan environment for resource opportunities

and threats, attempt to strike favorable bargains so as to
minimize dependence and maximize autonomy /
certainty.

Key component of the perspective. Exchange partners

and external relations more salient than internal
dynamics;
Bridging more relevant than buffering.

Focal organization and other organizations with

resource interdependence,

Goals
(what probs to
resolve)
Social
Structure

Participants

Matching, sense-making / meaning-making, or

where actors seek to express beliefs, norms, and
values via a variety of practices and externalize
them in artifacts depicting shared understandings /
notions of appropriateness.
Actors within the organization, and those salient to
meaning-making.

External adaptations in order to increase autonomy

and/or decrease dependence (see management).

Focal organization with input/output concerns that

cannot be resolved without considering the
environment.
For the most part, organizations are considered unitary
actors (some of the struggles/internal divisions are
minimized) in order to highlight the interactions with
suppliers and clients.

Exists when there is a focal actor interested in

decreasing dependence, increasing autonomy,
increasing power, and (possibly) increasing efficiency.
Preferences and goals are unclear except in relation to
dependence.

Resource Dependence Theory (RDT)

Technology
(how solutions
get decided)

Key Organizational Elements

Exists when the cognitive and normative aspects of

social structure are of concern and seem to guide
organizational decisions (sense-making) and
outcomes.

When does it
apply?

Organizational Culture

Design network to deliver service (select partners

and alliances wisely for aligned values / goals);
establish informal, active communication
channels; coordinate member activities (group
processing skills align members culturally,
remove internal competition, create open
information, form joint governance/shared
decision making, and get them to focus on discrete
functions / coordination of actual tasks); reinforce
norms of collaboration and reciprocity.

Action = identify complementary strengths, form

alliances, establish collaborative/reciprocal norms,
create open-ended mutual benefits where possible,
outsource secondary tasks (to focus on core) all
for survival and creation of positive network
environment that delivers service.

Boundaries no longer clear. Networks apply to

within and between firm relations.

Goal is delivery of service via collaboration and

outsourcing aspects not central to technological
core..
Formal and informal roles, relations, and
communication channels. Patterns of relations
influence behaviors. Deep structure consists of
values and beliefs in sharing, communication, and
collaboration (trust).

All stakeholders in an organizational field.

Linking / coordinating /allying in order to deliver

service and outsourcing / subcontracting /
partnering in order to focus on core technology.

Organizations focus on network relations,

positions, and larger context in developing
strategy. Multiple types of networks are feasible
and they can guide resultant exchanges.

When the wider context of organizational relations

influences organizational behavior and survival.

Network Organization

Bridging: Institutional Isomorphism (external pressures

via rationalized myths) occurs in effort to acquire
legitimacy. Three forms of isomorphism are coercive,
mimetic, normative

Buffering: Symbolic coding (systematizing and

classifying); Decoupling organizational elements (loose
coupling).

Action = Organizations in a field conforming to

normative and regulative environments; the process can
be strategic and planned or cognitive and taken-forgranted.

Cultural legitimacy and resources. Legitimacy in the

environment necessary for survival.

Formal structure conforms to the environment. Often,

the technical core is radically decoupled from
institutionally defined org structure (loose coupling).
The logic of confidence makes inspection less
necessary, and practice may be very different from
ceremonial classifications or structures.

Organizational survival through alignment with the

environment.

Organizations in a field, professionals, and the nationstate.

External adaptations in order to fit the environment and

insure survival. Professionals provide expertise and
consult to organizations.

Organizations in a field conform to cultural norms to

insure survival and to reduce ambiguity. Legitimacy is
a key resource and legitimacy can come at the
expense of organizational efficiency. Professionals and
the nation-state carry the modern cultural recipes and
influence the translation of these elements into the org
context.

Exists when the level of analysis is a field (not a focal

actor) and the focus is on conformity to cultural scripts
and/or normative constraints on action. Unlike
organizational culture, social structure is based at least
as much on external environment as on internal
dynamics.

Neoinstitutional Theory

Summary Table of Organizational Culture, Resource Dependence Theory (RDT), Network Organization, and Neoinstitutional Theory (NIT)

References
Anonymous. Intelligent Design Rears its Head.
The Economist, July 28, 2005.
Bai, Matt. The Framing Wars. New York Times
Magazine, July 17, 2005 (pp. 1-8).
Davis, Gerald F. and Walter W. Powell. 1992. A
selection from Organization-Environment Relations (pp. 342, 354-365). In Handbook of Industrial and Organizational Psychology, Vol 3 (2nd
ed.). Eds. Marvin D. Dunnette and Leaetta M
Hough. Palo Alto, CA: Consulting Psychologists.

Scott, W. Richard, 1995. Institutions and Organizations. Sage.

Scott, Richard. 2003 (5th ed). Institutional Theory (pp. 119-120) and Managing Institutional Environments (pp. 213-220) of Organizations: Rational, Natural and Open Systems, 5th Edition,
Englewood Cliffs, NJ: Prentice-Hall.
Wilgoren, Jodi. 2005. Politicized Scholars Put
Evolution on the Defensive. New York Times,
August 21, 2005.

DiMaggio, P. & W. Powell. 1983. "The Iron Cage

Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields." American Sociological Review 48:147-160.
Grossman, Robert J. 2013. Are Massive Open Online Courses in Your Future? HR Magazine, pp
30-36. August, 2013.
Lakoff, George. 2011. Don't Think of an Elephant!
Know Your Values and Frame the Debate. Chelsea
Green Publishing.
Metz, Mary Haywood. 1989. Real School: A Universal Drama Amid Disparate Experience. Politics of Education Association Yearbook
1989:75-91.
Meyer, John and Brian Rowan. 1977. "Institutionalized Organizations: Formal Structure as Myth
and Ceremony." American Journal of Sociology
83:340-363
Meyer, John W. and Brian Rowan. [1978] 2004.
The Structure of Educational Organizations. Pp.
201-212 in Schools and Society: A Sociological
Approach to Education. Eds. Jeanne Ballantine
and Joan Spade. Canada: Wadsworth.
Oliver, Christine. 1991. Strategic Responses to
Institutional Processes. The Academy of Management Review, 16(1):145-179.

162

10
Organizational Ecology

Source: http://commons.wikimedia.org/wiki/File:Selection.svg

Ecological Conceptions of Organizations

And then there are the main theorists we will focus on this week: Mike Hannan, John Freeman
and Glenn Carroll (Hannan and Freeman 1977,
1989; Carroll 1981, 1984). They take this metaphor and understanding of organizational populations to a new level in their construction of population ecology theory.

Population ecology focuses on organizational

change and explains it as the result of environmental forces acting on populations of organizations. It argues that social, economic, and political
conditions affect the relative abundance and diversity of organizations and accounts for their changing composition over time (Hannan and Freeman,
1977). Population ecology is a theory about Darwinian selection in populations of organizations
(Carroll and Hannan, 1995).
Now that we have a general sense for population ecology, lets look more carefully at its core
concepts. A core concept of organizational ecology concerns the definition of a population. If you
recall, neoinstitutional theory had an elaborate definition of organizational fields, and population ecologys notion of a population is similar, albeit with
less emphasis on self-awareness, and more emphasis on regional boundaries and competition.
A population of organizations is composed of
a class of organizations facing similar environmental vulnerabilities and sharing the same internal form (technical core). This shared internal
form is a consistent blueprint for action or pattern of activity. As such, it is akin to how Nelson
and Winter view SOPs and tasks. And shared environmental vulnerabilities refer to external sets
of relations and dependencies an organization has
in the environment. Last, the population is
bounded within a common system, whether by a
geographical (region), political (nation), or economic (market) boundary. Examples of a population could be financial institutions in Seattle, or
car dealerships in Houston, or now with the digital
era, an entire industry like the beer industry and
the niche of micro-brewing.

Population Ecology

Environmental Niche

Population ecology begins with several questions: Why are there so many kinds of organizations? What explains the diversity of organizations? Where do different organizational forms
come from? Notice these questions are the inverse
of neo-institutional theory, which asks, why are organizations so similar and stable?

Thus far, we have defined populations of organizations. Population ecology contends that the
environment can be partitioned into different kinds
of resource spaces where distinct populations of
firms can persist. They call these environmental
niches. Organizational ecologists describe two
types of environmental niches: fundamental and

In this chapter we continue our study of organizations as open systems whose survival and
success depends on their reaction to the environment. We introduce a 10th and final theory called
Population Ecology. There is a long history of
work that applies biological and natural selection
metaphors to organizations (Scott 2003:117; Davis
and Powell 1992:342-354), let alone to the study
of society.
Karl Weick (1979) is an organizations scholar
who described variation, selection, and retention
processes within human organizations and on an
organizational community level.
Arthur Stinchcombe talks of firm founding and
retention in epochs (1965).
Richard Nelson and Stanley Winter (1982) offer
an evolutionary account of how firms and industries change over time. They regard organizations as strings of SOPSs, which they view as
the genetic makeup of an organization, and then
firms experience random mutation and recombination in their tasks that lead some to outcompete and survive (i.e., selection and retention).

164

Bear Realized Niche!

Bear Fundamental Niche!

Figure. Example of
Fundamental and
Realized Niches ""
(Source - http://commons.wikimedia.org/wiki/
File:NAMAP.jpg)

realized. In biology, a fundamental niche is where a species of animal is

able to live and survive, and the realized niche is where the organism actually lives. For example, an animal might be able to live within the entire
forest, but because of human encroachment and noise, it might only live
in a small area of the forest. As such, the entire forest is a fundamental
niche and the realized niche is the small part of the forest the animal actually lives. With that in mind, lets consider a simple example of bears in
the United States. The fundamental niche for bears is very broad. But
most of their realized niche is in the Northern United States.
For organizations, a fundamental niche refers to a region of the resource space in which an organization can persist in the absence of competition. Examples of this might be entertainment, health, education, or
beverage industries. The realized niche is the subset of the fundamental
niche in which an organization can sustain itself in the presence of given
competitors. Examples of a realized niche might be music, dance, and
movies within the entertainment industry, or beer, wine, and soda in the
beverage industry.
The realized niche-width refers to the resource space a species of
firm gets that is not used by another species of firms. Hence, in the beer
industry, micro-brewing companies may find themselves in a partitioned
resource space where they can survive in spite of huge brewers like Budweiser (Carroll and Swaminathan 2000). Likewise, in education, there
may be a city where there are private schools, public charters, public mag-

165

nets, and traditional public schools. Each city may

afford a resource space in which only so many of
each type of school can survive before it competes
with others.
Process of Ecological Change
Environments are constantly changing, and

new organizations emerge to meet these

changes. As such, a healthy population has some
diversity (note how this relates to the concept of
variation in evolutionary theory). However, societies have limited carrying capacities for organizations. Therefore, in equilibrium, the surviving
populations of firms occupy a niche wherein organizations are isomorphic and fit the environment
(note how this relates to the concepts of selection
and retention in evolutionary theory). Firms that
deviate in form are eliminated as unfit. So we
have three features by which populations of organizations change variation, selection and retention
- and they all reflect concepts in evolutionary theory (e.g., speciation and natural selection).
According to population ecology, new organizational forms emerge all the time to cope with
perceived needs in the environment. Organizational variation is due to mutation (random genechange or accidental new ideas), recombination of
existing forms (mix and match old ideas together),
and cross-over of forms, In biology, a cross-over
case arises in chromosomes, but for organizations
it occurs when one idea is taken from one domain
and imported to another of say a biology metaphor of evolution being applied and extended for
use in conceptions of organizations.
In animals we see species vary within niches.
For example, there is variation in biodiversity. Below is an image of fungi from Saskatchewan, but
you can imagine the same for butterflies, birds, rodents, etc. In the same way, one can also imagine
variation in financial firms.

Figure. Variation in Fungi

(Source - http://commons.wikimedia.org/wiki/File:Fungi_of_Saskatchewan.JPG

Now some organizational forms suit the environment more than others and they survive. The
survivors are reliable and accountable (or favored
by selection). Population ecologists typically observe the selection of an organizational form as the
reproduction rate of an organizational form. Notably, organizational variation and selection does not
have to be optimal (e.g., the best mutation takes
off) nor Lamarckian (i.e., traits passed down from
predecessor organizations that enabled adaptation). The fit is more like that of satisficing described in chapter 2. An organizational form is selected, survives, ad spreads if its form fits (among
a variety that would work) and takes off.
Ultimately, some organizational forms are selected, reproduced and institutionalized as relatively permanent (e.g., Governments, schools,
franchises, etc). In biology we see these as animals that have a strong fit and reproduction rate
like Mallard Ducks, Starlings, and so on. Those
are birds that proliferate across the world.
Organizational ecologists identify retention
through a focus on the rates of organizational
founding and death. In the case of organizations,
we can observe this process of variation, selection
and retention in action when we consider the retail
industry in the United States. The environment
has greatly changed over the last 100 years. In the
1940s and 50s stores like Woolworth were com-

166

Retention of
organizational forms
-- From Woolworth to
Wal-Mart to Amazon
(Source - http://commons.wikimedia.org/wiki/
File:Woolworth-kassel.JPG; http://
commons.wikimedia.org/wiki/
File:Walmart_exterior.jpg; http://
commons.wikimedia.org/wiki/File:Amazon.comLogo.svg)

mon in the United States. These eventually gave way to another retail
firm, Walmart, and that spread and became common in the 1980s. Today, the environment continues to change, and a retailer like Amazon.com
has taken hold. In each era, the fundamental niche remains, but a new
form of organization rises, out-competes the others, gets selected, proliferates, and is retained.
Structural Inertia
So to this point, we have discussed populations, environmental
niches and processes of ecological change. Another key concept in organizational ecology concerns structural inertia (Hannan and Freeman 1977,
1977a; Hannan and Carroll 1995). Contrary to contingency theorists and
natural system views related earlier in the quarter, organizational ecologists contend organizations are inert, and at best slow to adapt and
change. There are a variety of pressures in place to make organizational
change difficult. Both internal and external constraints are at work: Internal constraints are things like investments in equipment, information limits, intra-organizational politics, and the institutionalization of organizational routines. All of these are sunk costs placed in internal technologies and social structures that make it hard to adapt them to new circumstances. External constraints are barriers to entry and exit, and legitimacy concerns. Inertia is often associated with organizational age since
many of these constraints build up over time, making it harder and harder
167

for a firm to adapt. Core organizational characteristics of mission, goals, forms of authority, core
technology, and market strategy are hardest to
change.
The implication of all this? Well, the greater
the inertia, the less an organization can adapt and
the more important environmental selection becomes. Instead of adapting, most firms die when
the environment changes. Therefore, the main dynamic of organizational change is the birth of new
organizations and the death of old ones. If you
want to change a niche, you need a new, better organizational form that can outcompete those firms
already present.
Population ecology studies the birth of new
organizational forms (diversification) and the
death of old ones. In this manner, it identifies the
core processes of population ecology. It asks, how
do prevailing social conditions determine what organizational forms are founded and their rate of
founding? It also asks the converse, how do social
conditions determine what organizational forms
die and their rate of death? I will cut to the conclusion a bit: in most instances, volatile times encourage the birth of new organizational forms, and this
frees resources for organizational founding.
A variety of sub-theories arise from organizational ecology to account for firm births and
deaths. One common theory in organizational ecology is called density dependence (Hannan and
Freeman 1989). The theory of density dependence argues that there is a curvilinear function
where social processes of legitimization found
firms and competition cull their numbers. In low
density or in a sparsely populated niche -, the
density dependence model predicts that the legitimization process will dominate to increase the organizational founding rate and decrease the mortality rate. At high levels of density or in a heavily
populated niche-, competition will dominate, leading to low founding rates and high mortality rates.
The inverted U curve here to my side shows
x=#foundings and y=population density. Legitimacy refers to the taken-for-grantedness of an organizational form: The more legitimate the form,
(1) the easier it is to acquire resources, and the (2)
mortality rate decreases. Competition refers to or-

ganizational forms that seek the same limited resources in a niche. When there are fewer resources to go around, competition grows intense,
so (1) the founding rate drops, (2) and the mortality rate increases. As such, competition is inversely related to population density.

Figure - Density and Founding Rate

Organizational ecology has a second theory of
firm death, called the liability of newness and it
concerns age-dependence and survival (Hannan
and Freeman 1989). The main idea of this theory
is that new organizations are most likely to fail
since their internal structure and external dependence relations are not well elaborated and established. This happens because, internally, the members must learn the new roles and relations that a
new organizational form requires, and it prevents
them from getting down to business (i.e., focusing
on input to output flows). Externally, they lack the
legitimacy and stable relations older firms have,
making it difficult for them to attract support.
Hence, with age comes a greater chance of survival. This is confounded with size, and many argue that is the deciding factor -- a demographic
trait determines organizational survival.
The liability of newness also applies to times
of crisis. In these periods, many organizations try
to change internally. When this happens the internal structure no longer reflects the firms accumulated history, and it is robbed of prior survival

168

value. Changes in the core features of an organization, like its mission and values, are more problematic and therefore help explain organizational
death something population ecology is keen to
explain. In contrast, changes in short-run strategies and peripheral features are more consistent
with adaptive perspectives of organizations.
A third theory about firm survival and death is
called niche theory (Hannan and Freeman 1989).
Here the general idea is that different environmental conditions favor specialist and generalist
organizational forms. A specialist firm is one that
focuses on a particular technology and takes the
risk of maximizing their exploitation of an environment, fully realizing it could change. A generalist
firm is one that exploits multiple environments
(niches) at lower levels so it has greater security in
the face of environmental change.
For example, in the wine-industry, mass production firms like Gallo produce wines like burgundy, chablis, claret, madeira, port, rhine, sherry
and tokay, named from geographic regions. In
many instances, they generate jug wines or lower
quality versions of these wines at a much lower
cost. Generalist production accounts for the bulk
of wine sales. In contrast, a specialist firm in the
wine industry is a farm winery that produces varietal wines, named on the basis a specific grape, and
labeled with appellation of origin (Carroll and
Swaminathan 2000).
There are several theories on how niches favor specialists and generalists. One is called niche
width theory and it was posited by Hannan and
Freeman in 1977. Niche-width theory focuses on
2 aspects of environmental variability to explain
differential survival of specialists and generalists.
In my earlier discussion of ecological processes, I
described how populations of organizations can occupy the same niche (or the same domain of
unique environmental resources) and depend upon
identical environmental resources. If two populations of organizations occupy the same niche
while differing in some organizational characteristic, the population with the less fit environmental
characteristic will be eliminated. Generalist and
specialist organizations respond to environments
differently however. Generalist organizations

have slack. They draw on different resources or

realized niches so they can survive changes in one
environment. In contrast, specialist organizations
are leaner and try to exploit resources of a single
environment, or realized niche.
Generalist firms are not optimally suited for
any single situation. As such, a wide fundamental
niche with many realized niches favors the generalist. A firm that spans 2 or more different parts of
the environment (i.e., a generalist) will be able to
respond to the environment regardless of what the
environment is doing. However, it incurs a cost
for covering a wider portion of the environment.
Therefore, generalist firms benefit from wide
niches and unstable markets because they have diversified their efforts and can handle volatility.
Specialist firms are suited to a narrow niche.
A narrow niche has resources suited to a small
range of products and therefore favors specialists.
Specialists succeed in narrow niches and stable
markets. In these contexts, they can exploit their
fit with a realized niche and ignore other niches.
In an environment that is stable, generalists cannot
compete with specialists because specialists maximize their share of the market and do not have to
pay an overhead cost. They can better fit special
interests, while generalists are ready to address
changes in environmental interests should one
niche grow in size over another.
A second type of niche-width theory calls into
question whether generalist or specialist firms succeed more in unstable environments. But the theory really has to do with the level of change and
variation one focuses upon. As such, niche-effects
on specialist and generalist survival depend on
whether one regards environmental variation as
fine or coarse-grained.
On the surface, it makes sense that stable environments encourage specialist organizations, and
that unstable environments lead to the mushrooming of generalist organizations. But Hannan and
Freeman argue this is not true for all cases.
Whether generalists or specialists are favored by
environmental change, is determined by a combination of the distance between two kinds of firms
(how specialized / general a firm currently is), and
the grain of environmental variations.

169

Figure. Topology of Normal and Abnormal Market

(Source - http://commons.wikimedia.org/wiki/File:Topology_of_US_market_before_and_after_transition.jpg )

Fine-grained variation: Fine-grained variations

have a typically short duration, compared to the
lifetime of the organization.
Coarse-grained variation: Coarse-grained variations refer to long-term changes (e.g., a change
in the political structure of the country, or the
passing of certain laws).
When environmental changes are rapid and
fine-grained, it is better to become a specialist.
When environmental changes are rapid and
coarse-grained, it is preferred to be a generalist because specialists may not survive long enough if
they incrementally shift to the optimal state. And
here is where the distance between types plays in
if you are a specialist firm that is far from being a
generalist, then fine-grained variation probably
will not get you to survive in an environment undergoing rapid coarse-grained change.

Lets consider a brief example. In developing

these lectures, I try hard to find you visuals that
can demonstrate the core ideas and concepts, but I
am always limited to what I can find on the creative commons. In searching around I found some
images where analysts tried to identify structural
changes in the stock market that reveal where it
was susceptible to systematic collapse. What was
interesting was that they found stable, normal market behavior to entail strong residual correlations
and the stocks were correlated in a segmented fashion. Two figures show this. The first is of a clustered network diagram, representing stocks that
are highly correlated. Notably in times of normal
market behavior, there is differentiation and. The
second figure illustrates change in correlations
across stocks, and here again, you see a switch
from segmentation to an undifferentiated set of correlations. In effect, the normal market is like a realized niche space, or narrow niches where specialists can win.

170

Figure. Stock
Correlations During
Normal and
Abnormal Market
Phases ""
"
"
(source - http://commons.wikimedia.org/wiki/
File:Dynamics_of_US_stock_market_correlations.
jpg)

In the case of an abnormal market, we see the stocks suddenly have

very strong correlations across the board, and they lose their correlation
with residuals. This means there is a great deal of volatility and interdependence. In this instance, we have an environment where realized
niches collide and firms compete for resources and here the specialists
will die off if the niche does not fit them and generalists will survive to
see another day. Now the details in this are not so important. What I
want you to get is an image, or a conceptual representation you can hold
onto so as to understand the theory, that is all.
A third type of niche-theory is called resource partitioning theory,
advanced by Carroll (1985). This theory describes niche-width dynamics
to explain the differential survival capabilities of specialists and generalists. Prior formulations argue for fitness to a set, and predicts that for a
given population one optimal strategy exists. In contrast, Carroll proposes that competition among large generalist organizations to occupy the

171

center of the market frees resources at the periphery that can be used by small specialist firms without engaging in direct competition with the generalists. Swaminthan has a paper on how this occurs
in wine (2001) and Carroll describes how it occurs
for beer (Carroll and Swaminathan 2000). As you
can imagine in both industries there are generalist
and specialist firms: for wine Robert Mondavi is a
generalist, and then farm wineries like the one up
the road here called Page Mill is a specialist.
And for beer, Anheiser Busch is a generalist and
our local beers, Anchor-Steam, Red Seal Ale, etc.,
are specialist micro-brews focused on varietals.

nisms is a process of customization. The argument

here is that small firms are more flexible and can
customize their products to particular consumer
tastes. In contrast, large firms are slow and unable
to adapt quickly to changing tastes. But this creates a second issue of identity. Who you are matters. That is, even though major brewers can copy
the technical aspect of microbreweries, they would
not be as successful because they are not independently owned businesses.
In this case, it seems to matter greatly who
you are and not what you can do. Consumers purchase identities and seek customized products.
Swaminthan and Carroll conjecture why (2000,
2001):
Consumers put great faith in small producers to
make quality products.
Consumers might be reacting to mass society
and its production techniques.
Consumers may be purchasing as a form of selfexpression.
They may see their purchase as a forum for
status generation and expression.

Figure. Generalist and Specialist Beers

(Source - http://commons.wikimedia.org/wiki/File:Budweiser2.jpg
http://commons.wikimedia.org/wiki/2/21/SFBayAreaMicrobeers.jpg)

Carroll and Swaminthan both find that a

crowding of generalists in the market creates opportunities for specialists. So Budweiser, Coors,
Heineken, Miller, etc, are all generalist beer companies that compete, and they create room for
micro-brews to emerge and survive. The same occurs for wine. You have Robert Mondavi, Gallo,
Sutter Home, Woodbridge, and all the other mass
produced wineries competing and opening up farm
winery production. Both papers find a growth in
these specialist firms over the last 2 decades.
Carroll and Swaminthan argue that the competition between generalists and opening up of market space for specialists is a process of resourcepartitioning. Distinct mechanisms give rise to the
specialists in these circumstances. The first mecha-

So with a glut of general brews and wines, there

arises specialty markets for quality. And these
market-segments seem to thrive on customization
and identity.
On the surface, resource-partitioning theory
seems to counter density dependence. For example, density dependence argues that with greater
density, legitimacy becomes less salient and competition decides firm fates. In contrast, resource
partitioning argues the environment is partitioned
-- competitive culling exists for the generalists, but
in this environment, specialists can thrive and appeal to quality and identity. This occurs because,
the generalists are trying to appeal to the largest
realized niche within the fundamental niche.
Hence, recourse partitioning helps explain how
density dependence can hold for the main industry,
but then there are smaller niches wherein legitimacy mechanisms can still apply (markets at the
margin).
One question that remains unclear is whether
generalist firms can eventually figure out a micro-

172

brew and farm-wine product so as to win over customers. For example, Robert Mondavi and Miller
Beer have developed wines and beers that challenge the quality of specialist wines and beers.
Such products have won critical acclaim, and this
has enabled generalist firms to operate in both generalist and specialist segments.

tion ecology explores the relationships of organizations to their environment from a selection perspective. Compared to the others, the population ecological view is more environmentally deterministic.
Management

Summary of Theory
In sum, population ecology concerns populations of organizations and it considers environmental features that drive firm birth and death.
Lets quickly summarize the main points. Organizational ecology assumes an open system perspective, which regards the environment as inseparable
from the birth and death of an organization.
For organizational ecologists, organizations
do not catch up with environmental change (internal adaptation is uncommon [is organizational
learning is a myth?!]), and the mechanism of organizational change is organizational selection
(birth and death). Much like Stinchcombe argued
in 1965, an organizations form is pretty constant,
and consistency is the result of the founders imprint, sunk costs in internal and external relations,
and the process of selection. If two populations of
organizations occupy the same niche while differing in some organizational characteristic, the organizations likely will not adapt very fast, and the
population with the less fit environmental characteristic will be eliminated.
The diversity of organizational forms is isomorphic to the diversity of environments. In equilibrium, each environment has only one organizational form: the one that adapted to the demands of
the environment (this is the idea of a realized
niche). Each unit in the environment faces the
same environmental constraints, and hence, it has
to have the same organizational structure to survive. Isomorphism results from selection. Here,
the environment selects only those organizations
that fit the environment.
Resource dependency theory, contingency theory, and neoinstitutional theory are all about how
organizations adapt to their environment. Popula-

Now that you have a good sense of the theory,

how might a manager apply it? In many regards, I
see population ecology as a theory of forecasting.
Here you have a theory at the population or industry level, and you can use that to ascertain the environment. What potential does your firm have for
survival? Given the stability of organizational
forms, it is frequently hard to change a firm, unless it is small. So managers of organizational
ecology focus outside the firm.
The main focus is on trying to find an environmental niche where the firm can reasonably survive. This means the manager considers the population of firms they are part of. Who do they compete with? What technology do they share with
others? And what resources do they rely on?
Then they ask what is the composition of these
firms and how do they differ? Are some generalists doing many things and are others specialists
that focus on one thing? How similar or different
are they from your firm? Do they rely on the same
resources? Then one looks around and asks how
many of these firms can the niche hold? And how
fast is the environment changing? Are the changes
rapid and coarse-grained or slow and fine-grained?
All of this information will inform you as to
whether your specialist or generalist firm can survive. It should inform you about where your key
competition resides. And as you consider the competition, you may come to see the resource space
partitioned, and that some forms of competition
between other firms can create opportunities for a
small specialist.
Finally, you will consider your firms history,
identity and liability of newness, recognizing that
legitimacy may work for you in certain circumstances (e.g., resource partitioned markets, slow
stable ones with room for specialists), and out-

173

competing other firms (exploitation) in others

(e.g., fast paced, volatile markets). And in general,
you will know it may be better to start over and reinvent than to adapt.
Criticisms and Limitations of Population Ecology

As with all the theories presented, organizational ecology has certain limitations and can
be critiqued. Perhaps the most common criticism
is that the population ecology perspective is so environmentally deterministic that there is a loss of
human agency (Baum 1996; Davis and Powell
1992; Hannan and Freeman 1989). Adaptation
and decision making are after-the-fact considerations for population ecology when the reality of
managers seems different.
Another critique is that the concept of population density assumes all organizational members
are equivalent. In terms of measuring competition, for example, not all organizations are equally
competitive. Some firms are huge and others are
small. Therefore the number is questionable in the
instantiation of competition.
In terms of density dependence, the term density refers to an absolute number. This could be
misleading since populations vary in size, so a density of 100 firms in a small population could be a
large number, while in a huge population 100
could be low. What might be more useful is a relative, normalized notion of density. An environment of that size is more dense than usual, may be
a better comparison.
Population ecology also neglects the role of
globalization and technology in linking different
populations. A variety of complications arise in
terms of niche definition with the advent of the
internet and telecommunications. How can the
size of a population or resource space be determined in this case? Is it fair to call Seattle financial firms a niche? Also, what happens when a
firm is a generalist in the local market but they are
a specialist in the global market because of different cultural definitions and tastes?

Case: Linda Renzulli and Charter Schools

So now that we have described the theory and
highlighted some potential trouble-spots, lets consider an application. Renzullis work does an actual empirical study of charter schools and when
they are proposed (2005). In effect, the question
focuses on how this new organizational form of
schooling (charter schooling) has come about and
by what factors. It is a nice piece in that it compares several theories on organizational environments such as resource dependence theory, neoinstitutional theory and population ecology. Her specific question is -- Why have charter school applications grown?
Now if you drive by a charter school there is
nothing especially noticeable about them. For
those of you who may not know, charter schools
are publicly funded, primary and secondary
schools in the United States that are not subject to
the same rules and regulations as typical public
schools. They are expected to produce certain results, as laid out in their charter, and they are attended by choice as an alternative to other public
schools. If a charter is over-prescribed, attendance
is allotted by a lottery. Charter schools vary.
Some offer a curriculum that specializes in a particular field (e.g., math, arts, or vocation), others
offer a general curriculum and attempt to be more
efficient, cost-effective and outperform usual public schools (e.g., KIPP schools - Knowledge Is
Power Program). Last, all charter schools are
open to inspection and accountability through standardized testing.
Again, Renzulli asks what theory explains the
growth in charter school applications? In a way,
her answer helps future educational entrepreneur
know where he or she should consider opening a
new charter school! To answer the question, she
assesses resource-dependence, neo-institutional,
and population ecology arguments. She renders
her analysis a horse-race between proxy characteristics for each theory.
For resource dependence theory, she tests
whether they arise in districts that give enough
money to open a charter school. This is measured

174

by the instructional expenditure per student that a

new charter school would get.
For neoinstitutional theory, she assesses
whether there are social / political pressures to
open a charter school. She measures this via legislative / union pressure (stronger versus weaker legislation in the state concerning charter schools);
age of founding legislation (charter law exposure);
and number of administrators in a district.
For population ecology she looks at local competition (as proxied by density-dependence in the
# of district charters), state legitimation (# charters
in state), and niche promotion (which she measures as the # private secular schools since those
will promote the demand for charter schools for
the poor).
Her results suggest that educational organizational environments are indeed key in the process
of generating charter school applications. She
finds strong evidence in support of population ecology: e.g., nonreligious private schools increase the
submission of charter school applications (niche
promotion), while the density of extant charter
schools in local districts (or saturation, in general)
decreases the submission of applications (i.e., competition). She also finds some evidence for neoinstitutional theory and resource dependence theory
explanations -- local political environments, solid
funding, and legislative support induce the application for a charter.
So you are an educational entrepreneur
where do you open a charter? Open it in state with
many charters, a neighborhood with many secular
private schools, and a district with few competitors (e.g., few catholic schools). Make sure the district is top-heavy with administrators, that laws are
in place to support charters, and that student expenditures are high.

175

176

Action = identify complementary strengths, form alliances,

establish collaborative/reciprocal norms, create open-ended
mutual benefits where possible, outsource secondary tasks (to
focus on core) all for survival and creation of positive network
environment that delivers service.

Design network to deliver service (select partners and alliances

wisely for aligned values / goals); establish informal, active
communication channels; coordinate member activities (group
processing skills align members culturally, remove internal
competition, create open information, form joint
governance/shared decision making, and get them to focus on
discrete functions / coordination of actual tasks); reinforce norms
of collaboration and reciprocity.

Dominant
Pattern of
Inference

Management
Strategies

Formal and informal roles, relations, and communication

channels. Patterns of relations influence behaviors. Deep
structure consists of values and beliefs in sharing,
communication, and collaboration (trust).

Boundaries no longer clear. Networks apply to within and

between firm relations.

Formal structure conforms to the environment. Often, the technical core

is radically decoupled from institutionally defined org structure (loose
coupling). The logic of confidence makes inspection less necessary, and
practice may be very different from ceremonial classifications or
structures.

Goal is delivery of service via collaboration and outsourcing

aspects not central to technological core..

Goals
(what probs to
resolve)
Social
Structure

Environment

Organizational survival through alignment with the environment.

Linking / coordinating /allying in order to deliver service and

outsourcing / subcontracting / partnering in order to focus on core
technology.
All stakeholders in an organizational field.

Technology
(how solutions
get decided)
Participants

Organizations in a population. Within a population, organizations with same form

(pattern of activity) and resource dependencies in the environment occupy market
niches.

Organizations in a field, professionals, and the nation-state.

Action = Organizations in a population competing to fit an organizational niche

(set of other orgs engaged in same form of activity and relations of
interdependence) and become isomorphic with others in it.

Internal management of core doesnt really apply, but peripheral changes (shortrun strategy) are not inconsistent with theory. Instead, main effort is to be
competitively isomorphic in organizational niches. Organizations can succeed by
recognizing their fit with an environment what population you are in, what the
composition is, what change is occurring, and then whether is makes sense to
adopt a generalist or specialist orientation. Also consider own orgs history and if
your changes will evoke liability of newness. Also consider if your founding
entailed too much of an innovation so that you dont fit a niche (and will die).

Action = Organizations in a field conforming to normative and regulative

environments; the process can be strategic and planned or cognitive and
taken-for-granted.

Buffering: Symbolic coding (systematizing and classifying); Decoupling

organizational elements (loose coupling).
Bridging: Institutional Isomorphism (external pressures via rationalized
myths) occurs in effort to acquire legitimacy. Three forms of
isomorphism are coercive, mimetic, normative

Key feature is environment: Relations of dependence, population composition

(generalists vs specialists; density / carrying capacity), rate of change (coarse or
fine grained), and combinations there from.

Cultural legitimacy and resources. Legitimacy in the environment

necessary for survival.

Core structure of firm is harder to change inertia present making very stable,
hence selection, not adaptation decides organizational fate. Core structure consists
of SOPs, mission, goals, values, etc.

Organizational survival through environmental fit.

External selection in order to fit the environment and insure survival. Population
composition and niche density / carrying capacities determine selection.

Organizations in a population vary in times of change / volatility / crisis, and then

form niches of isomorphic fitting organizations that establish environmental
equilibrium. Variations come in the form of mutations & recombination of forms,
and then they are replicated until carrying capacity and population needs met.

Exists when the level of analysis is an organizational population (not a focal

actor) and the focus is on the variation, birth, survival, death of organizational
forms. Unlike organizational learning, the core structure of organizations is seen
as following structural inertia and therefore unable to adapt much internally.
Adaptation occurs at the population level as firms are selected on the basis of their
static structural forms that mutate (randomly) with each new founding.

Population Ecology

External adaptations in order to fit the environment and insure survival.

Professionals provide expertise and consult to organizations.

Organizations in a field conform to cultural norms to insure survival and

to reduce ambiguity. Legitimacy is a key resource and legitimacy can
come at the expense of organizational efficiency. Professionals and the
nation-state carry the modern cultural recipes and influence the translation
of these elements into the org context.

Organizations focus on network relations, positions, and larger

context in developing strategy. Multiple types of networks are
feasible and they can guide resultant exchanges.

Summary or
Basic
Argument

Exists when the level of analysis is a field (not a focal actor) and the focus
is on conformity to cultural scripts and/or normative constraints on action.
Unlike organizational culture, social structure is based at least as much on
external environment as on internal dynamics.

Neoinstitutional Theory

When the wider context of organizational relations influences

organizational behavior and survival.

When does it
apply?

Network Organization

Summary Table of Network Organization, and Neoinstitutional Theory (NIT), and Population Ecology

References

Baum, J. A. C. 1996. Organizational ecology

Handbook of Organization Studies, eds., S.
Clegg, C. Hardy, and W. Nord (London: Sage, pp.
77-114).
Carroll, G.R. 1981. Dynamics of Organizational
Expansion in National Systems of Education.
American Sociological Review 46, 5:585-599.
Carroll, G.R. 1984. Organizational ecology, Annual Review of Sociology, 10: 71-93.

Hannan, M.T. and J. Freeman. 1989. Organizational Ecology. Cambridge, MA: Harvard University Press.
Hannan, Michael T., and Glenn R. Carroll 1995.
An introduction to organizational ecology. In
Organizations in Industry. Oxford University
Press, pp. 17-31.
Nelson, Richard and Winter, Sydney. 1982. An
Evolutionary Theory of Economic Change. Belknap Press of Harvard University Press.

Carroll, G.R. 1981. Dynamics of Organizational

Expansion in National Systems of Education.
American Sociological Review 46, 5:585-599.

Renzulli, Linda. 2005. "Organizational Environments and the Emergence of Charter Schools in
the United States." Sociology of Education 78: 126.

Carroll, Glenn R. and Anand Swaminathan.

2000. "Why the Microbrewery Movement? Organizational Dynamics of Resource Partitioning
in the U.S. Brewing Industry." American Journal
of Sociology 106, 715-762.

Stinchcombe, A. L. (1965). "Social Structure and

Organizations". In March, J. G. Handbook of Organizations. Chicago: Rand McNally & Co.
pp. 142193.

Davis, Gerald F. and Walter W. Powell. 1992. A

selection from Organization-Environment Relations (pp. 342-354). In Handbook of Industrial
and Organizational Psychology, Vol 3 (2nd ed.).
Eds. Marvin D. Dunnette and Leaetta M Hough.
Palo Alto, CA: Consulting Psychologists.
Gray, Virginia and David Lowery. 1996. The
Population Ecology of Interest Representation.
Ann Arbor: University of Michigan Press (applies
pop ecol to interest groups and their influence on
the legislative / policy-making process nice comparison with Hula).

Swaminathan, Anand. 2001. Resource Partitioning and the Evolution of Specialist Organizations:
The Role of Location and Identity in the U.S.
Wine Industry. The Academy of Management
Journal 44, 6:1169-1185.
Weick, Karl.1979. The Social Psychology of Organizing (Topics in Social Psychology Series).
McGraw-Hill Humanities/Social Sciences/
Languages; 2nd edition.

Hannan, Michael T. and John Freeman 1977.

"The population ecology of organizations," American Journal of Sociology, 82: 929-64.
Hannan, Michael T. and John Freeman. 1977a.
"The population ecology of organizations." Pp.
176-200 in Sociology of Organizations, Oscar
Grusky and George Miller eds..
177

11
Summary of Theories

NM

EN
T)

EN
VIR

EN
VI

RO

ORGANIZATION)

ON
ME

NT
)

Social)Structures)
Technology)

Goals)

VI
EN

Par6cipants)

RO
T)
EN

NM
Figure. Features of an Organization
(adapted from Leavitt 1965: 1145; Scott 2003:18)

)
T
EN
M
N
O
R
I
V
N
E

Summary of Theories
If you consider this text for a moment, you
will realize that you have come a very long way.
You were introduced to something called organizations and their behavior. You were given a wide
array of cases to study from governmental organizations, lobbying groups, technology companies, classrooms, high schools, school districts, universities, reform movements, online educational
courses, multi-player online games, and national
policies.

Magnet school reform (Metz)

Cuban Missile Crisis (Allison)
Chicago public school reforms (Bryk, Shipps)
Hurricane Katrina
Milwaukee parental choice plan (Quinn, Witte)
Lobbyists (Hula)
School desegregation efforts (Weiner)
Legislative efforts NCLB (Kingdon)
Academic senates (Birnbaum)
Xerox machine workers (Suchman)
World of Warcraft (Seeley-Brown)
Learning communities (Louise, Kruse,
Leithwood)
Tech (Kunda)
Ozco (Martin & Meyerson)
University of Chicago-Northwestern merger
(Barnes)
Attempted Union Strike Silicon Systems
(Krackhardt)
Classroom and school networks (McFarland)
National park service (Eggers & Goldsmith)
Charter school networks (Smith & Wohlstetter)
Schools and High Schools (Metz, MeyerRowan)
Intelligent design and teaching of evolution debate (NY Times)
Presidential platforms (Bai)
Massive Open Online Courses (MOOC)
U.S. microbrewery (Carroll)
Wine industry (Swaminathan)
Charter School movement (Renzulli)

You were given a variety of organizational

features to consider the environment, social

structure (behavioral, normative, and cognitive),

participants, technology, and goals. And these afford you a language and checklist by which to consider the complexity of organizations more deeply.
You were also given a variety of theories by
which to consider how those features work together. These conceptions of organizations reflected rational systems (Scott 2003), where administrators made rational decisions in ideal
means-end ways, in realistic boundedly rational
ways of satisficing, and even in rule-matching
manners of duty driven behavior. We then developed more natural system views of how firms
acted like an organism with many internal contingencies. We observed how firms often followed
organizational processes and rules, how others
only coordinated when they had performed political wrangling and formed coalitions, and yet others seemed to follow an anarchic process of flows
into and out of meetings. We also came to view
firms as self-reflective, learning organizations and
as potentially putting in place a social structure
that could sustain that form of association. With
organizational culture we dug deeper into normative and cognitive principles guiding action and
we learned that firms had their own ethos and
styles, which greatly shaped the members experiences.
The past few chapters we looked at organizations as open systems and extended our focus into
the environment and how it influences firm behavior and survival. With resource dependence theory
we focused on dependence relations between
firms; with network forms of organization we
looked at the larger arrangements and patterns of
coordinated actions; with neo-institutionalism we
looked at the deep structures and cultures in the environment and how firms succeed by mirroring
them; and with population ecology we looked a
hard forms of environmental determinism and natural selection due to inter-firm competition.
We went from micro level agents driving an
organization as a unified actor; to meso-level
groups of persons being coordinated by rules, politics, meanings and feedbacks (or sense-making);
to macro-level environments where resource constraints, network contexts of reciprocity and trust,

179

and sociopolitical patterns of belief impinge upon

the organization. We went from micro to macro,
and from rational to natural to open system conceptions.
Along the way, you were given management prescriptions. Each theory saw the world
in certain ways, as being driven by certain facets
and this suggested a variety of maneuvers by
which organizational creation, change, and stability could be accomplished. In some cases, it was
the resources in the environment, in others it was
the beliefs and contexts of relations. In other
cases, the internal dynamics of the firm or school
mattered more: whether people bought into the
companys goals and rituals; and if they did not,
how they could be persuaded or how a decision
could be made anyway. You also learned how
these hard fought lessons of organizing could be
remembered or forgotten, how they could be harnessed toward a learning organization. In my
mind, you have a real toolkit now to be serious researchers and managers. You just need to consider
how these frames or perspective apply, when,
and why.
For any particular case that we have covered,
it is likely we will want to apply multiple theories. Are there ways we can combine our application of these theories? One way of considering
this is to ask how they vary. Some clearly are
more ideal than real so perhaps better suited to
planning than implementation. Similarly, some
are more limited in scope and focus on decision
moments and administration, while others concern
the organizational context and conditions of decisions. Some presume an internal capacity to
change and adapt (i.e., organizational learning, resource dependency theory) while others are deterministic matching efforts (neoinstitutional, population ecology). Some are more focused on deep
structures and culture (neoinstitutional, organizational culture) while others rest at the surface (resource dependence theory, population ecology, rational actor models).
As a manager and analyst, you can combine
theories in various ways. In some instances you
may want to apply them in a (1) staged manner,
such that some theories best apply to the planning

stage (rational actor) and other apply more to the

implementation stage (coalitions).
In other cases, you may find the theories embed
nicely with one another due to their different
scopes. For example, micro theories rational decision making may fit within a natural system view
of the organization. And meso-theories may fit
within larger macro ones focused on the environment (open).
Many of theories seem to have some semblance
with our logics of appropriateness and consequence. As such, they can be applied in a differentiated fashion. We have theories that concern the
logic of appropriateness, sense-making, and culture; and we have theories that concern the logic
of consequence, notions of efficiency and resource
allocations, and mean-end calculations.
Last, many of the presented theories are industry relevant. You can guess which theories might
apply more to certain industries
Finance = consequence-based theories of rational actor, resource dependence, and population ecology.
Knowledge = organized anarchy, organizational
learning, organizational culture, neo-institutional
theory.
Politics = bureaucratic politics, coalition theory,
resource-dependence theory, and network forms
of organization.
Bureaucracy = organizational process and rulefollowing,
In summary, there are at least four master narratives or schemas by which you can utilize the
theoretical tools in your toolkit. I suspect you can
generate more and that is what is nice about having theories as tools. You can rearrange their use
in myriad ways to great effect and become a master analyst and manager!

180

181

Unitary actor or team that

confronts a problem, assesses
objectives (goals) with regard to
it, identifies options, the
consequences of said options,
and then chooses option that
minimizes costs.
Variant: Bounded rationality
and satisficing. Recognize
imperfect info, ambiguity, and
select first satisfactory option
(good enough).
Maximization of options
(solutions).

Summary or Basic Argument

Know alternatives and their

consequences for the shared
goal, and select wisely. Improve
information and analysis.
Management by consequences.

Not salient except as

influencing consequences of
options.

Environment

Management Strategies

Formal roles, hierarchical.

Social Structure

Action = Maximization of
means to ends.

Goals are defined in regard to

problem.

Goals
(what probs to resolve)

Dominant Pattern of Inference

Unified team or actor

Participants

Technology (how solutions get decided)

Exists when there is a unified

actor with consistent
preferences, lots of information,
and clear goals (and time
calculate).

When does it apply?

Rational Actor (RA)

Summary Table of Five Theories to Date:

Know SOPs, what problems they

go with (matching), and who cues
them. Improve rules and matching
with problems. Management by
rules.

Action = output close to prior

output (path dependence), cueing
of SOPs appropriate to problem.

NA

Actors in hierarchical
organizational positions. Cue
sequential routines that accomplish
task or solve problem by routines
available (supply issue).

Objectives compliance to SOPs,

match with problem parts.

Organizational positions

Matching identity and SOPs

(solutions) / programs / repertoires
to problem.

Dividing up problem, coordinating

/ activating organizational actors
who have special capacities /
SOPs for parts of problem,
conducting sequential attention to
objectives (localized searches until
problems resolved). Action guided
by processes / available routines.

Exists when the decision is guided

by a logic of appropriateness
matching problem to actors with
procedures for handling it (routineprocess focus).

Organizational Process (OP) /

Limited Problem Solver (LPS)

Bargain with players (log-roll,

horse-trade, hinder oppositions
coalition formation, etc). Learn
others interests / weaknesses so
you know how to manipulate and
win. Direct management of
relations via bargaining.

Action = result of political

bargaining.

Deadlines and wider array of

stakeholders.

Coalitions enemy/friend

Parochial priorities, goals/interests,

stakes / stands.

Players in positions

Bargaining, or playing the game

(within its rules), or political
maneuvering.

Focus on the players occupying

various positions; their parochial
interests (their conceptions of
problems and solutions); their
resources (expertise, money,
people) and stakes in game; and
bargaining processes between them
that establish agreements /
coalitions.

Exists when there are multiple

actors with inconsistent preferences
and identities, and none of whom
can go it alone without assistance
of others.

Coalitions /
Bureaucratic Politics (BP)

Time when your solution is raised (to

coincide with right participants and
cycle of problems) to maximize energy;
abandon entangled initiatives; know
how to overload system for policies you
detest; and generate choice opportunities
that work to your interests
(access/timing). Indirect managing of
situations.

Action / decision = result of streams

collision in choice arena.

Deadlines and other choice arenas (e.g.,

decision in current arena may be means
of access to another choice arena)

Access rules segmented, hierarchical,

or democratic.

Problems stream determined by public

opinion, prominence / vocalness of
problems in firm, etc.

Confluence of multiple streams, such

that solution is connected to problems
and enough actor-energy to see it
through.
Participant stream shaped by political /
career cycles & unplanned departures.

Focus on choice arenas (when choice

opportunities / windows arise); the
distinct and decoupled streams of
problems, solutions, and participants;
and their access rules to the arena
(whether structural or timed).

Exists when solutions are unclear,

participants turn over, and
preferences/identities are inconsistent.

Organized Anarchies /
Garbage Can (GC)

182

Acknowledges routines, but focuses on practices within

them that enable their continual adaptation and change
to fit reality i.e., practices reflecting organizational
intelligence.

Summary or
Basic
Argument

Action = result of local actors collaborative search (trial

& error / transfer) and adapting rule to situation.

Find ways to create lateral ties among workers so

knowledge is passed / transferred more readily /
quickly (if possible, quickly), create means to
organizational memory of what works. Create applied,
social learning experiences with means to retaining and
transferring expertise. Want communication, collective
improvisation, practice and knowledge sharing to arise.

Dominant
Pattern of
Inference

Management
Strategies

Informal, lateral relations, communication, negotiation,

& collective improv. Actor identities (demand)
important. Network of practice (professional identity /
reach) & community of practice (cohesive group).

Find ways to confer ideology and lead others to identify with

it (using a variety of practices and artifacts), but dont make
it so explicit / fanatical that cynicism emerges. Give room
for autonomy and self-expression so distancing is
unnecessary, and encourage members to generate a culture
of their own (~org learning culture NE to Tech culture which
is top-down engineered).

Action = result of deep structure or culture that is generated

in the organization, but which is mediated by the members
relation to it.

Many elements of culture have origins from outside, and

they are transported in, then translated to the local culture.

Application problems pattern recognition not there (no

fit).

Goals
(what probs to
resolve)
Social
Structure

Source of inter-organizational knowledge / tricks /

transfers.

Deep structure composes the elements of culture themes

(beliefs & norms), their expression via practices (rituals,
etc), and their manifestation or expression in artifacts
(reports, mission statements, etc).

Members of organization doing work / SOPs

Participants

Environment

Create intrinsic motivation (sense of fulfillment), and

remove differentiation / cynicism in most cases.

Internal adaptation, or where actors alter routines for the

better and fit reality (knowledge).

Matching, sense-making / meaning-making, or where actors

seek to express beliefs, norms, and values via a variety of
practices and externalize them in artifacts depicting shared
understandings / notions of appropriateness.
Actors within the organization, and those salient to meaningmaking.

Actors seek expression and fulfillment of identity, and

organizational culture is the medium for such
expression/sense-making.

When the cognitive and normative aspects of social structure

are of concern and seem to guide organizational decisions
(sense-making) and outcomes.

Organizational Culture

Technology
(how solutions
get decided)

Key Organizational Elements

Exists when there are clear feedback loops, adaptations,

memory, and support of actor-expertise / adaptations of
rules to local reality.

When does it
apply?

Organizational Learning (OL)

Summar y Table of Resource Dependence Theory (RDT)

Buffering: protecting technical core from environmental threats

(coding, stockpiling, leveling, forecasting and adjusting scale).
Bridging: security of entire organization with relation to the
environment. Total absorption via merger (vertical, horizontal, and
diversification), partial absorption (cooptation [vertical or horizontal],
interlocks, joint ventures, strategic alliances, associations)

Action = scan environment for resource opportunities and threats,

attempt to strike favorable bargains so as to minimize dependence and
maximize autonomy / certainty.

Key component of the perspective. Exchange partners and external

relations more salient than internal dynamics;
Bridging more relevant than buffering.

Formal roles, standard operating procedures, inter-organizational

bargaining / politics.
(note: coalition approach emphasizes individuals and interests. Here,
the organization is the main actor and exchanges are with other
organizations.)

Goal is organizational survival through external adaptation (certainty

and autonomy).

Focal organization and other organizations with resource

interdependence,

External adaptations in order to increase autonomy and/or decrease

dependence (see management). Comply / adapt, avoid / manage.

Focal organization with input/output concerns that cannot be resolved

without considering the environment.
For the most part, organizations are considered unitary actors (some of
the struggles/internal divisions are minimized) in order to highlight the
interactions with suppliers and clients.

Exists when there is a focal actor interested in decreasing dependence,

increasing autonomy, increasing power, and (possibly) increasing
efficiency.
Preferences and goals are unclear except in relation to dependence.

Resource Dependence Theory (RDT)

183

Action = identify complementary strengths, form alliances,

establish collaborative/reciprocal norms, create open-ended
mutual benefits where possible, outsource secondary tasks (to
focus on core) all for survival and creation of positive network
environment that delivers service.

Design network to deliver service (select partners and alliances

wisely for aligned values / goals); establish informal, active
communication channels; coordinate member activities (group
processing skills align members culturally, remove internal
competition, create open information, form joint
governance/shared decision making, and get them to focus on
discrete functions / coordination of actual tasks); reinforce norms
of collaboration and reciprocity.

Dominant
Pattern of
Inference

Management
Strategies

Formal and informal roles, relations, and communication

channels. Patterns of relations influence behaviors. Deep
structure consists of values and beliefs in sharing,
communication, and collaboration (trust).

Boundaries no longer clear. Networks apply to within and

between firm relations.

Formal structure conforms to the environment. Often, the technical core

is radically decoupled from institutionally defined org structure (loose
coupling). The logic of confidence makes inspection less necessary, and
practice may be very different from ceremonial classifications or
structures.

Goal is delivery of service via collaboration and outsourcing

aspects not central to technological core..

Goals
(what probs to
resolve)
Social
Structure

Environment

Organizational survival through alignment with the environment.

Linking / coordinating /allying in order to deliver service and

outsourcing / subcontracting / partnering in order to focus on core
technology.
All stakeholders in an organizational field.

Technology
(how solutions
get decided)
Participants

Organizations in a population. Within a population, organizations with same form

(pattern of activity) and resource dependencies in the environment occupy market
niches.

Organizations in a field, professionals, and the nation-state.

Action = Organizations in a population competing to fit an organizational niche

(set of other orgs engaged in same form of activity and relations of
interdependence) and become isomorphic with others in it.

Internal management of core doesnt really apply, but peripheral changes (shortrun strategy) are not inconsistent with theory. Instead, main effort is to be
competitively isomorphic in organizational niches. Organizations can succeed by
recognizing their fit with an environment what population you are in, what the
composition is, what change is occurring, and then whether is makes sense to
adopt a generalist or specialist orientation. Also consider own orgs history and if
your changes will evoke liability of newness. Also consider if your founding
entailed too much of an innovation so that you dont fit a niche (and will die).

Action = Organizations in a field conforming to normative and regulative

environments; the process can be strategic and planned or cognitive and
taken-for-granted.

Buffering: Symbolic coding (systematizing and classifying); Decoupling

organizational elements (loose coupling).
Bridging: Institutional Isomorphism (external pressures via rationalized
myths) occurs in effort to acquire legitimacy. Three forms of
isomorphism are coercive, mimetic, normative

Key feature is environment: Relations of dependence, population composition

(generalists vs specialists; density / carrying capacity), rate of change (coarse or
fine grained), and combinations there from.

Cultural legitimacy and resources. Legitimacy in the environment

necessary for survival.

Core structure of firm is harder to change inertia present making very stable,
hence selection, not adaptation decides organizational fate. Core structure consists
of SOPs, mission, goals, values, etc.

Organizational survival through environmental fit.

External selection in order to fit the environment and insure survival. Population
composition and niche density / carrying capacities determine selection.

Organizations in a population vary in times of change / volatility / crisis, and then

form niches of isomorphic fitting organizations that establish environmental
equilibrium. Variations come in the form of mutations & recombination of forms,
and then they are replicated until carrying capacity and population needs met.

Exists when the level of analysis is an organizational population (not a focal

actor) and the focus is on the variation, birth, survival, death of organizational
forms. Unlike organizational learning, the core structure of organizations is seen
as following structural inertia and therefore unable to adapt much internally.
Adaptation occurs at the population level as firms are selected on the basis of their
static structural forms that mutate (randomly) with each new founding.

Population Ecology

External adaptations in order to fit the environment and insure survival.

Professionals provide expertise and consult to organizations.

Organizations in a field conform to cultural norms to insure survival and

to reduce ambiguity. Legitimacy is a key resource and legitimacy can
come at the expense of organizational efficiency. Professionals and the
nation-state carry the modern cultural recipes and influence the translation
of these elements into the org context.

Organizations focus on network relations, positions, and larger

context in developing strategy. Multiple types of networks are
feasible and they can guide resultant exchanges.

Summary or
Basic
Argument

Exists when the level of analysis is a field (not a focal actor) and the focus
is on conformity to cultural scripts and/or normative constraints on action.
Unlike organizational culture, social structure is based at least as much on
external environment as on internal dynamics.

Neoinstitutional Theory

When the wider context of organizational relations influences

organizational behavior and survival.

When does it
apply?

Network Organization

Summary Table of Network Organization, and Neoinstitutional Theory (NIT), and Population Ecology