Title Page
 Table of Contents
 Executive summary
 The need for a research initiative...
 A typology of agri-support...
 Defining and classifying farmer...
 Identifying incentives for collective...
 Potential areas for applied...

Group Title: The potential role of farmer organizations in increasing the productivity and income-earning capability of small-farmer agricultural systems in developing countries : : a call for research
Title: The potential role of farmer organizations in increasing the productivity and income-earning capability of small-farmer agricultural systems in developing countries
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00075306/00001
 Material Information
Title: The potential role of farmer organizations in increasing the productivity and income-earning capability of small-farmer agricultural systems in developing countries a call for research
Physical Description: i, 41 leaves : ill. ; 28 cm.
Language: English
Creator: Byrnes, Kerry J., 1945-
Publisher: s.n.
Place of Publication: S.l
Publication Date: 1990
Subject: Institution building   ( lcsh )
Organizations, Agricultural   ( ltcsh )
Cooperatives, Agricultural   ( ltcsh )
Agricultural development   ( ltcsh )
Popular participation   ( ltcsh )
Genre: non-fiction   ( marcgt )
Bibliography: Includes bibliographical references (leaves 39-41).
Statement of Responsibility: by Kerry J. Byrnes.
General Note: Cover title.
General Note: "Shortened version of a paper originally drafted for the Office of Rural Development, Bureau for Science and Technology, Agency for International Development."
 Record Information
Bibliographic ID: UF00075306
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 29683070

Table of Contents
    Title Page
        Title Page
    Table of Contents
        Table of Contents
    Executive summary
        Page i
    The need for a research initiative on farmer organizations
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
    A typology of agri-support factors
        Page 7
        Page 8
        Page 9
        Page 10
    Defining and classifying farmer organizations
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
    Identifying incentives for collective action
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
    Potential areas for applied research
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
Full Text

The Potential Role of Farmer Organizations in Increasing

the Productivity and Income-Earning Capability of

Small-Farmer Agricultural Systems in the Developing Countries:

A Call for Research1


Kerry J. Byrnes2

'This is a shortened version of a paper originally drafted
for the Office of Rural Development, Bureau for Science and
Technology, Agency for International Development. The author is
grateful for the comments made by reviewers of the earlier paper
as well as the assistance provided by Francis C. Byrnes and John
P. Mason in editing the present paper. The views expressed in the
paper are the author's and do not necessarily reflect those of the
Agency for International Development.
2Senior Social Science Analyst, Program and Policy Evaluation
Division, Center for Development Information and Evaluation, Agency
for International Development, SA-18 Room 220, Washington, DC
20523. Telephone: (703) 875-4961.

Table of Contents

Executive Summary

1. The Need for a Research Initiative on
Farmer Organizations

2. A Typology of Agri-Support Factors

3. Defining and Classifying Farmer Organizations

3.1 Examining Some Evidence
3.2 Defining the Problem

4. Identifying Incentives for Collective Action

4.1 The Olson Perspective
4.2 The Runge Perspective

5. Potential Areas for Applied Research


List of Boxes

Box 1. A Typology of Agri-Support Factors.
Box 2. Potential Tasks of Farmer Organizations.
Box 3. Conditions for Group Action.
Box 4. Factors Influencing Farmers to Contribute to
a Farmer Organization Which Seeks a Group Good.

List of Figures

Figure 1.
Figure 2.
Figure 3.

Figure 4.

A Typology of Agri-Support Factors.
Path of Development of Farmer Organizations.
Major Variables in the Core Set of Nine
Model of Variables Affecting the Contributions
of Local Organizations to Rural Development.

List of Tables

Table 1.

Access to Agri-Support Factors (Production
Assets & Production Services) by Small
Farmers in Zimbabwe.


Donors now recognize that the cornerstone for generating
broad-based increases in agricultural productivity and farmer
income in small farm agriculture in the LDCs lies in improving
farmer access to essential resources. While small farmers make
productive use of available land and labor, they typically have
little or no control over many other essential agri-support
factors (e.g., technology, credit, fertilizers, and market
information). Decisions to supply these factors, and to make
them available to small farmers, all too frequently are made by
state bureaucracies insensitive to the small farmer's needs.

This report sets forth the hypothesis that there is a large
potential catalyst role for farmer organizations to help LDC
small farmers improve their access to and use of essential agri-
support factors. Farmer organizations can greatly improve the
capability of their members to access and manage essential pro-
duction and market resources. The question for future research
is: Under what conditions can the potential catalyst role of
farmer organizations be most effectively developed and nurtured
by donors and assistance agencies? As a step toward defining an
agenda for research, this report reviews what is known about the
conditions under which developing country small farmers will
engage in collective (group) action to obtain improved access to
agri-support factors. An analytical framework is proposed for
conducting applied research that could assist donors to learn how
to harness most effectively the potential role of farmer organi-
zations in the developing countries and, thereby, realize the
development potential of small farmer agriculture.

The paper is organized in five sections, as follows: Sec-
tion 1 introduces the need for a research initiative on farmer
organizations. Section 2 introduces "a typology of agri-support
factors" that identifies a number of agricultural production sup-
port factors around which farmers can, through collective (group)
action, create and participate in farmer organizations that enable
their members to improve their access to those factors having the
greatest utility to member farmers. Section 3 defines the
concept of farmer organization, reviews schemes for classifying
types of farmer organization, examines evidence on the impact of
farmer organizations in the LDCs, and identifies nine problems or
questions to be addressed in a program of research on the role of
farmer organizations in small farmer agricultural development.
Section 4 reviews two approaches to the problem of identifying
incentives for collective action--Olson's "logic of collective
action" and Runge's "assurance problem." Section 5 outlines a
potential approach for an applied research initiative to identify
the "incentive structures" that will support the development of
farmer organizations in the LDCs.

1. The Need for a Research Initiative on Farmer Organizations

Today, donors, technical assistance agencies, and host
developing country governments search for solutions to institu-
tional and organizational problems that impede the pace of
development of small farmer agriculture. These problems include
but are not limited to development of site- and situation-
relevant information (e.g., technology), dissemination of this
information to farmers in scattered and remote locations, and
timely delivery of various goods and services. Another problem
demanding attention is that of providing farmers with opportuni-
ties to market their production at a fair price and to grow non-
traditional crops with greater market demand than those tradi-
tionally grown. The search has led to a recognition of two
ideas. While not yet universally accepted, these ideas have
significant implications for the design of agricultural
development strategy. They may be stated as follows:

1. Small farmers, often with a large share of production for
subsistence, as distinct from large-scale, primarily
commercial farmers, can help a developing country meet its
requirements for increased production of food and other
agricultural commodities.

2. Small farmers will not increase the production and produc-
tivity levels of their agricultural systems unless such
increases enable them to achieve a significant and relative-
ly assured increase in the income which they derive from
such production.

For those who accept the validity of these two ideas, the
implication is clear for the design of agricultural development
strategy: If small farmers are to significantly contribute
toward a developing country's requirements for increased produc-
tion, the country's strategy for agricultural development must
enable its small farmers to increase the productivity and the
income-earning capability of their agricultural systems (Johnston
and Tomich, 1984).

A strategy to stimulate development of small farmer agricul-
ture could include many potential components. This paper focuses
on one of these components, namely, farmer organizations, and
their potential to address the institutional and organizational
problems involved in increasing the productivity and income-
earning capability of small farmer agricultural systems. In view
of the magnitude of these problems as well as the continuing
weakness and inefficiency of state agricultural bureaucracies in
dealing with them, and the fiscal deficits that confront these
bureaucracies, there has been increased interest in questions
concerning farmer organizations. These questions include but are
not limited to the following:


1. Can small farmers, through their participation in a
farmer organization,

a. Improve their level of access to the various goods and
services essential for increasing the productivity and
income-earning capability of their farms?

b. Reduce the levels of risk and uncertainty which they must
face in making decisions about the management of their

c. Increase the productivity and income-earning levels of
their farms?

2. Why are some farmer organizations more effective than others
in helping small farmers to meet their needs for agricul-
tural goods and services?

3. What elements determine a farmer organization's capability
to help small farmers get their needed agricultural goods
and services?

4. Which of the elements identified in answering question 3 can
be controlled and how can these elements be manipulated so
as to increase the productivity and income-earning capabili-
ty of small farmer agricultural systems?

5. What elements lead or cause individual small farmers to
engage in collective (group) action to establish new farmer
organizations, to support and participate in existing farmer
organizations, or to improve the capability of existing
farmer organizations to help them meet their needs for
agricultural goods and services?

6. Which of the elements identified in answering question 5 can
be controlled and how can these elements be manipulated so
as to achieve a desired level of small farmer support of and
participation in a farmer organization?

While there is increased interest in questions such as these
(see Esman and Uphoff, 1984:22-23 for references), this interest
stems from a practical concern over a basic question. Can farmer
organizations play a larger and more effective role in the devel-
oping countries (LDCs) by helping these countries to identify and
implement ways of accelerating the pace of developmental change
in small farmer agriculture?

The potential role farmer organizations can play in the LDCs
in creating a demand for and supply of agricultural support
institutions that effectively reduce constraints to technology
utilization in small farmer agriculture has been highlighted by
Hayami and Ruttan (1984:56; Ruttan and Hayami, 1984:211, 217-

218). They find that technological change depends on a mix of
technical and institutional reforms that reallocate resources "so
as to remove those resource constraints that are most inelastic
and those institutional constraints that are most restrictive of
growth and development" (Ruttan, 1978:413). Addressing the ques-
tion of which types of technical and institutional reforms will
be required for technological change in LDC agriculture, Hayami
and Ruttan (1984:56) cite Grabowski's (1981) statement that, in
addition to developing improved seed, research activities:

...must be directed at...improving cultivation practices and
irrigation techniques in order to increase cropping
intensity. Credit must be made available to allow farmers
with small farms to irrigate their land and thus increase
their cropping intensities. Larger farmers' privi-
leged access to machinery must be eliminated. All of
these require an increase in power and influence of farmers
with small farms, relative to those with large farms, on
government decisions concerning agricultural research and
credit priorities. This could possibly be accomplished
through land reforms or, a less radical solution, the
organization of small farmers into groups which could put
pressure on government agencies to recognize and respond to
the interest of small farmers [emphasis added].

While asserting that technical and institutional reforms are
"clearly desirable," Hayami and Ruttan (1984:56) ask: "But what
are the conditions that make them [the clearly desirable reforms]
economically and politically viable?"

Several studies provide evidence that farmer organizations
can play a positive role in support of agricultural development.
Bratton (1983) found that almost half of 500 randomly-sampled
households in Zimbabwe's tribal areas were members of agricul-
tural organizations. Maize farmers who were members of these
organizations consistently outproduced individual (non-member)
maize farmers (Bratton, 1983:17). Oxby (1983:54) reported higher
yields among rainfed group farmers in Kenya compared to non-group
farmers assisted by extension workers. Another example comes
from the World Bank-sponsored URADEP project in Ghana. This
project encountered difficulties for several years after failing
to set up the farmer groups which the project design had included
as a means of helping farmers to operate credit, seed, and ferti-
lizer programs from farmer service centers. The project began to
get "on track" only after such groups were established (Uphoff
and Van Dusen, 1984:42).

The evidence reported by Bratton, Oxby, and others raises
questions about the potential role that farmer organizations
could play in helping LDCs to bring about developmental change in
small farmer agriculture. Because answers to such questions are
not readily available, there is need for a framework that would

address how the potential role of farmer organizations could be
most effectively identified and developed.

The need for such a systematic framework or methodology also
was identified by Cernea (1981) in a study of peasants' groups in
164 World Bank-financed rural development projects initiated in
the period 1973-1977 in 80 countries. Each project's appraisal
report (the design at the beginning of the project) was reviewed
to determine whether the report: (1) considered patterns of tra-
ditional cooperation based on labor exchange, group management of
water, or savings and credit; (2) indicated that group action in
the form of community "self-help" would be encouraged; or (3)
indicated that farmers' groups related to productive or marketing
activities or other types of "pre-coops" would be established
during project implementation. More than 40 percent of the
projects in 37 countries contained one of these elements.

Data on these projects also were collected through review of
other project documents. The following findings are significant
in terms of the general state-of-the-art (Cernea, 1981:132-133).
The design of most projects has not been guided by a policy or a
methodology for identifying traditional, or for establishing new,
forms of farmer organization or for incorporating them into the
mechanisms of planned and financially supported development ini-
tiatives. Despite the absence of any policy to involve farmer
organizations, the creation (or strengthening) of a variety of
quasi-formal small farmer groups was "an important proportion" of
the projects reviewed. The "main reason" identified for this
consideration of farmer organizations, in the absence of any
formal policy on this matter, was "the intuitive perception or...
conviction of the individual staff involved in the design or
appraisal of projects that the development process needs to rely
upon and promote the structured self-organization of small pro-
ducers for their own interest." But these projects faced a major
constraint, namely, the absence of a "practical methodology" or
"social technology for building farmers' groups, for identifying
and effectively supporting (i.e., financially and technically)
existing traditional organizations." Even when project design
and appraisal incorporated provisions relating to farmer organi-
zations, they only allowed these organizations "a rather second-
ary function in the overall project effort." Even when projects
took "a relatively open-minded stand on the potential of farmers'
organizations," most "appeared to have a development mechanism
not centered around the self-organization of the producers for
attaining their own self-sustained growth."

Support for these findings comes from a study of ex-post
evaluation reports on World Bank-assisted rural development
projects in the 1960s-1970s (Kottak, 1985). While the develop-
ment potential of group action and local social organization was
recognized in the design of some of the projects reviewed by
Kottak, this recognition was rarely effectively harnessed during


project implementation. Kottak (1985:39-40) found that social
design flaws in 36 (53%) of the 68 sampled projects caused many
projects to fail or to achieve less than could have been
achieved. (See also Cernea, 1987).

Three reasons for this poor track record in project design
and implementation were: (1) inadequate socioeconomic knowledge
in project preparation, (2) lack of social skills in project man-
agement units to carry out the project goals, and (3) use (often
unconscious) of culturally-biased and therefore often incompati-
ble social designs for innovation. The lack of accurate social
and institutional knowledge as an input to project design led to
deficient or inappropriate social strategy for project implemen-
tation (e.g., ignoring traditional social organization and impos-
ing non-traditional organizational forms). As a result, (1)
appropriate groups with development potential were ignored; (2)
inappropriate, unworkable, or unnecessary new organizations were
formed; and (3) assumptions about individual motivations were
made that conflicted with traditional communal values.

Further, as the A.I.D. Policy Paper on Local Organizations
in Development states, the "literature examining the results of
[A.I.D.] experiences and other donor attempts to deliver assis-
tance to local organizations is...often contradictory" (A.I.D.,
1984:1). Thus, it is not surprising that top-down initiatives to
establish farmer organizations have often proceeded on little
more than an ad hoc basis; as a result, these organizations all
too frequently have failed, been short-lived, or performed at a
level far below that which had been expected (cf. Galjart and
Buijs, 1982:3; Esman and Uphoff, 1984:164-165). Initiatives to
create or revive such organizations often were undertaken without
consideration of potentially applicable theory, past research,
and/or relevant field experience. In the process, some

...efforts have been very successful, whereas others have
been notable failures. Some types of cooperatives have
worked well in one country but not at all in another. One
form of cooperative may have thrived in a country, while
another form did not. There is still no general body of
theory, based on empirical examination of experience, that
provides adequate guidelines for planning and implementing
successful cooperative projects (D.A.I., 1984:2).

Indeed, it is generally the case that research on farmer
organizations has not resulted in any systematically-organized
body of knowledge that would provide guidelines for establishing
the necessary and sufficient conditions to support the develop-
ment of farmer organizations that effectively help farmers to:
(1) improve their level of access to the various goods and ser-
vices that are essential for increasing the productivity and
income-earning capability of the agricultural systems which they
operate; (2) reduce the levels of risk and uncertainty which they

must face in making decisions about the management of their agri-
cultural systems; and (3) increase the productivity and income-
earning levels of these agricultural systems.

Given this dearth of knowledge about farmer organizations in
agricultural development, there is a need for research that would
provide guidelines on the conditions under which farmer organiza-
tions can facilitate developmental change in small farmer agri-
cultural systems. While there is recognition of the development
impact that could be achieved by utilizing farmer organizations
as catalysts for technical and institutional change in LDC agri-
culture, donors and technical assistance agencies have lacked
adequate knowledge about how they or their intermediaries (e.g.,
private voluntary organizations) can most effectively support
farmer organizations in this catalyst role. Thus, as a recent
Office of Technology Assessment report noted:

Major donors have been largely ineffective working at the
local level. Many donors have failed to tap the potential
of local organizations and sometimes have made overwhelming
demands on local groups and thus, undermined the groups'
effectiveness. Yet the needs of local groups are large
enough that they may require the resources available only
from major donors (OTA, 1988:16).

What is needed is a commitment to develop the knowledge that
will enable donors and development assistance agencies to know
when (under what conditions) and how farmer organizations most
effectively can support development and transfer of agricultural
technologies that enable small farmers to increase the productiv-
ity and income-earning capability of their agricultural systems.

This, basically, is the rationale for a research initiative
on LDC farmer organizations. Through a properly-focused research
program a donor such as A.I.D. would learn how that Agency's
"four basic programmatic components" (A.I.D., n.d.:4) could be
used to: (1) tap and increase the overall development impact of
farmer-controlled resources--the strategy of reliance on the
private sector and market forces; (2) identify and find means of
relaxing technical and institutional constraints to technology
development and transfer in LDC small farmer agriculture--the
strategy of technology research, development, and transfer; (3)
upgrade the service capability of rural institutions (credit,
production input supply, commodity markets, etc.) to support an
accelerated pace of growth and development in LDC small farmer
agriculture--the strategy of institutional development and
training; and (4) improve communication between host country
governments and a major rural constituency, namely, small
farmers--the strategy of policy dialogue.

Finally, the knowledge that could be gained via an applied
research initiative on farmer organizations in the LDCs would
provide a donor country such as the U.S. with a major cornerstone
for addressing a range of foreign assistance objectives, such as
growth, environmental sustainability, poverty alleviation, and
pluralism, as proposed in the Hamilton Task Force Report (1989:
31-33) on U.S. foreign assistance.

2. A Tvpology of Agri-Support Factors

The small farmer's ability to increase his (her) productivi-
ty and income depends on their ability to manage their agricul-
tural systems. But small farmer decisionmaking increasingly has
become more complex in terms of the public and private sector
institutions and organizations that provide and/or regulate
essential resources. This has raised the levels of risk and
uncertainty small farmers face (cf. McInerney, 1978). These
"resources" or agricultural production support factors (agri-
support factors) are defined in the typology of agri-support
factors in Box 1.

Only some of the agri-support factors (e.g., seed as a pro-
duction input) actually enter physically into a farmer's agricul-
tural system. But a farmer's decisions must take into account
all agri-support factors directly or indirectly involved in
production and marketing of a crop. Direct factors enter the
production process (e.g., production inputs such as fertilizers),
while indirect factors (e.g., market prices) "condition" the
decisions farmers make about direct inputs. Whether a farmer
chooses to take a production credit loan (capital) to buy ferti-
lizer (production input) depends on his or her perceptions of how
much surplus production that fertilizer will yield above the
farmer's consumption requirement and whether it will be possible
to sell this surplus at a profit (markets).

When looking at the farm management decisions a farmer
makes, it is useful to remember that a farmer's demand for any
agri-support factor is a derived demand. An example of the
derived demand concept is provided by a production input such as
fertilizers that enters directly into production. The quantity
of fertilizer used is determined by the quantity that will bring
the farmer a favorable return when the agricultural surplus pro-
duced by using that quantity is sold. Thus, the quantity of
fertilizer a farmer will use varies inversely with the price of
this input and directly with the price consumers (or marketing
intermediaries) are willing to pay for the surplus produced.

Thus, the level of economic incentive a farmer requires as a
condition for using an agri-support factor is a key part of the
total structure of incentives (cf. Runge, 1984a; Runge, n.d.)
involved in motivating farmers to increase their productivity.
Figure 1 depicts the relationship between the income-earning
capability of the farm family household, farm management deci-
sionmaking, and the typology of agri-support factors. Household
"income-earning capability" is based on non-farm (including off-
farm) employment and the agricultural production activities in
which household members engage. Agricultural production is
directly influenced by the farmer's management decisionmaking
which includes three components: (1) informal research deci-
sions--those decisions involved in conducting informal experi-
mentation to identify improved agricultural production technolo-
gies; (2) technology choice decisions--those decisions involved
in selecting, among technologies, a technology for production of
a crop or animal; and (3) production implementation decisions--
those decisions involved in putting a traditional agricultural
production technology into practice.

Box 1. A Typology of Agri-Support Factors.
o Land -- to provide the physical structure (soil) and
nutrients essential for crop or animal production.
o Access to Unit of Production -- to provide the right to
use a particular unit of production (e.g., a parcel of
land) for crop or animal production and to derive a
benefit from such use.
o Water -- to provide essential moisture for crop or
animal production.
o Technology -- to provide the how-to knowledge for
producing a particular crop or animal.
o Production Inputs -- to provide the seeds/seedlings,
fertilizers, chemicals (herbicides, insecticides, and
fungicides), traction, and implements required for
producing a crop or animal according to a particular
o Capital to provide the financial means (in cash or
credit) to purchase the production inputs used in
producing a crop or animal according to a particular
o Labor -- to provide for the performance of the various
physical tasks and cultural practices involved in
producing a particular crop or animal.
o Markets and Market Information -- to provide outlets for
the sale of surplus production, awareness-knowledge
(information) on market conditions (e.g., market prices),
and performance of processing, storage, and other
marketing functions.
o Infrastructure -- to provide the physical base for the
communication of technological and market information;
and the transportation of people, production inputs, and
o Policy -- to provide a favorable environment for invest-
ment in agriculture (e.g., currency exchange rates,
crop/input price ratios, institutional support for
contracts, etc.).


Figure 1. A Typology of Agri-Support Factors.




Local Government
(village council)
Farmer Associations

Construction Committee
Local Development
Associations (LDA)

Marketing Cooperative
Marketing Association
Pre-Coop Producer Group

Labor Exchange Group
Mechanization Group

Credit Union
Credit Cooperative/Group
Saving Club

Input Supply Cooperative
Buying Club/Pre-Coop

Producers Association
FSR Group
T&V System Group
Farmer School

Water Users Assocation
Irrigation Channel Group

Joint/Group Farming
Production Cooperative

Land Clearing Group

Local Administration


Commodity Markets
(Assembly, Whole-
saling, Retailing)

Wage Market

Banks / S&Ls

Retailers & Other
Sales Agents
(e.g., stockists)

Informal Information
Exchange Network
Formal Research &
Extension System

Traditional Regula-
tion Mechanisms

Land Tenure (owned,
fixed rent, etc.)

Land Market



Next, the various agri-support factors enter into the farm-
er's decisionmaking and, thereby, influence the productivity and
income-earning capability of the farmer's agricultural system.
The potential for a farmer's production implementation decisions
to be effective in achieving a particular production or income
objective set by a farmer depends on the farmer having adequate
access to and control over the quantity and quality of agri-
support factors required to meet the farmer's objective. How
effectively small farmers access and control the agri-support
factors they require determines to a significant extent how suc-
cessful they are in increasing the productivity of and income
derived from their agricultural systems. Typically, small farm-
ers individually exert limited control over the quantity and
quality of the agri-support factors they require. Major control
over the supply of many agri-support factors required by farmers
normally lies in the broader agri-support system over which small
farmers have little or no individual power. To garner counter-
vailing power, farmers have engaged at times in collective action
to improve their access to and control over various agri-support
factors. Such action has resulted in various types of organiza-
tional forms; more specifically, farmer organizations (Fig. 1).

A farmer group may provide its members with improved access
to or control over essential agri-support factors by (1) devel-
oping the group's capability to deliver essential factors to
member farmers and/or (2) acting as the group's agent to negoti-
ate with traditional, market, and/or bureaucratic systems for
improved delivery of these factors. To the extent that collec-
tive action by farmers is effective in establishing a farmer
organization and improving the access of member farmers to agri-
support factors, such improved access constitutes what we may
call a "group good." A group good may be defined as a benefit
available only to individual members of a group, regardless of
whether any particular person in the group has contributed to the
development or attainment of the group's ability to provide the
benefit to the group's members. Where the lack of collective
action by farmers effectively precludes them from accessing agri-
support factors essential for increased productivity and income-
earning capability, such farmers likely will continue to incur
what may be termed a "group bad" or the "costs of not contri-
buting" to support collective action.

Effective collective action by farmers to attain a group
good can be jeopardized by the presence of one or more institu-
tional problems: the free rider problem, the inconsequentiality
problem, and the organizing problem. In the free rider situa-
tion, a group member does not perceive any individual utility in
supporting the collective action required for a group to make a
group good available to its members; yet the group member is
willing to take advantage of the group good as long as it is
available. If every potential or current group member is only
willing to be a free rider and is unwilling to contribute in any


way to support the group-oriented effort, then there likely will
not be a sufficient level of contribution to sustain the group;
nor will the group good be provided for long.

In the inconsequentiality situation, potential or current
members of a group or organization perceive that their individual
contributions to support collective action to attain a group good
represent such a small portion of the total resources required to
make the good available that these contributions will be inconse-
quential in affecting the outcome of the group's effort to pro-
vide the good. Therefore, such individuals will not make the
individual contribution necessary to support effective collective
action by the group to attain the good in question.

In the organizing situation, potential or current members of
an organization perceive that the costs of organizing collective
action to attain a group good are sufficiently high so as to
preclude the possibility of the group (organization) being able
to cover the full costs involved in making the good available.
Accordingly, the collective action required to attain the good
in question will not be undertaken.

A group's collective action effort, if it is to be success-
ful, must provide current and potential group members with suffi-
cient "incentive" to overcome or preclude the potential dampening
effect of these three problems on an individual's motivation.

3. Defining and Classifying Farmer Organizations

The literature on farmer organizations documents the variety
of farmer organizations in the developing countries.3 The initi-
ative for establishing a farmer organization may arise from any
of several sources. For example, individual farmers may estab-
lish some form of organization to gain a measure of countervail-
ing power in relation to traditional (e.g., landlords), market,
or bureaucratic institutions and organizational forms. A govern-
ment may require farmers to establish a formal organization as a
condition for gaining access to a particular agri-support factor
(e.g., that farmers organize into a "credit society" or a water
users association as a condition for obtaining a group loan or
resources to line a watercourse). An outside agent may sponsor
or promote organizational efforts to establish some organization
which the agent believes will facilitate and motivate farmers to
gain access to one or more agri-support factors.

3See Uphoff and Van Dusen, 1984; Esman and Uphoff, 1984; Oxby,
1983; Johnston and Clark, 1982; Cernea, 1981; Lele, 1981; Nesman,
1981; Obern and Jones, 1981; PCARR, 1981; and Morss, et al., 1967.

While the source of initiative for establishing a farmer
organization is a potentially important variable, more important
is the question of whether such an organization can help its
member farmers to achieve three interrelated objectives: (1) To
gain an increased measure of access to and control over essential
agri-support factors not readily available to individual farmers
through existing traditional, market, or bureaucratic organiza-
tions; (2) to reduce the levels of uncertainty and risk which
small farmers must face in making decisions about the management
of their agricultural systems; and (3) to increase the productiv-
ity and income-earning capability of agricultural production
systems operated by the small farmer.

The relationships among these objectives are readily
apparent. Certain farmer organization activities may focus on
getting (gaining access to) specific agri-support factors (ASFs),
while other focus on using ASFs or helping farmers to make better
farm management decisions once they have access to ASFs. Gaining
access to ASFs (objective #1) may directly contribute to produc-
tion and income (objective #3). Reducing risk and uncertainty
(objective #2) may come from accessing ASFs (e.g., timely avail-
ability of production inputs) or from collective action to use
ASFs (e.g., coordinated planting dates to reduce pest/disease
losses or joint crop protection measures against animal damage).

With these objectives in mind, a farmer organization may be
defined as any group of farmers enabling its individual members
to achieve more effective control over the supply of, access to.
and/or proper use of one or more agri-support factors entering
directly or indirectly in agricultural systems operated by group
members individually or jointly. This definition includes formal
and informal groups, regardless of a group's sex composition.

Farmer organizations may be classified in different ways
(Torgerson, 1977). Robert C. Flick (personal communication) of
Agricultural Cooperative Development International, identifies
four kinds of farmer organizations: (1) a business organization
that seeks to improve farmer income by providing services needed
by the farmer to produce (e.g., an input supply cooperative); (2)
a supply management organization that seeks to enforce production
quotas to control the amount of a particular crop reaching a mar-
ket (e.g., the supply of California navel oranges is controlled
by a marketing order); (3) a research, promotion, quality (grades
and standards), and advertising organization that seeks to im-
prove varieties, develop quality guidelines for the market, and
expand the market (e.g., The California Almond Association); and
(4) a representational, lobbying organization that seeks to im-
prove farmer income by advancing farmers' interests on the poli-
tical and governmental front (e.g., National Council of Farmer
Cooperatives, National Farmers' Union, American Farm Bureau
Federation, Land O'Lakes Political Action Committee).

Another classification scheme (Box 2) proposes a typology of
potential tasks (or functions) which a local organization, more
specifically, a farmer organization could undertake. These tasks
can be viewed as four pairs, constituting a continuum from initi-
ating activity (A) to influencing the external political-admini-
strative environment (D). In between are activities pertaining
to the resources and services that constitute the inputs and
outputs of organization.

A third approach is found in Bratton's (1985) study of the
role of farmer groups in food production in Zimbabwe. Bratton
hypothesized that the type of collective action in which farmers
engage is largely determined by the nature of the resource around
which collective action takes place. Bratton focused on whether
and to what extent a scarce resource is influential in determin-
ing the type and likelihood of collective action by farmers to
obtain improved access to and control over that resource.

Box 2. Potential Tasks of Farmer Organizations (Adapted
from Esman and Uphoff, 1984:72-82. 295).
A. Intra-Organizational Tasks
1. Planning and Goal-Setting: assessments of com-
munity or group needs and of various problems,
means, and strategies; and formulation of plans
to deal with needs and problems.
2. Conflict Management: efforts to resolve con-
flicts within community or organization, to
facilitate production or maintain social harmony.
B. Resource Tasks
3. Resource Mobilization: gathering community
resources or gaining resources from outside
sources for development efforts.
4. Resource Management: efficiency and correctness
in resource use, including financial, organiza-
tional, and natural resource management.
C. Service Tasks
5. Provision of Services: delivery or distribution
of services, either those of the farmer organi-
zation or from outside sources farmer organiza-
tion involvement.
6. Integration of Services: coordination of serv-
ices, either of a farmer organization or outside
sources, so they most efficiently and effectively
meet members' needs.
D. Extra-Organizational Tasks
7. Control of Bureaucracy: efforts to make govern-
ment staff work harder, more flexibly, and more
cooperatively with and for members to ensure
attendance at office, field visits, lack of
corruption, etc.
8. Claim-Making: efforts to get government
decisionmakers to deal with farmer problems
and needs; may include getting rules altered,
budget allocations changed, etc.

Bratton classifies agricultural production support factors
(or production resources) into (1) production assets which are
basic material goods vested in the farm family through ownership
or use rights); and (2) production services which are supplied by
public and/or private agencies but are not owned by the household
though it may purchase and consume them. In terms of the typolo-
gy of agri-support factors (Box 1), production assets are land
and access to unit of production, labor and draft power, and pro-
duction inputs (seed). Production services are water, technology
(information), production inputs (fertilizers), capital (credit),
markets and market information, infrastructure, and policy.

Next Bratton distinguishes between the level of production
(level at which farmers mobilize and allocate production assets);
and level of exchange (level at which farmers enter transactions,
usually using money, to access production services outside the
household or village). Thus, we have two major types of farmer
groups--those that "pool" production assets from household
resources at the production level, and those that "bulk" demand
for production services from the state or market at the exchange
level. Bratton highlights four specific types of farmer groups--
information, labor, market, and multipurpose. An information
group brings farmers together for information or educational
purposes. A labor group pools labor assets of individual house-
holds. A market group bulks purchases and sales of production
services (e.g., credit, production inputs, and agricultural
produce marketing). A multipurpose group engages in both pooling
for production and bulking for exchange.

Through their groups farmers are able to obtain new ideas
and practices, supplement a meager household resource base, and
create an effective demand for services. To the extent that farm-
er groups improve member farmer access to agri-support factors,
reduce uncertainty and risk, and increase productivity and
income, these groups increase their members' control over an
uncertain and risky physical and institutional environment.
These groups may be arrayed developmentally, some engaging in
more "complex tasks," others being more dependent on "member
cooperation." The interaction of these variables--task complex-
ity and member cooperation--provides a model (Figure 2) of how
farmer organizations grow and change. According to this model,

market groups are more complex than labor groups. They have
more members, keep more formal records and must coordinate
their activities with outside agencies. Labor groups, how-
ever, are more cooperative than market groups. Although
they are smaller, member contact is more regular and inten-
sive and involves the commitment of assets. Multi-purpose
groups only come about when labor groups add supply and
marketing to their repertoire of functions. They are the
most complex and cooperative organizations of all (Bratton,


Figure 2. Path of Development of Farmer Organizations (Adapted
from Bratton, 1985).

Supply &
(no pooled

(pooled private

owned assets)

NORMAL Multipt

LOW Scale of Member Cooperation


3.1 Examining Some Evidence

Table 1 summarizes key findings from Bratton's survey of 464
randomly selected households in four rural districts of Zimbabwe.
Of the sampled households, 204 (44%) belonged to one or more of
the four identified group types, with group membership played a
significantly greater role in facilitating farmer access to "pro-
duction services" than to "production assets." The improved
access to production assets" afforded by group membership had a
significant impact on the productivity and income-earning capabi-
lity of farmers who were members of groups. In the 1981/82 sea-
son, group farmer maize yields were 64% to 137% higher per acre
and production levels (in tons) were 89% to 172% higher than
individual farmers; and group farmers sold two to seven times
more maize and at a higher average price than did individual
farmers. Further, as groups develop along the path hypothesized
in Figure 2, farmer organization played an increasingly smaller
role in helping members "pool" production assets and an increas-
ingly greater role in helping members "bulk" demand for produc-
tion services. Bratton concludes that small farmers will more
readily organize at the exchange than the production level and
that the "nature of the resource" around which farmers will most
readily organize is more likely to be a "central service" (e.g.,
credit) than a "household asset" (e.g., land).


e I-



Table 1. Access to Agri-Support Factors (Production Assets &
Production Services) by Small Farmers in Zimbabwe
(Adapted from Bratton, 1985).
All Individual Group
Farmers Farmers Farmers
(----------- -- -- )
Production Assets
Land 12 11 13
Labor 38 21 46*
Draft oxen and implements 44 48 40
Production Services
Advice from extension workers 55 31 86*
Advice from missions and
fertilizer companies 16 54
Loan from credit scheme 18 7 32*
Recovery rate -- 54 1-92
Production inputs (fertilizer) 61 48 77*
Obtained directly from
manufacturer -- 38 77
Ordered early for cash rebate
and early delivery -- 44 58
Paid cheaper bulk road
haulage rates (**) -- 44 59
Markets and Market Information
Sold crop through an official
marketing agency 39 25 57*
Infrastructure (transportation)
Complaints about transport -- 77 61
Complementary package (tech-
nology, fertilizer, and
market) -- 14 57
p .001
** "...in Wedza in 1984 group farmers paid an average of 13%
less for a bag of compound maize fertilizer than their
individual counterparts. In multipurpose groups in Zwimba
the savings were as high as 33%" (Bratton, 1985:16).

This finding is consistent with a D.A.I. study on "coopera-
tives in development" that found that a cooperative, to be suc-
cessful, must be organized around a key resource that an institu-
tion can efficiently mobilize, provide, or market. "Agricultural
cooperatives appear to be most successful when they are organized
around a key stage in the production cycle that responds well to
scale or technology--which an institution can provide. .
This is typically in agro-industry, storage, marketing, and key
crops." (D.A.I., 1984:xiv, 56-57). This analysis indicates how
differences in an agri-support factor's characteristics influence
not only a farmer's incentive to join an organization (or engage
in collective action to obtain improved access to and control
over a particular type of agri-support factor) but also the type
of group organized. But the analysis has a major weakness in not
being based on a theoretically-specified model of the necessary
and sufficient conditions for farmers to engage in collective
action to assist group members to obtain improved access to agri-
support factors.




This weakness was addressed by Doherty and Jodha (1979) who,
in a study of adoption of new technology by farmers in India,
used Olson's "logic of collective action" (Olson, 1965; see Sec-
tion 4.1) as a theoretical base for identifying conditions for
group action by farmers. Defining "group action" as "measures
adopted or taken by a group to provide its members with common
benefits," Doherty and Jodha (1979:2) hypothesized seven condi-
tions as essential for group action. A careful reading of their
analysis, however, indicates at least 10 conditions which may be
hypothesized as essential for collective (group) action by
farmers (Box 3).

Box 3. Conditions for Group Action (Adapted from Doherty
and Jodha, 1979).
Group sanction: that utilization of collective action to
seek a common benefit or good through a group-oriented
approach is sanctioned within the society in which a group
is to function--this condition includes such variables as
government policies on and attitudes toward farmer
Group identity: that the actual or potential members of a
group (or organization) see or identify themselves as being
members or potential members of the group by virtue of
having one or more characteristics in common.
Group size: that the group's size (i.e., number of members)
is appropriate relative to the specific benefit or good
provided by the group to its members (e.g., in a marketing
cooperative, the individual member's share of market power
increases as more farmers become members).
Group structure: that the group has an adequate organiza-
tional structure in terms of leadership as well as mana-
gerial, administrative, and financial procedures.
Group good: that it is impossible to exclude any group
member from consuming a group-provided benefit or good, if
one member consumes it.
Organizational good: that the good will not be available to
the group's members unless they (the potential benefici-
aries) organize themselves to provide it.
Functional identity: that a group good functions in the
same way for all of a group's members (e.g., in an irriga-
tion scheme, everyone must receive the same amount of water
per unit area, regardless of farm size).
Divisible good: that it is possible to divide a group good
so that individual members of the group may utilize it.
Individual profit: that individual or collective use of a
group good by group members enables each individual to earn
a minimum acceptable net return after costs and risks are
taken into account.
Compensatory profit: that the level of return enjoyed by a
group member's use of a group good is sufficiently high to
cover not only the required level of individual profit but
also the transaction costs and loss of individual discretion
involved in joining, cooperating with, and supporting the


In their study, Doherty and Jodha (1979:222) analyzed dif-
ferent types of farmer groups, including marketing cooperatives
(for improved access to markets and market information), credit
cooperatives (for improved access to capital), and other types of
groups engaged in such activities as land shaping and construc-
tion and maintenance of surface runoff reservoirs or tanks (for
improved access to land and water). They found that "group good"
and "functional identity" were the most important conditions in
determining whether there was scope for collective action in any
given case. Whether the benefit sought was an "organizational
good" was important in determining whether "the farmers' active
participation or simple acquiescence to an administered plan" was

While noting the importance of profits ["individual prof-
it"], Doherty and Jodha (1979:222) point out that, to generate
group action, the profits had to be divisible among individuals
["divisible good"], and that these individuals had to be compen-
sated by some further increments ["compensatory profit"] for the
loss of discretion experienced in adhering to the group's rules.
The researchers concluded that small groups of 5-15 farmers can
be effective as task groups but that "only larger groups in the
neighbourhood of 100 farmers can maintain the momentum and
enforce the rules necessary to keep up group action on their own
over the long term" (Doherty and Jodha, 1979:222). Finally,
referring to "group sanction," Doherty and Jodha emphasize the
importance of an "ultimate guarantor" such as the government (or
a similarly impersonal institution) that backs collective action
in concept and in practice.

These hypothesized conditions provide a "framework for the
evaluation of the likelihood of successful cooperation by farm-
ers" seeking a common benefit or good; however, there are certain
weaknesses in this framework. Doherty and Jodha did not specify
how certain conditions (e.g., group size) should be interpreted
or specified in terms of function or weight. Olson's "logic of
collective action" suggests that, as group size increases, social
incentives become less effective as motivators and collective
action becomes less likely. Yet Doherty and Jodha distinguish
between "task groups" (5-15 members) and "larger groups" (around
100 farmers) and propose that "only larger groups...can...enforce
the rules necessary to keep up group action...over the long
term." Also, the Doherty and Jodha (1979) framework focused more
on basic or generic conditions for group action and less on how
an agri-support factor's specific characteristics shape or modify
the conditions or structure of incentives for individual farmers
to engage in group action to improve their access to an agri-
support factor.

3.2 Defining the Problem

Given the preceding introduction, definition of concepts,
and review of evidence, the problem may be stated as follows:
What are the organizational processes through which small farmers
are able to realize, or could be able to realize, a more effec-
tive measure of control over the supply of, access to, and/or
proper use of the various agri-support factors essential for
increasing the productivity and income-earning capability of
their agricultural systems? This problem can be broken down and
restated in terms of a number of more specific problems or ques-
tions which are really component parts of the general problem.
Nine such component problems or questions are identified.

Questions 1 through 4 relate to the problem of identifying
the characteristics that determine the structure of incentives
that motivate and facilitate individual farmers to engage in
collective (group) action to support and participate in a farmer

1. Which characteristics of specific agricultural commodities
and agricultural production systems determine the structure
of incentives for and the likelihood of collective action by
farmers to support and participate in a farmer organization?

2. Which characteristics of specific types of agri-support
factors determine the structure of incentives for and the
likelihood of collective action by farmers to support and
participate in a farmer organization?

3. Which characteristics of a farmer organization determine the
structure of incentives for and the likelihood of collective
action by farmers to participate in a farmer organization?

4. Which characteristics of the farmer's social system deter-
mine the structure of incentives for and the likelihood of
collective action by farmers to participate in a farmer

Questions 5 through 7 focus on the extent to which a farmer
organization adequately provides farmers improved access to
essential agri-support factors and the problem of identifying the
conditions under which farmers will take action to improve a
farmer organization's performance capability.

5. Case A--Given the absence of any type of farmer organization
to provide farmers improved access to a particular agri-
support factor:

a. What factors would account for the absence of any such
farmer organization?

b. Under what conditions will farmers engage in collective
action to establish such a farmer organization?

6. Case B--Given the presence of a farmer organization that is
not adequately providing farmers improved access to one or
more agri-support factors:

a. What factors would account for this inadequate level of

b. Under what conditions will farmers engage in collective
action to resolve the problems which exist?

7. Case C--Given the presence of a farmer organization that is
adequately performing the function of providing farmers
improved access to one or more agri-support factors:

a. What factors would account for this adequate level of

b. Under what conditions will farmers engage in collective
action to upgrade the organization's capability to
provide farmers improved access to one or more additional
agri-support factors?

Questions 8 and 9 focus on the extent to which a farmer's
membership in a farmer organization has a positive impact on the
productivity and income-earning capability of the farmer's
agricultural system.

8. Under what conditions does being a member of and partici-
pating in a farmer organization reduce the level of uncer-
tainty and risk which a farmer must face in making farm
management decisions?

9. Under what conditions does being a member of and partici-
pating in a farmer organization have a positive impact on
the productivity and income-earning capability of the
farmer's agricultural system?

Figure 3 summarizes the major variables addressed in this
core set of nine questions.


Figure 3. Major Variables in the Core Set of Nine Questions.

I individual



Farmer Access
to Agri-Support

Agricultural Commodity Uncertainty &
g Risk Levels
Production System I
in-terelad o e e un Agricultural
Agri-Support Factor Productivity

Farmer Organization -
Social System Income

Questions 1-4__ j Questions 5-7 Questions 8-9



=* correspondence of "Local Organization" model (Figure 4 in
Section 4) to the variables in the core set of 9 questions.

4. Identifying Incentives for Collective Action

Improved knowledge of how the variables in Figure 3 are
interrelated would enhance understanding of the incentives
required to motivate individual farmers to support and partici-
pate in a farmer organization. As Section 2 notes, collective
action by farmers to improve their access to an agri-support
factor can be jeopardized by the "free rider," "inconsequentiali-
ty," and/or "organizing" problems. If farmers are to engage
successfully in collective action, sufficient incentive must be
present to overcome or preclude the potential dampening effect
which these three problems can have.

A number of approaches potentially could be applied to
develop a better understanding of the incentives required for
collective action. For example, Esman and Uphoff (1984:295-303)
proposed a model of "local organization" in rural development
(Figure 4). This model identifies intra- and extra-organization-
al elements that affect the contributions of local organizations
to rural development. Intra-organizational elements include
functional and structural variables such as service provision and
membership size, respectively. Extra-organizational elements
include performance (e.g., growth in income), participation
(e.g., resource contributions), exogenous (e.g., governmental
policies), and environmental (e.g., topography) variables.

Figure 4. Model of Variables Affecting the Contributions of Local
Organizations to Rural Development (Adapted from Esman
and Uphoff, 1984:69).




The "local organization" model identifies variables that may
be causally related to dependent variables. However, the model
does not specify the causal relationship between the dependent
variables and the independent variables which, appropriately
manipulated through policy and other interventions, would effect
the desired changes in the specified dependent variable(s). If
research on a given set of independent variables is to produce
results useful for designing policies and programs to support the
development of the potential catalyst role of farmer organiza-
tions, then such research must be developed within a theoretical
framework that provides an ex ante basis for predicting how and
explaining why a particular independent variable is related to a
specified dependent variable.

A potentially useful theoretical framework for this task is
provided by public choice theory (Erickson-Blomquist and Ostrom,
1984; Sproule-Jones, 1982; and Hardin, 1982). We now examine two
approaches to public choice theory: the "logic of collective
action" (Olson, 1965), and the "assurance problem in collective
action" (Runge, 1981, 1983a, 1983b, 1984a, 1984b, n.d.).

4.1 The Olson Perspective

Several studies in a developing country context have applied
public choice theory to natural resource problems (Popkin, 1981;
and Russell and Nicholson, 1981). West's (1983:46) study of the
collective adoption of natural resource conservation practices
usefully points out that "collective adoption implies not just
decision but also collective action [emphasis added]."

Nearly 25 years ago, Mancur Olson (1965) proposed a theory
or "logic of collective action" that deals with situations where
a set of individuals (e.g., farmers) can obtain a potential
benefit (e.g., irrigation) only if they come together into some
form of organization (e.g., a water users association) which has
the ability to provide the benefit to its members. Olson distin-
guishes between two benefit classes--public goods and private
goods. The consumption of a private good can be restricted to
those who have paid for it (e.g., a privately-owned well and
pump), whereas a public good, if available for consumption by one
person, is available to all potential beneficiaries whether or
not they have contributed to making the good available (e.g.,
water along an irrigation canal). Given this distinction, Olson
proposes that public goods may not be provided even if everyone
concerned might actually be better off making the required
contribution and receiving the benefits.

In deciding whether to contribute, the rational, self-
interested individual will consider the expected utility of his
contribution and compare it with alternative uses to which it
might be put. The expected utility of the contribution is not
the benefit the individual will obtain, for if the benefit is a
public good, it by definition will be available to all potential
beneficiaries. What is at issue is not simply the benefit in
question but rather the individual's perception of the difference --
which his (her) contribution will make to the probability that
the benefit actually will be provided. The rational individual,
Olson proposes, reasons as follows: "Since I get the benefit
whether or not I contribute, it is only worth contributing if it
makes a significant difference to the chance of the benefit
indeed being made available."

Thus, the rational individual considers the difference which
that individual's contribution will make to the likelihood that
the benefit will be provided and then multiplies this by the
utility of the benefit in question. This may be illustrated by
the following example. Suppose that it is technically and
economically feasible for farmers to form a produce marketing
cooperative, and that the cooperative can negotiate a contract
with a vegetable processor that would enable the cooperative to
pay each member farmer an increase of $.10 per bushel in the
price of each bushel of cucumbers a member farmer sells to the

cooperative. Suppose, too, that a participation fee (membership
dues) of only $.01 per bushel sold, if paid by all potential
beneficiaries, would finance the cooperative. It would seem that
all the members of the cooperative would thus gain $.09 per
bushel sold. But Olson, to the contrary, would argue that it may
be quite rational for no one to contribute and thus for no price
increase to be obtained.

Olson's argument would be based on the following reasoning.
The farmer, being a rational, self-interested individual, will
consider the difference that his contribution would make to the
likelihood of the benefit being provided and then multiply this
by the utility of the benefit in question. If the farmer thinks
that contributing'$.01 per bushel sold will increase his chances
of getting the $.10 per bushel price increase by only .01, then
the expected utility of the contribution is equivalent to 0.01 x
$.09, or $.0009 (assuming that utility is a linear function of
money). The farmer, under these conditions, clearly would be
better off not paying $.01 per bushel sold and instead spending
the money thereby saved on something else. All the other poten-
tial contributing farmers will reason likewise; none of them will
be willing to pay the $.01 per bushel sold; no cooperative will
be formed; and no price increase will be obtained.

This example illustrates Olson's counterargument to the
conventional wisdom which asserts that individuals with common
interests will join together to further those interests. It will
not be enough, to use a different illustration, for debt-ridden
farmers to realize that, to break the traditional pattern of
farmer dependence on the village moneylender for loans, it is in
their common interest to engage in collective (group) action to
organize themselves into a "savings and loan society". Such
realization will in no way guarantee the occurrence of effective
group action, for each farmer will also realize that an individ-
ual's own efforts will have no noticeable effect on the success
of the organizational effort. Thus, in the absence of special
conditions (to which we shall come shortly), rational, self-
interested farmers will not act to achieve their common or group
interests. More specifically, such farmers will not engage in
collective action to organize a cooperative or any other type of
farmer organization.

But since farmers do sometimes support and participate in a
wide variety of farmer organizations (Fig. 1), one must conclude
or at least one may hypothesize that such farmer support of farm-
er organizations depends on a set of special conditions. As a
basis for identifying such special conditions, Olson proposes the
concept of a latent group. When we think about a large number of
farmers (e.g., the population of small farmers in a village), we
are thinking about what Olson would call a "latent group." Such
a large number of farmers is called a "latent group" because the
individual farmers have a potential (latent) capacity for action.

Olson identifies the concept of a "selective incentive" as
the first special condition. Such an incentive is selective in
that it can distinguish between those individuals who support
collective (group) action in the common interest and those who do
not. The potential or latent capacity for individuals in a
latent group to engage in collective action can be realized or
"mobilized" only with the aid of selective incentives. Olson
(1965:51) writes: "Only a separate and 'selective' incentive
will stimulate a rational individual in a latent group to act in
a group-oriented way." Under such circumstances,

group action can be obtained only through an incentive that
operates, not indiscriminately, like the collective good,
upon the group as a whole, but rather selectively towards
the individuals in the group. The incentive must be
"selective" so that those who do not join the organization
working for the group's interest, or in other ways
contribute to the attainment of the group's interest, can be
treated differently than those who do (Olson, 1965:51).

Olson next examines the type of selective incentive required
to ensure that individuals in a latent group support and partici-
pate in collective action. Here Olson focuses on "social incen-
tives" as distinct from "economic incentives." A "social incen-
tive" is a positive or negative non-monetary sanction that can
either reward those who act in the group interest or punish those
failing to bear an allocative share of the costs of group action.
As Olson (1965: 60) states, "everyday observation reveals that
most people value the fellowship of friends and associates, and
value social status, personal prestige, and self-esteem."

Because it is the group which decides whether to give an
"individual, noncollective good" to a person, and because most
persons normally value this type of good, the group is in a stra-
tegic position to use "social incentives" (positive and negative
sanctions) to encourage an individual to do his (her) part toward
achieving the group's goal. Because social incentives, adminis-
tered by the group's leaders or members, distinguish among indi-
viduals, such social incentives are a type of selective incentive
that may be used to mobilize a latent group or segments thereof
to support and participate in collective (group) action.

A second condition for stimulating individuals to act in a
group-oriented way to achieve a public good is that the number of
individuals needed to provide it be relatively small. Generally,
social incentives are effective only in groups sufficiently small
that groups members have face-to-face contact with one another.
Such contact, as one observer notes, has been one key to the
success of the new irrigation (water users) groups in Sri Lanka's
Gal Oya water management project, where the farmers are organized
around a "turn-out" (the irrigation system's lowest level).

The condition that a group be "relatively small" for effec-
tive collective action is based on four considerations. First,
"in any large.group everyone cannot possibly know everyone else,
and the group will ipso facto not be a friendship group; so a
person will ordinarily not be affected socially if he fails to
make sacrifices on behalf of his group's goals" (Olson, 1965:62).
By contrast, in a small group, it is difficult for an individual
not to be affected socially if he fails to make sacrifices on
behalf of the group's goals. Second, in a relatively small
group, each individual's contribution makes a much larger impact
on the likelihood of the benefit being provided. Third, in a
relatively smaller group, there is greater scope for communica-
tion and strategic bargaining. Fourth, the greater the number of
individuals involved in organizing a group, the greater the
organizational costs will be and the less likely that any one
individual could be found who would be prepared to act as

Olson concludes that the larger the number of individuals
required to participate if a public good is to be provided, the
less likely they are to do so. It is not assumed or expected
that individuals in a large or latent group will organize for
coordinated action merely because, as a group, they have a reason
(e.g., an economic incentive) for doing so, or that social incen-
tives alone will provide sufficient motivation for the individ-
uals in a latent group to engage in collective (group) action to
obtain a collective goal.

Olson's theoretical framework assumes rationality and self-
interest, that the rational, self-interested individual considers
the expected utility of his contribution to the provision of a
public good and compares it with alternative uses. It is there-
fore not unreasonable to conclude, as in the case of the Alphan
farmer, that the farmer, as a rational, self-interested individ-
ual, will not contribute $.01 per bushel sold to a cooperative
that would obtain a $.10 per bushel price increase, if he per-
ceives that there is only a slim (e.g., a .01 or one in a hun-
dred) chance that the $.01 will make all the difference between
the cooperative being a success or it being a failure. The
farmer is better off keeping his money and spending it on some-
thing else; and the cooperative accordingly will not be formed.

This example, however, is based on rather arbitrary assump-
tions about the contribution's required size ($.01), the prob-
ability (0.01) that it will make a difference, and the size
($.09) of the ultimate benefit. If these assumptions are not
made or, more generally, if one cannot specify on an a prior
basis the type and level of "selective incentive" that will be
necessary to stimulate a farmer to act in a group-oriented way,
then one cannot readily predict whether the cooperative or, more
generally, a farmer organization will be formed. Therefore, it
becomes clear that ability to predict whether a farmer will

contribute to a group-oriented effort or, more specifically, will
engage in collective action to support a farmer organization
depends on ability to identify some type and level of "selective
incentive" sufficient to motivate potential member farmers to
provide the essential level of support or contribution.

One approach to developing an ability to predict what would
constitute a sufficient incentive for collective action is pro-
vided by Mitchell's (1979) "revised rational choice theoretical
paradigm of collective action" (Box 4). The paradigm identifies
factors determining an individual's level of incentive to engage
in collective action. These factors include the costs of contri-
buting, the benefits of contributing, and the costs of not con-
tributing. In the case of farmer organizations in the LDCs, we
are interested in the potential role which such organizations can
play in increasing the level of incentive current or potential
member farmers have to participate in and support the collective
action required.

Box 4. Factors Influencing Farmers to Contribute to a
a Farmer Organization That Seeks a Group Good
(Adapted from Mitchell, 1979:100).
I. Costs of Contributing
A. Money
B. Time
C. Loss of social status and reputation
II. Benefits of Contributing
A. Possible increase in a group good
1. Utility of the group good for the farmer
2. Amount of the group good the farmer
expects to receive personally
3. Perceived effectiveness of collective
action to achieve the group good
B. Receipt of selective goods
1. Goods and services
2. Sociability
3. Social status
4. Self-esteem
III. Costs of Not Contributing
A. Possible continuance of or increase in a group bad
1. Disutility of the group bad for the farmer
2. Amount of the group bad the farmer has
received or expects to receive personally
3. Perceived effectiveness of farmer contri-
butions in preventing the bad
B. Receipt of selective bads
1. Loss of goods and services
2. Reduced social status
3. Guilt

The communication of information about the factors listed in
Box 4 can play an important role in determining how potential and
actual group members perceive these factors. For example, Moe
(1977) found that small contributions made by members of a group
become more understandable if one does not assume that potential
contributors have perfect knowledge of the marginal costs, mar-
ginal benefits, and the prevailing supply of relevant group goods
(or bads) when they contemplate making a contribution. Thus, if
potential members underestimate marginal costs and/or overesti-
mate benefits, the likelihood is increased that such individuals
will contribute to the group. Mitchell (1979:120-121) suggests
that groups working to achieve a group good may more effectively
utilize communication to change potential contributors' percep-
tions of the potential benefits of contribution (and the poten-
tial costs of not contributing) than perceptions of the costs of

4.2 The Runge Perspective

Olson's "logic of collective action" implies that farmers
will not support a group-oriented effort (e.g., a farmer organi-
zation) without some form of "selective incentive." Since there
are numerous instances where farmer organizations exist without
any apparent provision of a selective incentive, one must ask
what other factors) may account for such organizations. Runge's
(1984b) answer to this question is that the organizational
problem is not really one of providing selective incentives but
rather one of providing potential contributors with "assurance"
that each and every individual will contribute his (her) fair
share toward ensuring that the organization in question will be
able to provide the desired group good.

Returning to the cucumber marketing example, the individual
Alphan farmer should know that his (her) real choice is not
between (a) supporting a marketing cooperative and getting the
benefit of organization and (b) not supporting the cooperative
but yet getting a benefit which others foolishly make possible
for him. Rather the farmer's real choice is between (a) and (c)
not supporting the organization and not getting the benefit, with
the likelihood of (a) subject to the level of assurance which
individual farmers have that all other farmers will provide the
necessary support required for the group to be able to provide
the desired benefit (Kimber, 1981).

Runge (1983a:1) identifies institutions as key providers of
information that reduces uncertainty over and provides assurance
as to the expected actions of others. An institution is any pub-
lic system of rules or "rules of the game" that provides assur-
ance of others' actions, thereby "making possible greater cooper-
ation and coordinated action" in a particular choice environment

(Runge, n.d.:l-2). Institutions provide for certain penalties
and defenses when violations occur. "Operationally, institutions
guide the behavior of people with respect to each other, and to
their own and others' belongings, possessions, and property"
(Runge, n.d.:l).

Note, for example, the reliance in developing countries on
common property arrangements, in which a community's members have
certain rights to be included in access to resources. Institu-
tions sharing this inclusivity property reduce uncertainty and
risk in environments where a low subsistence level from season to
season makes life chances precarious and places a premium on an
institution like common property. "Where weather and natural
calamity dominate the pattern of life, the assurance that mis-
fortune will not lead to certain death is provided by social
institutions which spread risks by means of the right to be
included" (Runge, 1983a:9).

Institutions, in Peter Dorner's words, "assure a degree of
security with respect to the accepted procedures of human inter-
action and response" (cited in Runge, n.d.:l). By playing a role
in setting expectations, institutions have the impact of "confer-
ring expected value to the stream of future benefits associated
with human activity. By defining rights and privileges, respon-
sibilities and obligations, ...institutions fix people's expecta-
tions of the future" (Runge, n.d.:l). In short, institutions
serve to reduce uncertainty and increase the potential for
cooperation and coordinated action. Depending on the circum-
stances, institutions may act as obstacles to change or "may
facilitate change by altering people's views of the likely
consequences of certain actions" (Runge, n.d.:2).

In Olson's perspective, free rider behavior is inevitable in
the absence of selective incentives, regardless of what others
are expected to do. Expectations about or uncertainty over the
behavior of others are irrelevant. Even if all agree not to take
a free ride, there is no incentive to keep the agreement; it will
eventually be broken and free rider behavior will resume. The
Runge perspective brings the Olson perspective into question by
asking: Is it reasonable that farmers would decide whether or
not to join a cooperative regardless of what they expect other
farmers to do? It seems more reasonable to argue that farmers
considering whether to join a cooperative will base their deci-
sion at least in part on what each farmer expects of other
farmers. The problem each farmer faces is uncertainty about
other farmers' actions.

The idea that an individualistic behavior pattern (i.e.,
free rider behavior) as distinct from other behavioral alter-
natives (e.g., individual support of collective action) is always
and necessarily the dominant behavior pattern also may be ques-
tioned. Consider an environment characterized by mutual inter-

dependence, in which expectations of others' behavior are
relevant. For example, if each Alphan farmer expects other
farmers to graze as many sheep as possible on a small pasture,
each may decide tograze as many sheep as possible, completely
disregarding the potential harm overgrazing could bring to the
community and individual farmers. But an individual farmer's
decision to graze as many sheep as possible is not one that has
been made on a purely individualistic basis with complete dis-
regard to expectations of other actors' behavior. If each farmer
were assured by some institution (rule or custom) that the other
farmers would not attempt to take a free ride by overgrazing
beyond a set individual limit, each farmer is likely to restrict
grazing of sheep to the assigned limit.

This example highlights just how important expectations and
the problem of uncertainty are to the farmer. It also helps to
point out that selective incentives (e.g., "enforcement from
above") are potentially not the sole or primary factor that can
influence individual behavior. Indeed, selective incentives may
be far less important than agreement developed within a village
or community through a process of mutual accommodation. In such
a setting, a farmer organization may arise from the "bottom up"
(e.g., spontaneous, "grass roots" initiatives) as the result of
local rules and customs.

The possibility of individual farmers' expectations of
others' actions being made with greater confidence is enhanced if
some existing rule of behavior permits farmers' actions to be
predicted with accuracy (Ullman-Margalit, 1977). Runge (1983a:5)
refers to a variety of institutions, in different biophysical and
cultural environments, that provide assurance and reduce uncer-
tainty over others' actions. All too frequently, however, exist-
ing institutions remain unrecognized as development practitioners
seek to implement what they see as rational solutions to what
they have identified as the problems. Thus, an existing institu-
tion will more likely be identified on a post hoc basis as a
constraint to adoption of a development initiative (e.g., tech-
nology transfer) than on an ex ante basis that would permit the
initiative to be designed in such a way that the institution
supports rather than impedes the initiative in question.

In contrast to Olson's perspective, Runge's description of
the "assurance problem" (AP) provides an alternative approach to
identifying the determinants of collective action. Unlike the
Olson perspective, the AP does not assume either free rider
behavior as a predominant behavioral pattern or an inevitable
dependency on or primacy of selective incentives as a necessary
condition for collective action. Nor is it assumed that farmer
organizations can only be fostered through top-down imposition
and enforcement of new and essentially alien institutions.

As described by the AP, whether farmers will support a
farmer organization or attempt to free ride the benefits (group
goods) made available by such an organization depends on expecta-
tions. Where a farmer has assurance that other farmers will draw
no more water than each farmer's fair share, the possibility of
free rider behavior (drawing more water from the irrigation canal
than one's allotted share) will not be a dominant strategy for
any farmer.

In the face of uncertainty, expectations formed and informed
by appropriate institutions (rules) can make a difference. Where
there is rampant free riding on a group good (e.g., nonrepayment
of production loans drawn on a credit line exclusively for small
farmer credit groups), this behavior may stem from expectations
that are either uncoordinated by existing "rules of the game" or
coordinated by prevailing norms which lead individuals to expect
nonrepayment by others, leading each to take out a loan before
the credit line closes (or similarly to draw more water before
the canal runs dry).

The AP suggests that what is necessary is to develop a basis
for mutual accommodation and consent based on a set of rules. The
basis for this may lie in existing rules or potentially in rules
which can be negotiated by the group.

The lesson of the assurance game is to let individuals...
innovate self-binding rules which best serve their needs
before enforcing rules from outside. Rules will be better
suited to the needs of the group...and more likely to
succeed.... These rules may come in many shapes and forms,
not all of which are familiar. The institutional oppor-
tunity set of solutions...is much wider than we think...
(Runge, 1981:603-604).

Even if old institutions present apparent obstacles for estab-
lishing farmer organizations, there may be elements of these
institutions (rules) which, having certain strengths, are worth
preserving or adapting, or which can serve as the basis for
adding new rules. In terms of Runge's perspective (i.e., the
AP), there are potentially various institutional forms, depending
on the history, traditions, and biophysical resources of the
group involved, that can successfully coordinate expectations.
The problem for applied and adaptive research is finding the
appropriate institutions for a particular situation.

The "free rider," "inconsequentiality," and "organizing"
situations are problems of uncertainty and expectations. Solving
them entails, in part, providing farmers with information on
relative productivity and income-earning capability under current
institutional relationships (without access to agri-support
factors) as compared with alternative institutional arrangements
(with access to agri-support factors through collective action by

farmers). This information can provide farmers with a less
uncertain (more informed) basis for making allocational decisions
about which types of organizations to support.

Further, technical information must be complemented by know-
ledge of the capacity of local institutions to regulate farmer
organizations and their members. Relevant here is information
about the attitudes farmers have about organizations, joining and
being a member of such organizations, abiding by the rules of
these organizations, farmer confidence in existing or proposed
mechanisms for redressing wrongdoing, etc.

To attempt a pure "top-down" approach, seeking to impose
organizations on villagers, runs the risk that potentially valu-
able information contained in local institutions (rules) will be
ignored. Of course, "the best sources of this information are
the people themselves, with whom consultation can provide an
understanding of the institutions most compatible with tech-
nically efficient resource management" (Runge, 1983a:8). This
implies a dual role for donors, governments, and technical assis-
tance agencies that seek to promote greater utilization of farmer

First, practitioners of farmer organization development,
must approach the issues of farmer productivity and income-
earning capability with sufficient technical expertise to be able
to assure the affected--the farmers--that the information in
question (e.g., income-increasing potential of a particular type
of farmer organization) is sound. Second, they must approach
allocational and distributional issues with sufficient attention
to local definitions of fairness so that they are able to gain a
locality-based constituency.

The second implication follows from the first. Not only
consultation with local people but also guidance from existing
local institutions is essential for the successful identification
of institutional as well as technical innovations. Such utiliza-
tion of and reliance upon locally available knowledge is increas-
ingly being exploited by anthropologists studying "indigenous
knowledge systems" (Chambers, 1983; Warren, 1984) and by farming
systems research practitioners (Whyte and Boynton, 1983).

5. Potential Areas for Applied Research

This paper has highlighted the potential catalyst role which
farmer organizations can play in representing and advancing the
economic interests of their members. This role entails helping
an organization's member farmers gain improved access to agri-
support factors essential to increasing the productivity and
income-earning capability of their agricultural systems. Avail-

able evidence suggests that this potential catalyst role is not
currently being adequately realized and that little, in a system-
atic way, is known about the nature of the incentive structures)
required to foster development of farmer organizations that can
play this catalyst role effectively. Thus, there is an identi-
fied need for an applied research initiative (project) that would
develop an analytical capability to determine: (1) the types of
farmer organizations that can play a catalyst role in helping
their member farmers gain improved access to essential agri-
support factors; (2) the types of incentive structures that will
foster the development of farmer organizations that can play this
catalyst role effectively; and (3) guidelines for designing
interventions (policies, reforms, methods, etc.) that will be
effective in creating incentive structures that support the
development of farmer organizations.

The objective of this proposed research would be the devel-
opment of an assessment methodology that would enable a user, in
a given developing country, to identify: (1) the current status
and development potential for farmer organizations in that coun-
try; (2) the conditions which must yet be established in that
country to create an incentive structure that will foster the
development of existing and potential farmer organizations; and
(3) the interventions that could be used to establish an incen-
tive structure that would foster the development of existing and
potential farmer organizations.

The two public choice perspectives reviewed in Section 4
provide a starting point for designing applied research on the
potential catalyst role of farmer organizations. While space
limitations preclude developing more fully the many implications
these perspectives have for designing such an applied research
initiative, the present section outlines how the identified need
for research on farmer organizations in the developing countries
could be addressed within a public choice perspective.

The remainder of this section focuses on the concept of an
"incentive structure" and the need for further empirical work to
define more adequately the nature of the "incentive structure"
required to motivate individual farmers to engage in collective
action. Applied research to develop the "incentive structure"
concept as a diagnostic and analytical tool would provide the key
ingredient for developing an assessment methodology that could be
used by practitioners in a developing country to identify the
current status of and development potential for farmer organiza-
tions in that country.

The key element in the earlier reviewed perspectives on
public choice theory is the central role which an incentive or,
more specifically, an "incentive structure" (see Fig. 5) plays in
stimulating or motivating individuals to act in a group-oriented
way, where a group seeks via collective action to obtain a group

good for its members. Of course, the concept of "incentive" may
be interpreted as a "selective incentive" or as an "institution"
which provides a solution to the "assurance problem". Now, the
preceding section's review of the Olson and Runge perspectives
may have left the reader asking which is more important for
supporting the development of farmer organizations--"selective
incentives" or "institutions"? Also, the reader may be asking
whether the development of farmer organizations ultimately
depends on the initiative of farmers themselves or whether there
is a role which donors, technical assistance agencies, and LDC
governments can play in initiating and assisting in action to
support the development of such organizations. The way out of
this quandary is to recognize:

the problem as one of assurance, which can be provided by a
combination of bottom-up organization and top-down assis-
tance.... The problem is to find the right combination...
and rethink the conditions necessary for a combination of
bottom-up and top-down activities to generate farmer
organizations (C. Ford Runge, personal communication).4

A primary research issue is the need to specify exactly what
kinds of conditions, including both "institutions" and "selective
incentives," are required in an "incentive structure" to ensure,
where a group seeks to obtain a group good, that each group mem-
ber will act in a group-oriented way in support of the group's
collective action effort. In focusing on a particular type of
group, namely, farmer organizations, we are concerned with iden-
tifying the "incentive structure" required to ensure that such
organizations will have a favorable environment in which to
develop a "performance capability" to represent and advance the
economic interests of the member farmers, that is, to provide
their members improved access to agri-support factors.

The type of "incentive structure" involved in supporting
collective action to develop a farmer organization entails three
distinct levels of incentive (see Figures 1 and 3), as follows:

4Brubaker (1975:158] has stated that the "opportunities for
eliciting more nearly voluntary economic expression of individual
priorities for collective goods may be far greater than most of the
contemporary orthodox literature suggests. If so, it may be emi-
nently worthwhile to explore more carefully means to expand the
scope of voluntary arrangements for provision of collective needs
while perhaps in some measure of correspondence reducing reliance
on coercive institutions with their own potentially detrimental

Level 1: Those incentives required to motivate individual
action by a farmer to adjust his production implementation
decisions in a manner that permits the farmer to take
advantage of available agri-support factors, and thereby, to
increase the productivity and income-earning capability of
the farmer's agricultural system.

Level 2: Those incentives required to motivate individual
farmers to engage in collective action to establish a farmer
group or organization, with the objective of utilizing the
group as a means to improve the access of the group's member
farmers to a particular type of group good, namely, agri-
support factors that would otherwise not be available to
these farmers on an individual basis.

Level 3: Those incentives required to ensure that a farmer
organization can develop and maintain a self-sustaining
performance capability to provide its members with improved
access to agri-support factors.

The utility of studying the implications that agricultural
commodity and production system characteristics have for the
local organizational forms best suited to improve the farmer's
access to agri-support factors has been suggested in several
studies (e.g., Montgomery, 1983; Uphoff and Van Dusen, 1984).
These implications could be clarified by examining (1) the extent
to which the characteristics of agricultural commodities, agri-
cultural production systems, agri-support factors, farmer organi-
zations, and social systems combine to determine specific types
of incentive structures; and (2) the extent to which a given type
of incentive structure motivates individual farmers to engage in
collective action, via a farmer organization, to improve their
access to specific agri-support factors.

A possible hypothesis may be stated as follows:

1. That the degree of Level 1 incentive for individual action
(e.g., a change in a farmer's crop fertilization practices)
to take advantage of the availability of an agri-support
factor (e.g., fertilizer) depends, in part, on two key
characteristics of this factor vis-a-vis the agricultural
production system in which this input is used to produce an
agricultural commodity; these characteristics are individual
profit and divisible good.

2. That the degree of Level 2 incentive for collective action
depends, in part, on four key characteristics of a farmer
organization (e.g., a credit group) as a group-oriented
means to assist individual farmers to obtain improved access
to an agri-support factor (e.g., credit); these character-
istics are compensatory profit, functional identity,
organizational good, and group good.

3. The the degree of Level 3 incentive for performance
capability of a farmer organization depends, in part, on
four key characteristics of that organization, as a social
system; these characteristics are group structure, group
size, group identity, and group sanction.

In this hypothesis, the effective realization of the poten-
tial catalyst role of a farmer organization depends on the right
mix (level, type, and quantity) of incentives being in place in a
given agricultural situation. In this mix, certain characteris-
tics may be more important in establishing an incentive for cer-
tain types of behavior (e.g., production implementation deci-
sions) than for other types of behavior (e.g., collective action
or performance capability).

Thus, it becomes important to be able to identify precisely
which variables (characteristics) create an incentive for which
kinds of behavioral outcomes. The ability to identify precisely
which incentives are effective in achieving a desired behavioral
objective (e.g., individual farmer support of a group-oriented
effort to provide group members with improved access to agri-
support factors) is essential if we are to be able to identify
the incentive structure that would most effectively motivate and
facilitate collective action by farmers.

How can we begin to identify, for each hypothesized incen-
tive level, the incentives required to motivate the required
behavior on the part of individual farmers? The problem can be
broken down into three distinct yet interrelated components. As
in the hypothesis above, the first component may be termed as
individual action (Fig. 3). The required research would focus on
whether farmers, given the availability of a particular agri-
support factor (e.g., a market in which farmers can get a higher
price for their produce), have sufficient incentive--in terms of
perceived benefits, costs, and risks--to change their existing
production implementation decisions in such a way that they will
be able to take advantage of the availability of the particular
agri-support factor.

Relevant research would entail applying traditional agricul-
tural economic models (e.g., partial budgeting) to estimate the
level of individual profit farmers would require to justify
making the required adjustments in their traditional production
implementation decisions. This type of model may be extended to
include procedures for identifying existing natural or social
constraints that reduce the incentive a farmer has to make the
adjustments required to take advantage of the specific agri-
support factors under evaluation. Knowledge of the existing
conditions (e.g., moisture stress or negative attitudes about
buying inputs such as fertilizer on credit) that operate as
constraints to a change in the farmer's behavior (individual
action) is a first step toward being able to identify an effec-

tive strategy for removing or relaxing the constraining impact
which such conditions have on the farmer's ability to take
individual action to change production implementation decisions.

As earlier noted, the demand for any agri-support factor is
a derived demand; thus, if farmers perceive that making adjust-
ments in their production decisions, to take advantage of an
agri-support factor's availability, will not likely result in any
increase in benefit (or reduction in cost or risk), then such
farmers will certainly not have any incentive, when this agri-
support factor is not available, to incur the additional costs
and risks involved in seeking, through collective action (e.g.,
farmer organization), to improve their access to this agri--
support factor.

In this sense, we can say not only that the demand for an
agri-support factor is a derived demand but also that this same
principle applies in collective action. In other words, the
greater the derived demand for a particular agri-support factor,
the greater the derived demand for a group-oriented effort to
provide member farmers with improved access to this agri-support
factor. In other words, the utility of a farmer organization to
actual and potential member farmers lies in the organization's
performance capability to help its member farmers gain improved
access to those agri-support factors for which the derived demand
is the greatest.

A key starting point for research on farmer organizations in
the developing countries is to do the agricultural economic and
constraints analyses outlined above; such analyses will provide a
more informed basis as regards which agri-support factors, if
more readily available to farmers, would provide the best oppor-
tunities to increase their productivity and income-earning capa-
bility. This analysis would indicate what additional character-
istics (e.g., institutions), beyond increased individual profit
or a reduction in cost or risk, impact on the incentive a farmer
has to adjust his (her) agricultural system in a way that allows
the farmer to take advantage of a particular agri-support factor,
if indeed this factor can be made available to farmers through
collective action.

The second component, collective action (Fig. 3), entails an
analysis of the incentives effective in motivating individual
farmers in a group to engage in and support collective action to
obtain some common benefit or good (e.g., improved access to an
agri-support factor). Here the research would build, test, and
validate models that specify the necessary and sufficient condi-
tions (variables) for individual farmers to engage in collective
action. The conditions for group action identified in Doherty
and Jodha (1979)--compensatory profit, functional identity, or-
ganizational good, and group good--provide an initial hypothesis
of variables that may be key determinants of whether a group-

oriented effort (e.g., a farmer organization) is likely to be

Models incorporating these variables could be systematical-
ly evaluated in a variety of empirical contexts. Other variables
that potentially may need to be included in these models are spe-
cified in Mitchell's revised rational choice theoretical paradigm
(Box 4) and Esman and Uphoff's (1984) local organization model.

In developing a collective action model, attention should be
addressed to identifying the mix of incentives qua institutions
which, if present, provide assurance of the expected actions of
an organization's actual and/or potential members. Do the insti-
tutions in a farming community have the impact of preventing the
free rider, inconsequentiality, or organizing problems from
precluding collective action by farmers? Here there is a need to
develop a conceptual framework that could be used to identify the
relevant institutions in a farming community.

The third component, performance capabilty (Fig. 3), entails
an analysis of the variables influencing the capability of a
farmer organization to provide member farmers with improved
access to agri-support factors. There is a need to focus on
specifying and testing models of a farmer organization's perform-
ance capability. Based on the Doherty and Jodha (1979) study,
conditions or variables influencing performance capability would
likely include group structure, group size, group identity, and
group sanction.

The models previously mentioned, as well as others in the
literature, could assist in specifying an empirically testable
model as regards the types of variables determining an organiza-
tion's performance capability. However, the dependent variable
in this model would focus on indicators of the characteristics of
a farmer organization and not on individual farmer behavior (as
would be the case in models of individual action and collective
action). One performance capability indicator could be whether
the organization has been able to establish a capability to
market the produce of the members at a higher price than that
which nonmembers obtain. Table 1's summary of the potential
tasks of an organization is suggestive of other potential areas
in which indicators of an organization's performance capability
could be developed.

SByrnes (1986) provides a proposed agenda for an applied
research initiative (project), based on the incentive structure
concept, to develop an assessment methodology to determine the
current status and development potential for farmer organizations
in the developing countries.


1984 A.I.D. Policy Paper: Local Organizations in Development. U.S.
Agency for International Development, Washington, D.C. 20523.

n.d. Blueprint for Development: The Strategic Plan of the Agency
for International Development. U.S. Agency for International
Development, Washington, D.C. 20523.

Bratton, Michael
1963 "Farmer Organizations in the Communal Areas of Zimbabwe:
Preliminary Findings." An unpublished report.

1985 "A Share of the Plow: Small Farmer Organizations and Food
Production in Zimbabwe." Department of Political Science and
African Studies Center, Michigan State University, East
Lansing, Michigan.

Brubaker, Earl R.
1975 "Free Ride, Free Revelation, or Golden Rule?" The Journal of
Law and Economics, 18:147-161.

Byrnes, Kerry J.
1986 The Potential Role of Farmer Organizations in Increasing the
Productivity and Income-Earning Capability of Small-Farmer
Agricultural Systems in the Developing Countries: A Concept
Paper, prepared for the Office of Rural Development, Bureau
for Science and Technology, Agency for International

Chambers, Robert
1983 Rural Development: Putting the Last First. London: Longmans.

Cernea, Michael
1981 "Modernization and Development Potential of Traditional
Grassroots.Peasant Organizations," pp. 121-139 In Directions
of Change: Modernization Theory, Research and Realities, ed.
M.O. Attir, B. Holzner, and Z. Suda. Boulder, Colorado:
Westview Press.

1987 "Farmer Organizations and Institution Building for
Sustainable Development," Regional Development Dialogue,
8(2)(Summer) :1-19.

Doherty, Victor S. and N.S. Jodha
1979 "Conditions for Group Action among Farmers," pp. 207-223 in
Group Farming in Asia: Experiences and Potentials. Singapore:
Singapore University Press.

Erickson-Blomquist, William and Elinor Ostrom
1984 "Institutional Capacity and the Resolution of a Commons
Dilemma," Workshop in Political Theory & Policy Analysis,
Indiana University, 513 North Park, Bloomington, Indiana

Esman, Milton J. and Norman T. Uphoff
1984 Local organizations: Intermediaries in Rural Development.
Ithaca: Cornell University Press.

Galjart, Benno and Dieke Buijs
1982 Participation of the Poor in Development. Leiden Development
Studies No. 2. Institute of Cultural and Social Studies,
University of Leiden.

Grabowski, Richard
1981 "Induced Innovation, Green Revolution, and Income Distribu-
tion: Comment," Economic Development and Cultural Change,

Hamilton Task Force
1989 Presentation of the Task Force on Foreign Assistance to the
House Foreign Affairs Committee, February 1, 1989.

Hardin, Russell
1982 Collective Action. Baltimore, Maryland: Johns Hopkins
University Press.

Hayami, Yujiro and Vernon W. Ruttan
1984 "The Green Revolution: Inducement and Distribution," The
Pakistan Development Review, 23(1)(Spring):37-63.

Johnston, Bruce F. and Thomas P. Tomich
1984 "Feasibility of Small Farm Development Strategies," a report
submitted to the Bureau of Policy & Program Coordination,
Agency for International Development, Washington, D.C. 20523.

Johnston, Bruce F. and William C. Clark
1982 Redesigning Rural Development: A Strategic Perspective.
Baltimore: Johns Hopkins University Press.

Kimber, Richard
1981 "Collective Action and the Fallacy of the Liberal Fallacy,"
World Politics, 33:178-196.

Kottak, Conrad Phillip
1985 "When People Don't Come First: Some Sociological Lessons," in
Michael M. Cernea (ed.), Putting People First: Sociological
Variables of Rural Development, Oxford University Press.

Lele, Uma
1981 "Cooperatives and the Poor: A Comparative Perspective," World
Development, 9:55-72.

McInerney, John P.
1978 The Technology of Rural Development, Staff Working Paper No.
295, Washington, D.C.: The World Bank.

Mitchell, Robert Cameron
1979 "National Environmental Lobbies and the Apparent Illogic of
Collective Action," pp 87-121 in Collective Decision Making:
Applications from Public Choice Theory, ed. Clifford S.
Russell. Baltimore: The Johns Hopkins University Press.

Montgomery, John D.
1983 "When Local Participation Helps," Journal of Policy Analysis
and Management, 3(1):90-105.

Morss, Elliott R., John K. Hatch, Donald R. Mickelwait, and Charles F.
1967 Strategies for Small Farmer Development, 2 vols. Boulder,
Colorado: Westview Press.

Nesman, Edgar G.
1981 Peasant Mobilization and Rural Development, Cambridge,
Massachusetts, Schenkman Publishing Company, Inc.

Obern, Catherine C. and Steven D. Jones
1981 "Critical Factors Affecting Agricultural Production Coopera-
tives: A Review," Annals of Public and Co-operative Economy,

Office of Technology Assessment
1988 Enhancing Agriculture in Africa. A Role for U.S. Development
Assistance. OTA-F-357. Congress of the United States: Office
of Technology Assessment.

Olson, Mancur
1965 The Logic of Collective Action: Public Goods and the Theory
of Groups. Cambridge: Harvard University Press.

Oxby, Clare
1983 "'Farmer Groups' in Rural Areas of the Third World,"
Community Development Journal, 18(1):50-59.

1981 State of the Art: Farmers' Organization Research, Socio-
Economic Series No. 1, Philippine Council for Agriculture and
Resources Research.

Popkin, Samuel
1981 "Public Choice and Rural Development: Free Riders, Lemons and
Institutional Design," pp. 43-80 in Russell and Nicholson

Runge, Carlisle Ford
1981 "Common Property Externalities: Isolation, Assurance and
Resource Depletion in a Traditional Grazing Context,"
American Journal of Agricultural Economics, 63:595-606.

1983a "The 'Tragedy of the Commons' and Resource Management," a
paper for the Ministry of Local Government and Lands,
Gabarone, Botswana.

1983b "Sources of Institutional Innovation: An Intepretative
Essay," Discussion Paper No. 176., Center for Economic
Research, Department of Economics, University of Minnesota,
Minneapolis, Minnesota 55455.

1984a "Common Property and Collective Action in Economic Develop-
ment," Working Paper No. 1, Center for Natural Resource
Policy and Management Studies, University of Minnesota, St.
Paul, Minnesota 55108.

1984b "Institutions and the Free Rider: The Assurance Problem in
Collective Action," The Journal of Politics, 46:154-181.

n.d. "Learning about Institutional Innovation and Technology
Transfer: The Role of the Agency," a paper for the Agency for
International Development.

Russell, Clifford and Norman Nicholson (eds.)
1981 Public Choice and Rural Development. Baltimore: Johns Hopkins
University Press.

Ruttan, Vernon W.
1978 "A Postscript on Alternative Paths of Induced Institutional
Change," Chapter 14 in Binswanger, H.P., V.W. Ruttan, et al.,
Induced Innovation: Technology, Institutions and Development.
Baltimore: Johns Hopkins University Press.

Ruttan, Vernon W. and Yujiro Hayami
1984 "Toward a Theory of Induced Institutional Innovation," The
Journal of Development Studies, 20(4)(July):203-223.

Sproule-Jones, Mark
1982 "Public Choice Theory and Natural Resources: Methodological
Explication and Critique," The American Political Science
Review, 76(4):790-804.

Torgerson, Randall E.
1977 "Farmer Cooperatives," The Annals of the American Academy of
Political and Social Sciences, 429(January):91-102.

Ullman-Margalit, Edna
1977 "Coordination Norms & Social Choice," Erkenntnis, 11:143-155.

Uphoff, Norman and Katy Van Dusen
1984 Local Institutional Development for Agriculture, Special
Series on Local Institutional Development--No. 5, Rural
Development Committee, Cornell University.

Warren, Dennis M.
1984 "The Role of Indigenous Knowledge Systems and Indigenous
Organizations in Facilitating a Participatory Approach to
Agricultural and Rural Development," International Seminar on
the Role of Rural Organizations in the Process of Develop-
ment, University Centre of Dschang, Cameroon, July 15-20.

West, Patrick C.
1983 "Collective Adoption of Natural Resource Practices in
Developing Nations," Rural Sociology (48:1):44-59.

Whyte, William F. and Damon Boynton (ed.)
1983 Higher Yielding Human Systems for Agriculture. Ithaca, New
York: Cornell University Pron.

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs