Y673
Section #10276
INSTITUTIONAL ANALYSIS AND DEVELOPMENT: MICRO
Elinor Ostrom
Tuesday,
Workshop in Political Theory and Policy Analysis
513
SCHEDULE
OF TOPICS: FALL TERM 2004
Preface
The central question
underlying this course is:
To address this question we will
have to learn a variety of intellectual tools to understand how fallible
individuals can influence the rules that structure their lives. This is a particularly challenging question
in an era when global concerns have moved onto the political agenda of most
international, national, and even local governing bodies. Instead of studying how individuals craft
institutions, many scholars are focusing on how to understand global phenomena. It is also an era of substantial violence,
terrorism, and disruption. Many of the
problems we are witnessing today are due to a lack of understanding of the
micro and meso levels that are essential aspects of
global processes.
In our effort to understand
self-governance, we will be studying the four “I’s”: individuals, incentives, institutions and
inquiry.
To understand processes at
any level one needs to understand the individuals
who are participants and the incentives they face. When we talk about “THE” government doing X
or Y, there are individuals who hold positions in a variety of situations
within THE government. We had better
understand how individuals approach making decisions in a variety of situations
given the incentives they face. Those incentives come from a variety of
sources, but a major source, particularly in the public sector, are the rules
of the game they are playing. We define institutions as the rules that specify
what may, must, or must not be done in situations that are linked together to
make up a polity, a society, and economy and their inter-linkages. We are engaged in an inquiry that will never end.
The settings we study are complex and dynamic. Thus, we can learn a framework and an
approach to studying these complex systems.
And, we can learn a variety of theories (and models of those theories)
that help understand particular settings.
We cannot develop a universal model of all settings for all time. Thus, our task of inquiry is a lifelong task. And, the task of citizens and their officials
is also unending.
A self-governing entity is one whose members (or their
representatives) participate in the establishment, reform, and continuity of
the constitutional and collective choice rules-in-use or accept the legitimacy
and appropriateness of these rules. All self-organized entities C whether in the private or public spheres C are to some extent self-governing. In modern societies, it is rare to find any
entity whose members (or their representatives) have fashioned all
constitutional and collective-choice rules that they face. Some rules are likely to have come from
external sources. Many rules will have
come from earlier times and not been discussed extensively among those using
the rules today.
On the other hand, even in a totalitarian polity, it
is difficult for central authorities to prevent all individuals from finding
ways of self-organizing and creating rules of their own. Some of these may even be contrary to the
formal laws of the totalitarian regime.
Given that most modern societies have many different entities, let me
rephrase the first question we started with:
How can fallible individuals achieve and sustain large numbers of small,
medium, and large-scale self-governing entities in the private and public
spheres?
We cannot thoroughly understand the diverse processes
of self-governance in any semester-long or year-long course of study. How humans can govern themselves is a
question that has puzzled and perplexed the greatest thinkers of the last
several millennia. Many have answered
that self-governance is impossible. In
this view, the best that human beings can do is live in a political system that
is imposed on them and that creates a predictable order within which
individuals may be able to achieve a high level of physical and economic
well-being without much autonomy. In
this view, the rules that structure the opportunities and constraints that
individuals face come from outside, from what is frequently referred to as Athe
state.@
For other thinkers, rules are best viewed as
spontaneously emerging from patterns of interactions among individuals. In this view trying to design any type of
institution, whether to be imposed on individuals or self-determined, is close
to impossible or potentially disastrous in its consequences. Human fallibility is too great to foretell
many of the consequences that are likely to follow and efforts to design
self-governing systems rather than making adaptive changes within what has been
passed along from past generations involves human beings in tasks that are
beyond them.
The thesis that we advance in this seminar is that
individuals, who seriously engage one another in efforts to build mutually
productive social relationships – and to understand why these are important –
are capable of devising ingenious ways of relating constructively with one
another. The impossible task, however,
is to design entire social systems “from scratch” at one point in time
that avoid the fate of being monumental disasters. Individuals, who are willing to explore
possibilities and to use reason, as well as trial and error experimentation,
can evolve and design rules, routines, and ways of life that are likely to
build up to self-governing entities with a higher chance of adapting and surviving
over time.
Successful groups of individuals may exist in simple
or complex nested systems ranging from very small to very large. The problem is in a complexly interrelated
world, one needs effective organization at all levels ranging from the smallest
work team all the way to international organizations.
If the size of the group that is governing and reaping
benefits is too small, negative externalities are likely to occur. Further,
even in small, face-to-face groups, some individuals may use any of a wide
array of asymmetries to take advantage of others. Individuals who are organized in many small
groups nested in larger structures – a polycentric system – may find ways of
exiting from some settings and joining others or of seeking remedies from overlapping
groups that may reduce the asymmetries within the smaller unit. If the size of the only group that is
governing and reaping benefits is too large, on the other hand, essential
information is lost and, further, the situation may change from one of adaptive
problem solving to one of exploitation.
Scale and complex nesting are only part of the
problem. Another part has to do with how
individuals view their basic relationships with one another. Many individuals learn to be relatively truthful,
considerate of others, trustworthy, and willing to work hard. Others are opportunistic. Some approach governance as involving basic
problem-solving skills. Some approach
governance as a problem of gaining dominance over others. The opportunities for
dominance always exist in any system of rule-ordering where some individuals
are delegated responsibilities for devising and monitoring conformance to rules
and sanctioning rule breakers. Those who
devise self-governing entities that work well only when everyone is a Asaint@ find themselves invaded by Asinners@
who take advantage of the situation and may cause what had worked successfully
to come unglued and fail.
Thus, the answer I give to the question is:
self-governance is possible in a setting, if . . .
$
most individuals
share a common understanding of the physical worlds they face, of the
importance of trying to follow general principles of reciprocity and fairness,
and that artisanship can be used to craft their own rules;
$ most
individuals have significant experience in small to medium-sized settings where
they learn the skills of living with others, being responsible, gaining trust,
and holding others responsible for their actions;
$ considerable
autonomy exists for constituting and reconstituting relationships with one
another that varies from very small to very large units (some of which will be
highly specialized while others may be general-purpose organizations);
$
individuals learn
to analyze the incentives that they face in particular situations (given the
type of physical and cultural setting in which they find themselves) and to try
to adjust positive and negative incentives so that those individuals who are
most likely to be opportunistic are deterred or sanctioned;
$ the ideas and
principles used in constituting multi-tiered self-governing entities are
sufficiently understood as a Ascience of association@ as new individuals replace those who have taken
initial responsibility for trial and error learning. The theoretical ideas and principles are
continually articulated and learned by new participants.
The above is posed as a Apossibility@
not a determinate outcome. In other
words, we view self-governing entities as fragile, social artifacts that
individuals may be able to constitute and reconstitute over time. We can make scientific statements about what
kinds of results are likely if individuals share particular kinds of common
understandings, are responsible, have autonomy, possess analytical tools, and
consciously pass both moral and analytical knowledge from one generation to the
next. These are strong conditions! I have not tried to develop a formal argument
for these conditions, but I have thought about some of the impossibility
theorems that are implied in the above and promised myself that I will tackle
this problem in the near future.
With this view, self-governing entities may exist as
an enclave in the midst of highly authoritarian regimes. This may not be a stable solution, but
self-governance may provide opportunities to develop productive arrangements
for those who establish trust and reciprocity backed by their own willingness
to monitor and enforce interpersonal commitments. If the macro structure is not hostile, or
even supports and encourages self-organization, what can be accomplished by
smaller private and public enclaves can be very substantial. This is initially
a bottom-up view of self-governance.
Productive, small-scale self-organization, however, is difficult to
sustain over time in a larger political system that tries to impose uniform
rules, operates through patron-client networks, or uses terror to sustain
authoritarian rule. Having vigorous
local and regional governments and many types of voluntary associations is part
of the answer, but not sufficient in and of itself.
Simply having national elections, choosing leaders,
and asking them to pass good legislation, is hardly sufficient, however, to
sustain a self-governing society over the long run. Electing officials to national office and
providing them with Acommon budgetary pools@ of
substantial size to spend Ain the public interest@ creates substantial temptations to engage in
rent-seeking behavior and distributive politics. The central problem is how to embed elected
officials in a set of institutions that generates information about their
actions, holds them accountable, allows for rapid response at times of threat,
and encourages innovation and problem solving.
Solving such problems involves the design of a delicately balanced
system. It requires decisions from
sophisticated participants who understand the theory involved in constituting
and reconstituting such systems and share a moral commitment to the maintenance
of a democratic social order.
Now, what is the role of the institutional analyst in
all of this? Well, for one, it is
essential for those who devote their lives to studying the emergence,
adaptation, design, and effects of institutional arrangements to understand a
very wide array of diverse rules that exist in an equally diverse set of
physical and cultural milieus. To
understand how various rules may be used as part of a self-governing society,
one has to examine how diverse rules affect the capacities of individuals to
achieve mutually productive outcomes over time or the dominance of some over
others. Eventually, one has to examine
constellations of embedded institutional arrangements rather than isolated
situations. And one has to examine the
short-run and long-run effects of many different types of rules on human
actions and outcomes. Further, one has
to acquire considerable humility regarding exactly how precise predictions can
be made about the effects of different rules on incentives, behavior, and
outcomes achieved. Design of successful
institutions may indeed be feasible.
Designed institutions, which tend to generate substantial information
rapidly and accurately and allow for the change of rules over time in light of
performance, are more likely to be successful than those resulting from Agrand
designs@ for societies as a whole.
To be an institutional analyst, one needs to learn to
use the best available theoretical tools, while at the same time trying to
develop even better theories and conducting further empirical studies that
contribute to our theoretical understanding of self-governing systems. All tools have capabilities and limits. The task of the skilled artisan –
whether an institutional theorist or a cabinetmaker – is to
learn the capabilities and limits of all tools and how best to use a
combination of tools to address the wide diversity of puzzles that one comes
across in a lifetime of work.
We need tools to address the puzzle. Relevant tools are plentiful in the sense
that we do have an extensive body of political, social, and economic theory
that focuses on the impact of diverse rules on the incentives, behavior, and
likely outcomes within different settings.
These tools are limited, however, in that many of the most rigorous
theories make assumptions both about the individual and about the settings
within which individuals find themselves which may be problematic at least for
explaining behavior in some settings.
These explicit, and often implicit, assumptions may mask some of the
deeper problems of sustaining democratic systems over time. Many of the difficult problems that human
beings face in trying to develop and sustain democratic organizations are
assumed away when one starts with assumptions that individuals have complete
and perfect information and can make error-free calculations about expected
consequences for themselves and no one else in complex, uncertain worlds.
Further, when assumptions are made that the structure
of the situations facing individuals is fixed and cannot be changed by those in
the situation, little effort is devoted to addressing how individuals affect
their own situations. And yet these same
assumptions (full information and fixed structures) are useful when the analyst
wants to examine the expected, short-term outcomes of an institutional and
physical setting where the options available to individuals are narrowly
constrained and where individuals have many opportunities to learn about the
costs and benefits of pursuing diverse options. Learning which assumptions,
theories, and models to use to analyze diverse institutional arrangements is an
important aspect of the training of institutional analysts.
During this seminar, we will use a variety of
theoretical tools. These will help us to
understand the institutional analysis and development (IAD) framework that we
have been developing over many years at the Workshop. The skilled institutional analyst uses a
framework to identify the types of questions and variables to be included in
any particular analysis. The artisan
then selects what is perceived to be the most appropriate theory available
given the particular questions to be addressed, the type of empirical evidence
that is available or is to be obtained, and the purpose of the analysis. For any one theory, there are multiple models
of that theory that can be used to analyze a focused set of questions. Choosing the most appropriate model (whether
this is a mathematical model, a simulation, a process model, or the design for
an experiment) also depends on the particular puzzle that an analyst wants to
examine.
Objectives of the Course
Given the above background to the substantive focus of
this seminar, let me try to present the central objectives for the semester as
I see them. The objectives are:
$
To understand the
constraints and opportunities of human artisanship.
If self-governance within any particular
organizational setting is only a possibility and not a necessity, then students
of self-governance need to understand the constraints on choice presented by
the structure of a physical, biological, and social world at any particular
point in time as well as the opportunities of using human insight, reason,
persuasion, and vigilance to transform inherited structures.
$
To learn how to
use the Institutional Analysis and Design (IAD) framework as a tool for
understanding the commonalities underlying entities that are often treated by
diverse disciplines as fundamentally different things.
Markets and States are frequently posed as opposite
types of entities. Those who study the
American Presidency or the American Congress sometimes view what they study as
entirely different from European Parliamentary systems, or some of the national
systems of
$
To learn some
very basic elements of game theory as one of the theories that is
consistent with the IAD framework and to gain some knowledge of simple games,
but this is really a very basic introduction and not a course on game theory.
Game theory is emerging as one of the theoretical
tools in heavy use across all of the social sciences (as well as in
biology). Game theory is useful for the
institutional analyst when trying to understand the patterns of outcomes that
result from the operation of repetitive situations over time when the
motivational structure of participants is clearly understood. It also provides a theoretical tool for
analyzing what to expect when rules are changed. As will become obvious in the semester, there
are also many perplexing issues that are not yet resolved both about the theory
of games and its applications to the study of institutions. We will do some reading drawing more on an
evolutionary perspective and how this perspective combined with game theory
helps us understand some of the above issues.
$
To recognize core
problems that humans repeatedly face in a wide diversity of settings such as
those involved in providing and regulating the use of public goods and
common-pool resources, asymmetric information problems, adverse selection
problems, moral hazard problems, aggregation of preferences problems, team
coordination problems, principal-agent problems, and the problems of
constituting complex orders under incomplete information.
Learning how to recognize the key symptoms of the core
problems that humans repeatedly face is essential for institutional
analysts. Diagnosis of the source(s) of
the problems involved in a simple or complex setting is necessary prior to effective
advice about the types of rules, norms, and strategies that have a chance to
improve on outcomes.
·
To understand how polycentric political
systems, including but not limited to federal ones, operate based
on principles learned from this course.
An
irony exists. This problem is at the
heart of recent controversies about how to govern
$
To conduct an
institutional analysis of an important and interesting puzzle relating to human
behavior in a rule-ordered setting at a local, regional, national, or
international domain.
Each enrolled student and visiting scholar will write a
paper to be presented at the Mini-Conference on December 11 and 13 that is an
institutional analysis of a structured situation or linked set of situations
that generate outcomes that are either puzzling, deemed inefficient,
inequitable, unsustainable, or in need of change. It is also important to study situations that
have generated productive outcomes and are worthy of emulation and to identify
what aspects of the structure and human behavior within that structure that has
led to positive results.
Procedures and Requirements for the Fall Semester
During the fall semester of this year-long course, we
try to provide an overview of the literature focusing on the analysis of the
incentives facing individuals within various types of institutional
arrangements. Many of the topics covered
here in one week could well be the topic of a full semester=s work in some other course or seminar. Thus, once you have completed this fall=s work, you will have been introduced to a diversity
of work but you will not yet have gained mastery and will need substantial
further inquiry to gain that mastery.
Fortunately, there are several other courses offered regularly in the
Department of Political Science, the Department of Economics, or the
The assigned readings will either be available on
line, distributed at least one week in advance, or be available at the IU
bookstore. Starred books have been
ordered for the course.
*Jones, Bryan D. (2001) Politics and the Architecture
of Choice: Bounded Rationality and Governance
*McGinnis, Michael D., ed. (1999) Polycentricity
and Local Public Economies.
of
*Schmid,
A. Allan. (2004) Conflict and
Cooperation: Institutional and Behavioral Economics.
MA:
Blackwell Publishing.
*
Experimental
Research.
Graduate students taking the course for credit have
four types of assignments. First, each
student is expected to write a short (1-3 pages) memo to be distributed among
participants in the class every second week starting September 7. Students should
reflect on what they are currently reading and related topics. From time to time, I might ask for comments
on a particular subject. Memos will be
sent electronically, and they are due Sunday
evening by
Second, a take-home exam will be given out during the
week before finals. It will be due on
the Monday afternoon of finals week.
This exam is worth 30% of your grade.
You will be involved in preparing the study questions for this exam.
Third, a final paper is required. Each student and visiting scholar will be
expected to select either a type of problem (such as that of providing a
particular type of public good or common-pool resource) or a type of
decision-making arrangement (such as that of a legislature, a court, or a
self-organized collectivity) and undertake an analysis of how combinations of
rules, the structure of the goods and technology involved, and culture interact
to affect the incentives facing individuals and resulting patterns of interactions
adopted by individuals. The student may
focus more on an operational, a collective choice, or a constitutional-choice
level, but the linkage among these levels should be addressed. Some participants are interested in
large-scale phenomena and will want to examine international or national
regimes. Others will focus on a smaller
scale of organizations. Some may want to
address the Ascaling up@ and Ascaling
down@ question in institutional analysis.
This is an excellent opportunity to do a research
design for a dissertation that applies institutional analysis to a particular
problem. Students may wish to do the
first draft of a paper that eventually will be submitted for publication. All papers will be presented at a
Mini-Conference at the end of the semester.
The final paper is due at the beginning of class on Tuesday, November 30 and constitutes 50% of the final grade. Individuals who turn in late papers will be
required to make copies for participants of the Mini-Conference. Otherwise, the Workshop pays these
expenses. The very latest date that
papers reproduced by participants are accepted is Tuesday, December 7. Since learning how to make deadlines is an
essential skill for all academics, keeping to these deadlines is taken very seriously.
Fourth, active participation in the Mini-Conference
itself is expected. The Mini-Conference at the end of the year (December 11 and 13) is the occasion
during which visiting scholars, students in this seminar, and other Workshop
colleagues present papers summarizing their work for the semester. The final paper will be presented at the
Mini-Conference by someone other than the author who will then also provide an
initial critique, the author will have an opportunity for immediate response,
and there will be a general discussion of each paper clustered together on
relevant panels.
Appointments with Elinor Ostrom
If you would like to schedule a time to talk with me
during the semester, please contact
SCHEDULE OF TOPICS: FALL TERM 2004
Week 1: August
31 Overview of the Semester
Week 2: September
7 Studying Action Situations in
the Field and the Lab
Week 3: September
14 Animating Institutional
Analysis and Analyzing Norms and Rules
Week 4: September
21 Conceptualizing Patterns of
Order in Human Societies (Vincent Ostrom,
leading class discussion)
Week 5: September
28 Type of Rules
Week 6: October
5 Coping with Complexity
Week 7: October
12 Nature of Goods and
Community
Linked Situations
Week 8: October
19 Hierarchical Forms of
Organization & Their Control
Week 9: October
26 Why
Development Assistance Has Failed So Often
Week 10: November
2 Polycentric Systems
Week 11: November
9 Applications of IAD to your
Puzzles
Week 12: November 16 Collective
Choice Processes (Michael Ensley, leading
class discussion)
Class begins at
Week 13: November
23 Thanksgiving Break
Multiple Domains
Week 14: November
30 Constituting and Reconstituting
Multi-Agent, Multi-Level, Overlapping Realms of Local, National, and
International Regimes
Week 15: December
7 Global and Local - Scaling Up
and Down
Mini-Conference: December 11 and 13
COURSE
DESCRIPTION
Week 1: August 31
Overview of Semester: Frameworks, Theories, and Models
at Multiple Levels of Action
Each member of the seminar will be expected to read
the preface for the syllabus (and to have glanced at the rest of it) and to
have begun to think about how their own work might be related to the general
work to be covered during the fall semester.
We will discuss the general organization of the fall semester=s work. There
are several key issues that we will discuss during this class. They include:
$
The differences
among frameworks, theories, and modelsCand how
various theories (and models of these theories) can be used to analyze
particular questions using the institutional analysis framework;
$
The importance of
both static and dynamic analyses when thinking about institutional questions;
$
When single-level
analysis is appropriate and when multiple levels of analysis should be invoked;
$
Thinking about
whole systems and thinking about parts; and
$
Thinking about
impossible and possible rather than only necessary and sufficient.
A central theme of the entire year=s seminar is that human organization is the result of
layers and layers and layers of conscious and unconscious structuringCboth within the single individual and within any
organized polity. To study institutions,
there is no single, correct level of analysis.
To ask any particular theoretical or empirical question, however, an
analyst can generate more useful information by starting to address that
question at one level instead of others.
That is the central message of Hofstadter=s AAnt Fugue.@ His provocative way of developing this
argument will, hopefully, supplement the argument for multiple levels made in
other papers throughout the semester.
Lave and March provide a simple approach for
understanding the usefulness of producing models for social analysis.
Essential
Syllabus
Dyson, Freeman. (1998) AScience as a Craft Industry.@ Science 280 (May 15): 1,014-15. Posted to Oncourse.
Online: http://www.sciencemag.org/cgi/content/full/280/5366/1014
Hofstadter, Douglas R. (1979) AAnt Fugue.@ In Gödel,
Escher, Bach: An Eternal Golden Braid.
“A Psycho-Cultural Interpretation of an American
Sport,” from the
Week 2: September 7
Studying Action Situations in the Field and the Lab
In this week, we will address basic foundation at a
micro level. It involves understanding
adaptation, innovation, entrepreneurship as well as understanding how the human
being has evolved over a long period of time.
The concept of an action situation is one way to
identify the “smallest relevant unit of analysis” for comparative
research. These concepts have been used
to design (1) the various Workshop databases developed to study the effects of
institutions on incentives to provide and appropriate from common-pool
resources; (2) many of our qualitative studies; (3) game theoretical analyses
and (4) experimental studies in the laboratory.
One way of modeling a theory of how a particular action situation is
structured, the likely behavior of participants, the consequences that are
likely to be produced and an evaluation of those consequences is by using
formal game theory. The language of game theory is being used across the social
sciences to analyze a wide diversity of interesting questions.
Essential
de Waal, Frans
B. M. (2003) “The Chimpanzee’s Service Economy: Evidence for Cognition-Based
Reciprocal
Exchange.” In
Fehr, Ernst and Bettina Rockenbach. (2003) “Detrimental Effects of Sanctions on Human Behavior.”
Nature 422
(March): 137-140. Posted to Oncourse. PDF: http://www.nature.com/cgi-taf/DynaPage.taf?file=/nature/journal/v422/n6928/full/nature01474_fs.html&content_filetype=PDF
“Guide to Writing a Scientific Paper.” Prepared by Professor Mat McCubbins
and his students of University of
Harbaugh, William T., Kate Krause, Steven G. Liday Jr., and Lise Vesterlund. (2003) “Trust in Children.” In
Kurzban, Robert. (2003) “Biological Foundations of Reciprocity.” In Elinor
Ostrom and James Walker, eds. Trust
and
Reciprocity: Interdisciplinary Lessons from Experimental Research.
Chapter 4: 105-127.
Schmid,
A. Allan. (2004) In “Institutional and Behavioral Economics
Theory.” Conflict and Cooperation:
Institutional and
Behavioral Economics.
Bookstore]
Supplementary
The IAD framework has been
described by many Workshop colleagues.
Earlier developments are in:
Kiser,
Larry, and Elinor Ostrom. (1982) “The Three Worlds of Action: A Metatheoretical Synthesis of Institutional Approaches.” In Elinor Ostrom, ed. Strategies
of Political Inquiry.
Oakerson, Ronald. (1992) “Analyzing
the Commons: A Framework.” In Daniel W. Bromley, et al., eds. Making the Commons Work: Theory, Practice,
and Policy.
Ostrom, Elinor, Roy Gardner,
and James Walker (1994) Rules, Games, and
Common-Pool Resources,
Sproule-Jones, Mark. (1993) Governments at Work: Canadian Parliamentary
Federalism and Its Public Policy Effects.
Tang, Shui