<%BANNER%>

Cooperative Marketing and Transaction Costs Economics

Permanent Link: http://ufdc.ufl.edu/UFE0042207/00001

Material Information

Title: Cooperative Marketing and Transaction Costs Economics A Study of Bwanje Valley Rice Farmers in Malawi
Physical Description: 1 online resource (102 p.)
Language: english
Creator: Nyirenda, Lucy
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: cooperatives, farmers, malawi, marketing, probit, tce
Food and Resource Economics -- Dissertations, Academic -- UF
Genre: Food and Resource Economics thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science COOPERATIVE MARKETING AND TRANSACTION COST ECONOMICS: A STUDY OF BWANJE VALLEY RICE FARMERS IN MALAWI By Lucy Nyirenda August 2010 Chair: James Sterns Major: Food and Resource Economics The study investigated the role marketing cooperatives and transaction costs play in small scale farmers? marketing of agricultural produce, especially rice from Bwanje Valley, Dedza district in Malawi. Production and marketing cooperatives are currently being promoted in Malawi with a goal to reduce the problems faced by small-scale farmers in marketing their produce. The study specifically looked at the consideration that farmers give to transaction costs when making a decision to participate in the marketing cooperatives and consequently the effect that membership has on incomes earned from rice. The study also researched sources of the transaction costs that these farmers face and if the sources are associated with cooperative membership. A probit analysis was used to establish the relationship between transactions costs and farmers? decision to join the cooperative. The results show the effect and relationship between the transaction cost proxies that were used and the farmers? choice decision on whether to join the cooperative or not. The key conclusion from this analysis is that little consideration is given to transaction costs when the farmers are making a decision about participating in a cooperative. A regression model was used to determine the relationship between cooperative membership and incomes earned from rice marketing. The results from this analysis suggest that cooperative membership has a significant positive effect on the income earned by the farmers who are cooperative members.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Lucy Nyirenda.
Thesis: Thesis (M.S.)--University of Florida, 2010.
Local: Adviser: Sterns, James A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042207:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042207/00001

Material Information

Title: Cooperative Marketing and Transaction Costs Economics A Study of Bwanje Valley Rice Farmers in Malawi
Physical Description: 1 online resource (102 p.)
Language: english
Creator: Nyirenda, Lucy
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: cooperatives, farmers, malawi, marketing, probit, tce
Food and Resource Economics -- Dissertations, Academic -- UF
Genre: Food and Resource Economics thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science COOPERATIVE MARKETING AND TRANSACTION COST ECONOMICS: A STUDY OF BWANJE VALLEY RICE FARMERS IN MALAWI By Lucy Nyirenda August 2010 Chair: James Sterns Major: Food and Resource Economics The study investigated the role marketing cooperatives and transaction costs play in small scale farmers? marketing of agricultural produce, especially rice from Bwanje Valley, Dedza district in Malawi. Production and marketing cooperatives are currently being promoted in Malawi with a goal to reduce the problems faced by small-scale farmers in marketing their produce. The study specifically looked at the consideration that farmers give to transaction costs when making a decision to participate in the marketing cooperatives and consequently the effect that membership has on incomes earned from rice. The study also researched sources of the transaction costs that these farmers face and if the sources are associated with cooperative membership. A probit analysis was used to establish the relationship between transactions costs and farmers? decision to join the cooperative. The results show the effect and relationship between the transaction cost proxies that were used and the farmers? choice decision on whether to join the cooperative or not. The key conclusion from this analysis is that little consideration is given to transaction costs when the farmers are making a decision about participating in a cooperative. A regression model was used to determine the relationship between cooperative membership and incomes earned from rice marketing. The results from this analysis suggest that cooperative membership has a significant positive effect on the income earned by the farmers who are cooperative members.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Lucy Nyirenda.
Thesis: Thesis (M.S.)--University of Florida, 2010.
Local: Adviser: Sterns, James A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042207:00001


This item has the following downloads:


Full Text





COOPERATIVE MARKETING AND TRANSACTION COST ECONOMICS: A STUDY
OF BWANJE VALLEY RICE FARMERS IN MALAWI




















By

LUCY NYIRENDA


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2010

































Lucy Nyirenda
































To my husband, Lincoln, and my children Chawezi and Nancy, thank you for enduring
the hard times we went through and for your understanding and always being there for
me. It was not easy but we made it through together. To my late Dad and Mom, I am
what I am today because of the principles you instilled in me. Dad, Mom I made it and I
am sure I would have made you proud.
Lastly to my brothers and sister, Owen, Phyllis, Sunganani and Geoffrey, thank you for
your support throughout my life.









ACKNOWLEDGMENTS

I would like to thank USAID through the UILTCB program for their financial support

and Ministry of Agriculture and Food Security for letting me participate in the program. I

also would like to thank my brothers and sister and friends for their moral support and

encouragement they gave me throughout the study period.

My heartfelt thanks go to my committee chair Dr. J.A. Sterns and Dr. Z.F.Gao for

tirelessly helping me with my work. Their comments and criticisms were a great

contribution to the shaping and perfecting of this thesis. Many thanks also go to the staff

of Dedza DAO and especially Mr. Siyasiya and Mr. Mafosha from Mtakataka EPA for

their support and input during data collection and also to the farmers of Bwanje Valley

for their willingness to provide information which has enabled the accomplishment of

my work. I would also like to thank the following colleagues,Mr. M. Banda and Ms J.

Msosa from USAID Malawi, Mr. M. Kanjadza, Mr. K. Chaula, Mr. Mlangeni, Mr. Dzimbiri,

Ms H. Chima who helped me in one way or another preparing for my trip and gathering

information on my behalf while I was in the United States.

Many thanks also go to my friends, Ms F. Nkana, Mr. P. Soko, Mr. B. Kamwana,

Mr. I. Thindwa Mr. J Chiputula and Mr. T. Blare who made my stay in Florida

memorable.

Special thanks go to my husband Lincoln whose love, care, patience,

understanding and support gave me the courage to go on even when it was tough.

Above all, I would like to thank God almighty for all the blessings he showered upon me

and for making it possible for me to finalize my study. Without his will I would not have

made it.









TABLE OF CONTENTS

page

ACKNOW LEDG M ENTS ............... .................................... ......................................... 4

LIST O F TABLES .............. ............................................................................. 7

LIST OF ABBREVIATIONS ................................................ 8

A BSTRACT ........... ........ ............................................................. ..... 9

CHAPTER

1 INTRODUCTION, PROBLEM STATEMENT, study OBJECTIVES, TESTABLE
ASSERTIONS, ANTICIPATED benefits, AGRICULTURAL production in malawi .. 11

Introduction ...................... ...... ............... 11
Anticipated Benefits .................. ....... .......... ......... 13
Problem Statement ...................... ................. ... ............... 13
Study Objectives .................... .......... .......... ......... 14
Testable Assertions .................. ....... .......... ......... 15
A agricultural Production in M alaw i ........................................................ ........ 15

2 LITERATURE REVIEW .................. ................... ......... 18

Transaction Cost Theory............................................ ............... 18
Overview ................ .......... ............. ......................... 18
Bounded Rationality and Opportunism .............. ...... ............... 19
Asset Specificity .................... .............................. 20
Transaction Costs Studies .......... ........... .................... 21
Transaction Costs and Organization Theory............................................ 23
Marketing Cooperatives .......................................... 25

3 MATERIAL AND METHODS OF STUDY ...................... .................. 29

D ata C collection ......... .. ............................................................................ 29
S tu dy A re a ............................................................................................... 3 0
S am pling D esign............................................. 31
Analytical Framework .................................................. 33
Analytical Approach ............... ................... ......... 33
Descriptive Analysis .............. ......... ........... ...... ....... 33
Analysis of Transaction Costs and Cooperative Membership ............... ....... 34
Test of association ........................... ... ........... ........... 34
Regression analysis: Participation model .............................. ............ 36
Regression analysis: Incom e m odel ............................................ ......... ... ... 41
Expected R relationship ........... ................. ................. .................... ............... 44









4 SOCIAL AND ECONOMIC CHARACTERISTICS OF THE FARMERS ............... 46

Age and Gender of Farmers .................... ......... .................... 46
Farmers Education Levels ...................... ......... ....... ............... 47
Purpose of Producing Rice .................... ......... ...... ............... 48
F a rm e rs' S o urce of Inco m e ............................................................ ... .. ............... 4 9
Rice Production and Marketing ............ ..................... ........... ... .. ............... 51
Rice Production ...... .... .............. .......................................................... 51
Rice Marketing ................................... ......... ......... 53
M ajor Types of R ice Buyers in the A rea......................................... ... .. ............... 55
R ice M marketing C constraints .......................................... ................ .............. 58

5 TRANSACTION COSTS AND COOPERATIVE MEMBERSHIP ........................... 61

Transaction Costs Sources.................. ................................... 61
Market Information ............................. ........ ...... ......... 61
Availability of Few Buyers in the Market ......... ...... ....... ................ .......... 64
Poor Road Infrastructure .......... .. ............... ............. ........... ............... 65
Asset Specificity Levels: Land Use ...... ........... ........... ........... ... ............... 65
U uncertainty Leve ls ............... ................................................. ........... ...... 67
Frequency ......................................... 69
Cooperative Membership............................................. 70
Reasons for Participating in the Cooperative ..................... ...... ........... 71
Reason for Not Participating in the Cooperative............... .......... ........ 72
S o c ia l N e tw o rk ............ ............ ...................... ...... ............... 7 3
Membership Restriction .......... ................ .. ........ ....... .. ........ ......... 74
Regression Analysis Results ........... .............. ............................ ............... 75
Participation Model .................................................. 75
Inco m e M ode l......................................................................................... 78

6 CONCLUSIONS AND POLICY IMPLICATIONS..................................... 80

C o n c lu s io n ............................................................................................... 8 0
P o licy Im p licatio ns ............... ................................................. ................ 82

APPENDIX

A QUESTIONNAIRE FOR THE STUDY ......... ....... .............. ............. 86

B MAP OF MALAWI SHOWING THE STUDY AREA .................. .............. 97

LIST OF REFERENCES .............. ............................... 98

B IO G RA PH IC A L S K ETC H ............. ................. ................. .................. ............... 102






6









LIST OF TABLES

Table page

4 -1 A g e of fa rm e rs ............... .......................................................... 4 7

4-2 Education levels for farm ers ........................................................... ..... ......... 48

4-3 P purpose for rice production........................................................ ... .. ............... 49

4-4 Farmers' sources of off-farm income............................................... 50

4-5 Rice production quantities, reported in 50kg bags ................... .................. 52

4-6 Results for cross tabulation between coop membership and production ........... 52

4-7 Farm ers m markets .............. ....................................................................... 54

4-8 Type of major rice buyers in the area ............................. ......... .............. 55

4-9 Results for cross tabulation between buyer knowledge and sales volume......... 58

4-10 Rice marketing constraints ............ ................................... 58

5-1 S sources of M market Inform ation................................................... ... ................. 62

5-2 Information about which farmers are certain at time of sale ............................. 68

5-3 Frequency of transactions ............. ...................................... 69

5-4 Reasons why farmers joined the cooperative.............. ..... ................. 71

5-5 Reasons for not participating .................. ................ ............... 72

5-6 Frequencies on join responses.................................................. ............... 75

5-7 Participation m odel results ................................................................................... 76

5-8 Incom e M odel R results .............................. ........................ .............. 79













ADMARC

ADD

AEDC

EPA

Coop

GDP

GoM

ICA

IRB

JICA

MoAFS

TC


LIST OF ABBREVIATIONS

Agricultural Development and Marketing Cooperation

Agricultural Development Division

Agricultural Extension Development Coordinator

Extension Planning Area

Cooperative

Gross domestic product

Government of Malawi

International Cooperative Alliance

Internal Review Board

Japanese International Cooperation Agency

Ministry of Agriculture and Food Security

Transaction costs









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

COOPERATIVE MARKETING AND TRANSACTION COST ECONOMICS:
A STUDY OF BWANJE VALLEY RICE FARMERS IN MALAWI

By

Lucy Nyirenda

August 2010

Chair: James Sterns
Major: Food and Resource Economics

The study investigated the role marketing cooperatives and transaction costs play

in small scale farmers' marketing of agricultural produce, especially rice from Bwanje

Valley, Dedza district in Malawi. Production and marketing cooperatives are currently

being promoted in Malawi with a goal to reduce the problems faced by small-scale

farmers in marketing their produce. The study specifically looked at the consideration

that farmers give to transaction costs when making a decision to participate in the

marketing cooperatives and consequently the effect that membership has on incomes

earned from rice. The study also researched sources of the transaction costs that these

farmers face and if the sources are associated with cooperative membership.

A probit analysis was used to establish the relationship between transactions costs

and farmers' decision to join the cooperative. The results show the effect and

relationship between the transaction cost proxies that were used and the farmers'

choice decision on whether to join the cooperative or not. The key conclusion from this

analysis is that little consideration is given to transaction costs when the farmers are

making a decision about participating in a cooperative.









A regression model was used to determine the relationship between cooperative

membership and incomes earned from rice marketing. The results from this analysis

suggest that cooperative membership has a significant positive effect on the income

earned by the farmers who are cooperative members.









CHAPTER 1
INTRODUCTION, PROBLEM STATEMENT, STUDY OBJECTIVES, TESTABLE
ASSERTIONS, ANTICIPATED BENEFITS, AGRICULTURAL PRODUCTION IN
MALAWI

Introduction

Agricultural marketing for small scale farmers in Malawi became an issue after

several changes in policies were executed, following the Structural Adjustment Program

implementation in 1981. Malawi is an agro-based country with 74% of the population

depending on agriculture for food and income. Better, more effective marketing of

agricultural produce is a key strategy for improving the incomes of these farmers. The

agriculture sector is dual in nature with the small-scale and estate sectors. The small-

scale sector contributes a higher percentage of the production and employment

especially in the rural areas (more than 70% of the agricultural gross domestic product,

GDP) compared with the estate sector.

These small-scale farmers are mostly based in the rural areas where access to a

lot of facilities, like good markets, is limited. Before market liberalization the Agricultural

Marketing and Development Corporation (ADMARC), a state owned marketing board,

controlled the marketing of all agricultural produce in the country. Despite the

inefficiencies associated with the board, it worked to the benefit of the small-scale

farmers living in the rural areas by providing a market for their production.

The liberalization of agricultural marketing was expected to provide incentives for

the participation of the private sector, with consequences of competitive marketing

benefiting smallholder farmers through better marketing arrangements and higher

prices. However due to lack of supporting infrastructure like good roads, markets and a

proper system to support the liberalization, the results have not produced the intended









benefits. The evidence from rural Malawi does suggest that smallholder farmers,

particularly, the poor have benefitted less because of unfair trading practices and

monopsony power of private traders, and lack of reliable markets for agricultural

produce and inputs.(E. Chirwa et.al, 2006)

These small-scale farmers now have problems of access to better and reliable

markets. The farmers' inability to find alternative markets has eroded their bargaining

power for good prices. This coupled with other problems associated with small scale

production, like being in the rural areas where access to good roads, information, and

urban markets is limited, renders marketing of produce to be expensive, which

contributes to low incomes for these farmers.

One of the proposed solutions for improving marketing for small-scale farmers in

order to enhance their incomes was the use of institutional arrangements such as

cooperatives. These cooperatives are intended in part to minimize the cost of carrying

out transactions in marketing. The cooperatives offer an alternative market for produce

as well as providing access to inputs, such as fertilizers, for its members. Theory states

that use of institutional arrangements such as cooperatives can minimize transaction

costs. Some empirical studies conducted earlier, for example in Canada, have also

shown that cooperatives can minimize transaction costs. However most of these

studies were conducted in different scenarios as compared to small scale farmers in

Malawi.

Small-scale farmers in Malawi offer a different perspective of the transaction cost

problems because of other additional problems that they face like poor road

infrastructure and a general lack of organized forms of marketing, which farmers from









developed countries rarely face. These additional problems are likely to have an effect

on the choice to participate in a cooperative. Given this context, this study focuses on

the question: Do transaction costs, as compared to the other factors, have a significant

influence on the farmer's decision to join a cooperative in Malawi?

Anticipated Benefits

The Government of Malawi (GoM), through the Ministry of Agriculture and Food

Security (MoAFS), with support from development partners, is promoting the formation

of cooperatives as an institutional arrangement for assisting small-scale farmers by

helping them earn more revenue from their production. It is anticipated that this study

will assist policy makers interested in the appropriateness of using cooperatives as a

means to improving small-scale farmers' marketing of agricultural produce and

enhancing their incomes.

It is also anticipated that the results from the study will benefit the farmers in

Malawi, specifically in the area of study, by helping them make a more informed, better

choice when deciding whether to participate in cooperatives or not, if their goal is to

minimize transaction costs. Finally the study will contribute empirically to the literature

on transaction cost theory and cooperatives as an institutional arrangement for

minimizing cost, in relation to small scale farmers in a developing country.

Problem Statement

Malawi is characterized by widespread poverty with 52% of population living on

less than one dollar a day. The majority of the population, 80%, lives in the rural areas

surviving on subsistence farming with a small surplus to sell for income. National

surveys estimate that crop production accounts for 74% of all rural incomes and

agriculture is the most important occupation for 71% of the rural population (Chirwa,









2006). Considering the predominance of agriculture in the lives of many Malawians for

food as well as income, marketing strategies being implemented need to benefit the

farmers by lowering the cost of marketing and increasing their returns.

As indicated earlier, theory and empirical evidence in developed countries

indicate that cooperatives can work to minimize transaction costs incurred by its

members. However, small-scale farmers in Malawi offer a different perspective of the

transaction cost problem due to other additional problems faced in low income

countries. Looking at these inherent problems faced by the farmers, and also the

promotion of cooperatives by government and development partners as means for

improving small scale farmers' marketing, it is vital to find out empirically what role

cooperatives play in small scale farmers' marketing of their production in relation to

transaction costs, which are said to be high for small scale farmers. The focus was on

finding out if transaction costs influence the decision to join the cooperative, despite all

the inherent problems indicated earlier and the costs associated with membership. This

will give an indication as to whether cooperatives are an appropriate solution to dealing

with high transaction costs problems faced by small scale famers. It will also be

beneficial to find out if the cooperative membership has an influence on the incomes of

the small holder farmers. The outcome of the study will help to determine if improving

the marketing of small scale farmers' production through promotion of cooperatives is

an effective means to enhancing small scale farmers' income.

Study Objectives

This study assesses what role transaction costs and marketing cooperatives, as an

institutional arrangement in marketing, play in small scale farmers' marketing.

Specifically the study will look at the following objectives:









1. Identify the major sources of transaction costs that affect rice farmers
(cooperative as well as non-cooperative members).

2. Analyze the association between cooperative membership and transaction costs.

3. Determine if transaction costs influence the decision on whether farmers would
join the cooperative or not.

4. Determine if cooperative membership influences incomes earned from selling
rice for small-scale farmers.

5. Identify other factors that influence farmers' decision to join the cooperative.

Testable Assertions

This study aims to contribute to the understanding of the role of transaction costs in

small-scale farmers marketing of agricultural produce, specifically rice. The following

assertions are addressed in this thesis:

Assertion 1 Since small-scale farmers are located in remote areas far away from
service providers and major consumers of farm products, the distance to the
market, together with the poor infrastructure, poor access to other facilities and
information is manifested in high exchange costs due to hold up problems.

Assertion 2 Transaction costs prevail for small-scale farmers in rural Malawi. This is
reflected by the higher levels of asset specificity, information asymmetry and
transaction frequency leading to low returns from marketing produce.

Assertion 3 Cooperative members in the study area have lower transaction costs,
and this cost savings creates an incentive for farmers to join the cooperative,
positively influencing their decisions to join.

Assertion 4 Since cooperative members have lower transaction costs, the incomes of
cooperative members are higher than non-cooperative members.

Agricultural Production in Malawi

Agricultural production is still considered the mainstay of the country's economy. It

remains the major source of food and income for the majority of Malawian households

and also a major foreign exchange earner for the country, contributing over 80% of total

foreign exchange into the country. The importance of agriculture in Malawi is evident in









its share to the GDP for the country. Currently agriculture contributes over 35% to the

country's total GDP (National Economic Council, 2002).

The agricultural sector in Malawi is dual in nature with both a small-scale and an

estate sector, with the small-scale sector being the larger sector. Production for most of

the crops is by the small-scale sector. The major crops grown in the country include

maize, which is the staple food for the country and usually grown for home

consumption, tobacco, which is the major foreign exchange earner for the country, rice,

tea, cotton, coffee, and ground nuts.

Rice production and marketing in Malawi; Maize remains Malawi's staple food

with rice only coming second in most parts of the country. Rice is grown in districts that

are along the valleys of Lake Malawi and Lake Chiuta. Rice production is mainly by

small-scale farmers. Most of the rice producing sites were developed into irrigation

schemes during the post colonial era. Due to transition problems in management of the

schemes, the land was distributed to small-scale farmers for the production of rice in the

previously government owned schemes. In addition to these irrigation schemes other

farmers grow their rice in upland areas where it is usually swampy during the rainy

season.

Presently the irrigation schemes are being rehabilitated with the help of

development partners' funds to assist the small-scale farmers to stop reliance on rain-

fed production and start production through irrigation.

Before market liberalization the marketing of agricultural products was controlled

by the state owned agricultural marketing board. Farmers sold all their surplus produce

to the board. After market liberalization private traders constitute the majority of buyers









in the marketing of agricultural produce including rice. These private traders usually

purchase paddy which they later either sell to rice milling companies or polish it and sell

at retail to consumers in the city. The selling of paddy entails very low value addition to

the rice on the side of the farmers. This can have an influence on the price that the

farmer is offered. Rice sold to consumers by the middlemen is usually sold unpacked. It

is either weighed and sold per kilogram or it is sold using plates and charged according

to the size of the plate.

The promotion of marketing cooperatives by government has opened a new

market for farmers marketing rice. These rice marketing cooperatives process rice from

their member-farmers for sale directly to the consumers through supermarkets.

However private traders remain the biggest source of market opportunities for most rice

farmers due to low operational scale for most of the cooperatives, which then results in

them not being able to buy all the cooperative member farmers' as well other farmers'

rice.









CHAPTER 2
LITERATURE REVIEW

Marketing of produce involves carrying out transactions and transactions are said

to have a cost. Different frameworks have been provided to describe the aspects that

surround the cost of carrying out a transaction. Similarly literature has provided a

number of ways to manage or minimize the cost of transacting, one of which is the use

of alternative institutional arrangements besides market exchange. The paper reviews

some of the literature on transaction costs, both theory and empirical, and the ways to

minimize these costs for small-scale farmers.

Transaction Cost Theory

Overview

The transaction cost paradigm was pioneered by Coase in 1937. In his article, The

Nature of the Firm, Coase argued that market exchange is not without costs. He

recognized the role of transaction costs in the organization of firms, and other contracts.

Transaction costs include the costs of information, negotiation, monitoring, co-

ordination, and enforcement of contracts. He explains that firms emerge to economize

on the transaction costs of market exchange and that the boundary of a firm or the

extent of vertical integration will depend on the magnitude of the transaction costs.

The work of Williamson (1993, and 1996) on the economics of organization and

contracts follows on from Coase's line of thinking. Williamson combines the concepts of

bounded rationality and opportunistic behavior to explain contractual choice and the

ownership structure of firms. Opportunistic behavior manifests itself as adverse

selection, moral hazard, cheating, and other forms of strategic behavior. In Williamson's

framework, a trade-off has to be made between the costs of coordination and hierarchy









within an organization, and the costs of transacting and forming contracts in the market

(Drugger, 1983). This trade-off will depend on the magnitude of the transaction costs.

Williamson rests his arguments on the bounded rationality and opportunism

assumptions. He states that if these assumptions are not valid then the effects of the

variables that affect the transaction costs will not be valid.

Bounded Rationality and Opportunism

In transaction cost economics, all players are assumed to behave rationally. Their

bounded rationality results, in part, from the fact that they have limited resources, time

and energy. Bounded rationality also refers to the fact that people have limited

memories and limited cognitive processing power. We cannot assimilate all the

information at our disposal, nor can we accurately work out the consequences of the

information we do have. Herbert Simon originally defined bounded rationality as the

behavior that is intendedly rational but only limitedly (as cited by Williamson, 1979).

Others like Posner describe bounded rationality as not only being that information is

costly to acquire and to process, but also refers to the impossibility of thinking through

the complex but well structured problems (Posner R. A 1998)

Bounded rationality affects the certainty of a transaction. The transaction cost

economics approach focuses on how the characteristics of a transaction affect the costs

of handling it through markets, bureaucracies, and other forms of organization.

Williamson identifies the critical dimensions of characterizing a transaction and links

these to the institutional governance structure of transactions. The principle dimensions

describing a transaction are uncertainty, frequency of exchange, and the degree to

which investment are transaction-specific. Transaction costs include the costs of

gathering and processing the information needed to carry out a transaction, of reaching









decisions, of negotiating contracts, and of policing and enforcing those contracts. All

these transaction costs derive from a combination of bounded rationality (which reflects

both imperfect information and a limited capacity to analyze it) and opportunism, which

Williamson (1996) defines as "self-interest seeking with guile". Given imperfect

information about the future, all contracts are said to be necessarily incomplete. If

people were never opportunistic, however, incomplete contracts would not lead to

contract enforcement problems; contracts would simply state that if unforeseen

contingencies arose, the parties would act in a manner acceptable to all.

Asset Specificity

Williamson further develops Coase's idea. He recognizes that transaction costs

are more likely to be important when economic agents make relationship-specific

investments, that is, investments are specific to a particular group of individuals or

assets. The party which makes relationship-specific investments is susceptible to the

holdup problem. This leads us to the problem of asset specificity. This is because, once

this kind of investment is made, that party is somewhat locked in the relationship: there

is no competitive price for such investments once they are made. And furthermore,

Williamson claims that every contract is inherently incomplete in that people cannot

foresee all the contingencies of the future. The existence of incomplete contracts makes

the ex post surplus sharing sometimes unrelated to ex ante investments. This is similar

to what happens in a group setting. As a result, incomplete contract and relationship-

specific investment leads to under-investment, if the two parties are separate firms.

Williamson claims that integrating a transaction into the firm mitigates this opportunistic

under-investment.









Farmers with specific experience in production of a particular crop, and/or

production investments related to a particular crop, will have human and physical asset

specificity and are liable to hold up problems.

A study in Kenya used proxy indicators such as perishability of products and

period of production, to analyze the conditions of asset specificity and uncertainty for

each of Kenya's most important horticultural products, in order to determine the

expected institutional arrangement for linking producers and exporters/processors (D.

Tschirley et al., 2004). The study found that the dominant institutional arrangement for

coordination was that of long-term contracts and vertical integration, rather than spot

market exchange.

Transaction Costs Studies

There have been a number of fairly recent applications of transactions cost

economics in different fields of the food and agricultural sector. Examples of these

studies are Staal, et al. (1997), Key, et al. (2000) and Hobbs (1997). Very few empirical

studies have actually measured transaction costs to date. Staal, et al. (1997) asserts

that the limited empirical evidence on the nature and importance of transaction costs is

mainly caused by conceptual and measurement difficulties. For example, when

transaction costs are sufficiently high such that they prevent exchanges from occurring,

then, by definition, these costs cannot be observed because no transaction took place.

The available studies have tended to focus on distance to market as a single indicator

of transaction costs (Omamo, 1998).

One of the earlier studies in agriculture to carry out empirical measurement of

transaction costs was the innovative approach by Hobbs (1997). Hobbs carried out a

study about measuring the Importance of Transaction Costs in Cattle Marketing. The









study revealed that some transaction costs variables (such as grade uncertainty, risk of

not selling, time spent at the auction) were significant factors affecting the choice of

either live-ring auction or direct-to-packer sales.

A number of studies in Africa have employed a variety of techniques to measure

transaction costs in small-scale farming systems (Makhura, 2001; Staal et al, 1997;

Matungul, Lyne & Ortmann, 2001). Most of these studies have measured the effect of

transaction costs on small-scale farmers in relation to selection of marketing channel

and commercialization of the farmers. Few studies have considered the effect of

transaction costs on decision to participate in a cooperative. Participation in

organizational arrangements like a cooperative can be a result of many factors, one of

which could be transaction costs. If the farmer considers the transaction costs he incurs

to be high, then he will be expected to decide to join the cooperative in order to reduce

his transaction costs. This is based on the theory by Williamson that economic agents

will choose institutions, organizational forms and transactions that minimize the cost of

exchange. One way of determining if the theory applies to small scale farmers is

through assessing the impact of transaction costs on the individual farmer's decision to

opt for the organizational arrangement.

Transaction cost theory indicates that vertical integration can lower transaction

costs. However most of this research has been in relation to well established

organizations. For example a study in Canada for milk producers compared transaction

costs of farmers selling through the marketing board vis a vis farmers selling directly to

the processors. As will be discussed later in the proposal, this study provides a similar

scenario for farmers selling through the marketing cooperative and farmers selling









directly to buyers. The results from the study in Canada found that farmers selling their

milk through the marketing board incur less transaction costs compared to farmers who

sell directly to buyers (Royer et al, 2003).

There are many differences between these farmers and the small scale farmers in

a developing country. Small-scale farmers in Africa have other challenges, like low

literacy levels, poor infrastructure, small scale of production, lack or limited access to

production inputs, of which in most of the study scenarios are not major challenges.

These differences provide a different scenario to the studies conducted in developed

countries.

Transaction Costs and Organization Theory

Transaction costs theory examines the comparative economic costs of transactions,

which are the transfer of goods or services across a technologically separable interface

(Williamson, 1985). This theory explores the costs associated with the organizational

forms used to complete transactions when the exchanging parties are risk neutral,

predicting that the more efficient governance structure i.e., the organizational form

that reduces transaction costs more will be chosen. The general hypothesis of this

strand of the New Institutional Economics is that institutions or organizational forms are

transaction cost-minimizing arrangements that may change and evolve with changes in

the nature and sources of transaction costs. Coase (1937) pioneered this work when he

argued that market exchange is not costless.

Transaction cost theory is based on three behavioral assumptions: bounded rationality,

opportunism, and risk neutrality. This implies that a transaction is associated with

contractual risks, for example the opportunistic behavior of one of the contracting

parties, or the so called hold-up problem that "comes up if one contracting party tries to









exploit the other party's vulnerability connected to his asset specific investments"

(Royer, 1999 ). Asset specificity is important in transaction cost theory (Williamson,

1985) because rising asymmetric specific asset investments make markets less

competitive by reducing the number of potential trading partners, locking one partner to

the other and offering increasing incentives for renegotiation. As economic incentives to

behave opportunistically rise, the preferences of the contracting parties diverge: while

the exchanging firms have an incentive to jointly maximize long-term profits, each also

wishes to maximize its return in the short term by appropriating as much of the gains as

possible whenever contractual changes are required (Williamson, 1979). The condition

of diverging preferences is known as goal incongruence, or a lack of overlapping goals

(Ouchi, 1980). Cooperation is difficult to achieve in conditions of goal incongruence

because the individuals involved each follow their own differing objectives ( Mahoney

J.T &R.C McNally, 2004) The conceptualization of transaction costs theory assumes

goal incongruence is given, although some transaction costs theorists acknowledge that

incentive alignment or loyalty may reduce goal incongruence (Alchian & Demsetz, 1972;

Williamson, 1975)]. Organization theorists, on the other hand, place particular emphasis

on the ability to reduce goal incongruence through means other than incentives

because doing so may increase exchange efficiency (Ouchi, 1980). Organization theory

adds cultural control as a mechanism to reduce goal incongruence. Cultural control

focuses on aligning the views of governance structure members through organization

culture, which is a fairly stable set of assumptions, beliefs, meanings, and values that

individuals use to orient their thinking and to guide their actions (Scott, 1998).









Transaction costs theory shows that transaction cost problems can be dealt with

using the different forms of governance structures. Organizational theory forms the

basis for the concepts underlying the different governance structures. In trying to deal

with transaction costs we cannot run away from organizational theories. In Williamson's

framework, a trade-off has to be made between the costs of coordination and hierarchy

within an organization, and the costs of transacting and forming contracts in the market

(Williamson, 1985). This trade-off will depend on the magnitude of the transaction costs.

Marketing Cooperatives

Cooperatives and farmer organizations are institutional arrangements, the

importance of which has re-emerged recently as small scale farmers in developing

countries seek ways to organize themselves in the wake of agricultural market

liberalization. Rhodes (1995) defines a cooperative as a business firm owned and

operated by a voluntary association of member- patrons for mutual benefit. He goes on

to describe those members of a cooperative, that as owners they have the additional

perspective of patrons; and as patrons they have the owners' perspective that they can

affect services and activities.

Opportunism from the side of the processor and marketer is often mentioned as

one of the main reasons for farmers to set up a co-operative and carry out the

processing and marketing of farm products under their control. Farmers can prevent

being held-up by internalizing the transaction, that is, by integrating forward via the

creation of a proprietary co-operative firm. Whether farmers will do so, also depends on

the type of farm product (perishable or not) and the size of relationship-specific

investments (in relation to total investments). "The incentives for farmers to integrate

vertically via a cooperative firm to avoid opportunistic behavior are greatest where the









proportion of sunk costs to total costs at the time of the transaction is high and the

product is highly perishable, making its transfer to alternative markets on short notice

very difficult. Fruits, certain vegetables, and dairy products are examples."(Staatz,

1989).

Theory and some empirical studies suggest that the advantages of organizing

farmers into groups include, among other factors, a reduction in the transaction costs of

accessing input and output markets, as well as improving the negotiating power of

smaller farmers vis-a-vis large buyers or sellers (Iliopoulos & Cook, 1999). Other

studies looked at the characteristics of transactions between farmers and their

cooperatives and concluded that the cooperative "represents a hybrid organizational

mode blending market forces with elements of internal organization designed to

minimize transaction costs" (Iliopoulos & Cook, 1999).

The history of traditional cooperatives, on the other hand, suggests that

cooperatives have not always been successful at serving the needs of its members, and

their popularity had waned in the few decades preceding the 1990s. Cooperatives

suffered from various organizational problems and a lack of clearly defined property

rights assignments resulting in opportunistic behavior (such as free-riding, moral

hazard, agency problems, etc.), bureaucratic inefficiencies, and under-investment in the

cooperative (Cook, 1995; Cook & Iliopoulos, 1999).

Other research building on agency and game theory suggests that traditional

cooperative principles undermine optimal resources allocation and investment policies

(Vitaliano, 1983), as well as the stability of members' coalition (Sexton and Julie, 1993;

Staatz, 1989). In other words, major problems of cooperative farming appear related to









membership desertion (Barham and Childress, 1992), heterogeneous membership

occasioning free-riding behavior and limited investments and capital mobilization due to

horizon problems. However cooperatives remain one of the most promoted way of

dealing with small scale farmers marketing problems in developing countries.

The literature review has shown that transactions are not costless and that there

are many ways of looking at the cost of carrying out a transaction. The theory that was

developed in the early 1930's forms the basis for most of the transaction costs studies.

It has however been established that transaction costs are difficult to measure and most

studies look at the cost of transacting in one contractual form or the other to establish

the effects of transaction costs. One of the areas that were looked at is the use of

vertical integration or institutional arrangements to minimize transaction costs. Most of

the studies in this area have looked at the effects of transaction costs in the different

forms of organizational arrangements to determine the best organizational form to carry

out a transaction. For small-scale farmers, studies in this area have focused on the

effect of transaction costs on the farmer's commercialization. Not much information was

found from the studies conducted on whether the transaction costs are considered by

the small-scale farmers in deciding to participate in a cooperative, a specific institutional

arrangement that can potentially lower transaction costs for these farmers.

Institutions have also been said to be cost minimizing arrangements and that

economic agents choose institutions or organizational forms that minimize the cost of

transacting. Results from some studies that were carried out, found that organizational

forms such as cooperatives lower the costs of transactions for farmers. These lower









costs of transacting could be an incentive for farmers as economic agents to choose to

participate in such arrangements in order for them to minimize costs.

Since institutions and the institutional framework provide the incentives for

production and for people to engage in economic activity, an institutional analysis is

required to explain why the cost of transacting is so high in developing countries. The

frequent occurrence of market failure and incomplete markets (because of higher

transaction costs and information asymmetries) in developing countries cannot be

explained by conventional neo-classical economics and requires an institutional

analysis. Many of the institutions or formal rules of behavior that are taken for granted in

developed countries and that facilitate market exchange are absent in low-income

countries. Therefore, theories of institutional economics provide a useful framework that

could help determine the types of institutions needed (either formal or informal) to

improve agricultural marketing in developing countries.









CHAPTER 3
MATERIAL AND METHODS OF STUDY

Data Collection

Data was collected through one-to-one interviews using a semi- structured

questionnaire (See Appendix 1). The questionnaire was administered to sampled

cooperative members and non-cooperative member farmers. Data that were collected

include social economic characteristics of the farmers, types of markets the farmers

access, market information sources and use, type of rice buyers in the markets, period

taken to sell rice, frequency at which the farmers sell their rice, and rice prices. The

following information was also collected on cooperative membership: why they joined if

they are cooperative members and why not if they are non cooperative members,

period of membership, knowledge of members before joining the cooperative, their

intention on cooperative membership if they were given a choice to join.

Prior to data collection an application was made to the University of Florida's

Internal Review Board (IRB) for approval of the study. An application form together with

the questionnaire and consent forms were submitted to the IRB. The consent form was

translated into Chichewa which is a local language as was required by the IRB. Data

collection proceeded after IRB approved the study.

Pre-testing was done before the actual data collection. This was done to get

feedback on the quality of the questionnaire so that all omitted, irrelevant or

misunderstood questions and mistakes could be rectified. The pre-testing was done

with approximately 20 farmers in the same study area. Some minor changes were made

to the questionnaire following the findings and comments from the pre-testing.









The actual data collection covered a period of 3 weeks from 26th October to 11th

November, 2009. Two enumerators were hired to assist with the data collection. These

enumerators are Bunda College of Agriculture graduates who had just finished their

final year at college. Before going to the field for data collection, training of the

enumerators was conducted to reduce human error and also to ensure that the

questions were clearly understood and properly translated into Chichewa1 since they

were originally in English.

Study Area

The study was conducted in Malawi. Malawi is in the southern part of Africa. The

country is divided into three regions geographically Southern, Central and Northern

regions. The regions are further divided into 28 administrative districts.

In the agricultural sector, the country is divided into 8 divisions called Agricultural

Development Divisions (ADD). These ADDs are constituted of districts which are

controlled by the District Agricultural Offices (DAO). There are a total of 28 DAOs in the

country divided among the ADDs. These ADDs have between 2 to 7 DAOs in one

ADD. The highest number of DAOs in one ADD is 7 which is Blantyre ADD. Each

district is further divided into Extension Planning Areas (EPA) which are further divided

into sections. The sections are then organized according to villages in the area. The

extension workers are based in the sections and work with the villages that are within

the section.

This study was conducted within Dedza district, particularly Bwanje valley in

Mtakataka EPA. Dedza DAO falls under the Lilongwe ADD which is in the central


1 Chichewa is the local language through which the questionnaire were administered









region of the country. The district was purposively selected because the valley is a rice

growing region and also because they have an established rice production and

marketing cooperative. The cooperative was also purposively selected from the district

since it is the only rice production and marketing cooperative in the district. The study

targeted Bwanje Valley Rice Production and Marketing Cooperative (BVRPC). This

cooperative has been in operation for more than 10 years.

Refer to appendix B for a map of Malawi showing the study area.

Sampling Design

According to Cochran(1977), a formula for determining a sample size expressed

as a percentage is;

t )(P)()(3-1)
n )- ...)... .................. .................................................... (3 -1 )
(j2)

where t2 = the standard deviation score that represents the probability level of a

variable of falling within a confidence interval when the variable is normally

distributed

(p)(q) = Variance

j2= confidence interval

We yielded the following results after incorporating our data variables into the formula:

this is what were represented as follows:

(1.962)(.5)(.5) (3-2)
n = (3-2)
(.052)

n=384

The probability level and confidence interval of 1.96 and 0.05 respectively were used as

these are the commonly used estimates and normally accord estimation process









efficiency. The variables making up the variance represent the proportion of farmers

who are cooperative members and those who are non cooperative members., Czaja R.

and Jonny B. (1995) recommends that a 50% proportion for each is ideal in a situation

where the it is difficult to source the specific estimates for these proportions. The study

needed to collect a sample size of about 384 to be able to represent the target

population of our study. However, due to budgetary constraints, time and other factors,

the study managed to collect a sample size of 190 This is still a significant figure

considering that it's still a large sample and it was randomly collected.

A two-stage sampling was employed to come up with the sampling units. The

selection of Lilongwe ADD and Dedza district was the first stage. Dedza DAO has only

one rice production and marketing cooperative hence the cooperative was purposively

selected. The second stage was the selection of villages in the EPA where the

cooperative is based. The following villages were selected Mchanja, Fole, Mthembanji,

Bwanamakowa, Chatewa, Madziansatsi, Kafulama, Ndongwe. It was found out later

that Ndongwe and Chanja villages belong to Golomoti EPA. They were still selected

due to their closeness with the irrigation scheme and since most of the farmers from the

area were cooperative members who were farming in the irrigation scheme.

A stratified sampling technique was used where the farmers were stratified into

two strata, rice farmers in a cooperative and rice farmers not in a cooperative. Using a

list of farm families maintained by the Agricultural Extension Development Coordinators

(AEDC) in the area as a sampling framework, a total of 200 farmers were then randomly

sampled from the strata 110 from the cooperative and 90 non cooperative,. This

represents 3.3% of the total population of farm families in the villages from the area









under study. For coop members a total of 110 were sampled from 2062 cooperative

members representing about 5.3% of the total cooperative membership. For the non

coop members a total of 90 were sampled from a total population of 3919 farm families

who are not coop members in the area representing 2.3% of the population.

A total of 179 interviews were achieved due to other problems like movement, death

and unwillingness to participate in the study of the sampled respondents.

Analytical Framework

Analytical Approach

SPSS 15.0 and Stata were used for data analysis. Both qualitative and

quantitative analytical approaches were employed in analyzing the data. The two

approaches were used in order to widen the explanation base for the results that were

found.

The data have been analyzed in two parts. The first part includes the descriptive

analysis, cross tabulations and contingency tables. The Chi square is used to test the

presence of significant association between cooperative membership and proxy

variables that are used to determine the extent of transaction costs. The second part is

a regression analysis of survey data.

Descriptive Analysis

Frequencies, means, and percentages were computed to determine and rank the

constraints farmers face in marketing their rice. The descriptive statistics were also used

to summarize and categorize the information that was collected. Information on

marketing constraints, information sources and information use were analyzed to

determine the sources of transaction costs the farmers face in the area.









Analysis of Transaction Costs and Cooperative Membership

Test of association

Transaction costs, especially for small scale farmers, are difficult to measure

quantitatively. This research focused on assessing and documenting specific

dimensions of a transaction (asset specificity, transaction frequency and information

asymmetry). As noted in the literature review ( Chapter 2), proxies for information and

search costs such as availability and access to market information, bargaining and

negotiation costs such as number of available buyers, or how long it takes for farmers to

sell their produce, have been used to determine the extent of transaction costs farmers

incur while marketing their produce. When used in these studies, these proxies provided

keys to understanding organizational effects on transaction costs for individual farmers.

For this thesis, the analysis is comparative, assessing the cost of conducting

transactions in one organizational or contractual form relative to another. With this

concept in mind, the goal of the tests of association was not to measure the absolute

amount of transaction costs, but the relative ranking of transaction costs associated with

different organizational or contractual choices. Furthermore, as reported in numerous

project reports and related unpublished empirical studies, transaction costs are not

directly measured (Wang N., 2003) Certain proxies, such as uncertainty, transaction

frequency, asset specificity, opportunism, and so on, are used instead, which are

believed to critically affect the cost of transactions. A statistically significant relationship

between the chosen proxy and organizational governance suffices to make the point

clear that economizing on transaction costs is the unifying logic behind various

contractual arrangements of production. (Wang N., 2003). Hence, these studies are









able to move around the tricky question of quantifying the absolute level of transaction

costs.

For this thesis the following variables were used to approximate the extent of

information costs and bargaining costs. These variables are considered to have an

effect on transaction costs

Information and search

Knowledge about buyer
Knowledge about distant market
Time to sell in market
Knew price in market of transaction
Knowledge of prices in the other markets
Frequency of sale

Bargaining and negotiation

Average price received (K/kg)
Number of available buyers
Farm experience (years)
Land redeployment to other crops

Other control variables

Farmer's age (years)
Farmer's education (years)
Gender of the farmer
Membership in cooperative
Land holding size

The study concentrated on the search and negotiation costs to establish the

relationship between the transaction costs and cooperative membership due to the

nature of the data that was expected to be collected. The inability of the target

population to access well established markets and other limitations result in sales

without properly prepared contracts with buyers. This makes it difficult to identify

enforcement and monitoring costs and determine the relationship. Proxies will be used









in the to establish the existence of transaction costs among the farmers due to the

difficulties associated with the measurement of transaction costs.

Regression analysis: Participation model

The regression analysis will be used to determine the significance and direction of

the effect of the proxy variables on the choice decision to join a cooperative or not and

the effect of membership on income. During the pre-test it was discovered that

cooperative membership for Bwanje Valley Rice Production and Marketing Cooperative

is restricted to those farmers who have plots of land in the irrigation scheme. With the

restricted participation it would be difficult to establish the effect of the TC variables on

the decision to join the cooperative. In order to find the effect on ones decision

regarding cooperative participation, the study considered asking the farmers their

intention on choice given a scenario where membership is not restricted. The choice

that they indicated they would make was used in finding out the effect of the TC

variables on the choice decision of participating in a cooperative.

The TC variables that were used in the study are those that are said to affect the

three principle dimensions describing a transaction as identified by Williamson, 1979.

The dimensions include uncertainty, frequency of exchange, and the degree to which

investment are transaction-specific. The assumption is that a change in the variables

affecting these three dimensions will have an effect on the cost of the transaction. The

variables have been discussed below in relation to the transaction characteristics;

Asset specificity : This will be determined by the farmer's level of experience in

the production, ease of redeployment of the resource ( rice field ) and probability of

being held up by the buyer ( dependency on the buyer). Asset specificity has an impact









on the negotiation cost and higher levels of asset specificity lead to high negotiation

costs.

Uncertainty: Lack of sufficient information contributes to bounded rationality and

uncertainty about the transaction. Finding a buyer for the produce and assessing the

fairness of his or her offers requires availability of sufficient information. Search cost will

be determined by level of knowledge or the amount of information (about the markets,

buyers and prices) that the farmer has at the time of sale and source of information.

Frequency of transaction: Transaction costs are transaction specific. Farmers

who carry out many transactions each with a different buyer are likely to incur more

costs because for each of the transactions they carry out they will need to search for a

buyer. The cost of carrying out a transaction is also increased because there is a higher

probability of opportunistic behavior if there is no buyer-seller relationship. For such

farmer the higher the number of transactions, the more likely the farmer is a member of

the cooperative, as membership will help reduce the transaction cost incurred for each

transaction.

Other variables that were expected to affect the choice decision, to be a

cooperative member or not, were used as control variables in order to reduce the

omitted variable bias. The following variables were used as control variables -

education level of farmer, gender of farmer, age of farmer, and distance to the nearest

market. A regression was run to establish the relationship between transaction costs

and the choice of cooperative membership.

Econometrically the specification problem followed a latent regression model

y*= 3 X + (3-3)









Where y the latent variable is unobserved (i.e., the importance of transaction costs in

the decision process). What we observe is a dummy variable y defined by

y = 1 if y* > 0

y = 0 otherwise

Here X is the vector of transaction costs proxies and other variables that influence

cooperative membership choice decision and sl is the error term following normal

distribution. Probit method was used to estimate this equation. The likelihood function of

this model can be written as:

P (3, I yi, Xi ) = F < (- ) 1 ( -3' X)] (3-4)

The marginal effects of this model can be written as:


-a C( xp)Pk (3-5)

To test the hypothesis, a probit analysis was run on the model indicated below.

Join = Po + PAGE + 2AGESQR + f3 FAEXPER+ 94 BUYDEP + PsASSETRED

+ P6BUYINFO + i7MKTINFO + F1PRICINFO + P9EDUC + PloFREQ + P1GNDER

+ #12DIST+ P13TYPBUYER + P14MEMBER + FAEXPSQ + e1 (3-6)

In the model, Join was the dependent variable which took on a value of 1 if the

respondent indicated that they would join the cooperative and 0 otherwise. The

following are the independent variables that were used in the model,

AGE = Age of the respondent

AGESQR = Age of respondent squared

FAEXPER = Farmers experience in rice production

FAEXPSQ = Farmer experience squared

BUYDEP = Farmer level of dependency on buyers









ASSETRED = Ability to use land for production of other crops

BUYINFO = If farmer has information about buyer before sale

1= Has Information

0 = other wise

MKTINFO = If farmer has information about other markets

1 = has information

0 = otherwise

PRCINFO = If farmer has information on prices in other markets

1 = has information

0 = Otherwise

FREQ = Number of times farmers goes to sell rice

EDUC = Farmers level of education

DIST = Distance to the nearest main market

GNDER = Dummy for gender 1 = If farmer is a male

0 = If farmer is female

TYPBUYER = The type of buyer the farmer typically sells to

(Score variable; 1 = Wholesaler, 2= Wholesaler and

Retailer, 3= Retailer, 4= Retailer and Wholesaler)

Description of variables

Farmer's experience, level of buyer dependency, asset redeployment: These

affect the level of asset specificity. The TC framework states that substantial asset

specificity increases the costs of safeguarding inter-firm agreements because of the

prospects of opportunistic behavior of the trading partners (Williamson 1985). Such









costs are associated with haggling, verification problems, and other bargaining

difficulties that may arise because of inappropriate or ambiguous performance

measures, hidden information, or moral hazards. The variable TYPBYR which stands

for type of buyer was included in the model after establishing that it had some influence

on the negotiation abilities of the farmers. This was established during the discussion

held with the farmers on their ability to negotiate.

Knowledge about the buyers, knowledge about availability of markets,

knowledge about prices: In order to carry out a market transaction it is necessary to

discover who it is that one wishes to deal with. Therefore information becomes vital for

the transaction to take place. These variables have an effect on information

asymmetry. This can impact on the level of uncertainty and opportunistic behavior which

will have a bearing on the search cost. The expectation is that when faced with high

costs the farmer is likely to become a member of a cooperative.

Education: It is expected that farmers with more years of education will be able

to understand the benefits of membership to a cooperative and they are more likely to

join it.

Age and Agesq: Age is the age of the respondent and Agesq the age square

which was used to capture the diminishing effect of a respondents age

Expericence : Experience refers to the period the farmer has been growing rice

Distance: Distance to nearest reliable produce market is expected to have a

positive relationship with participation in the cooperative. The further away the nearest

reliable produce market is the more likely it will be for farmers to become cooperative

members in order to access a markets through the cooperative.









Regression analysis: Income model

The second model that was run was the income model. It was expected that the

incomes of the cooperative members will be higher than the incomes of non-cooperative

members, holding all other factors constant, following the low transaction costs that are

said to be associated with cooperative membership. Cooperative membership in this

case is considered to be an endogenous variable and as a result it will be included as

an instrumental variable in the income equation.

In an Instrumental Variable setting, we need to find a variable which should be

highly correlated with the treatment but not with the error term or the other explanatory

variables, i.e. Cov (z, Di)0, cov(zi xi) =0, and cov(zi u)=0 where u is an error term

(Wooldridge 2003).

Wooldridge (2003) showed that (IV) can consistently estimate the population

average treatment effects under weaker assumptions than those needed for two stage

least squares plug-in estimators proposed by Heckman (1998). Blundell (2002) also

indicated that given cross-sectional data, instrumental variables can be a reliable tool

for evaluating program impacts as long as it satisfies the required assumptions. Under

the required assumptions an (IV) provides the required randomness in the assignment

rule. Thus the relationship between the instrument and the outcome for different

participation groups identifies the impact of treatment avoiding selection problems.

Selection problems arise because ones decision to join a program may be determined

by other factors for example education. Higher education might influence their incomes

positively and entailing that the high incomes would be as a result of the high education

and not necessarily the program participation.

The problem could be written as follows;









Y2 = Pix + dlZmember +e1 (3-7)

z = P2 w + e2 (3-8)

Where Y2 will be the logarithm of rice incomes while z is a dummy for membership into

the cooperative and X contains age, education and gender of the farmer, land holding

size, membership to cooperative, production, off farm income, .

In the model Z will be the coop membership while w will contain Distance, Knew

coop members and experience, these variables will be used as instruments for

cooperative membership. It is assumed that these variables are correlated with

cooperative membership but are not regressors in the income model and the are not

correlated with the error term.

Following this, the empirical model for income can be written as;

Log income = #1 + If2quantity + f3age + f3agesqr + f4educ + P5 off farminc + F6

coop + p7 land + Pf male+e (3-9)

Where:

Dependent variable is the log income. This is the log of rice incomes the farmer has

attained in the previous growing season. The income was determined by multiplying the

quantity the farmer sold and the average price at which the farmer sold his rice. Log of

income was used in order to reduce the fluctuations in the income data.

The following are the independent variables:

Quantity: The quantity the farmer produced during the period. Since the study is

focusing on rice incomes then the quantity the farmer is able to produce will positively

affect the income they realize from the rice sales. This variable was later dropped









because it was highly correlated with the land variable which is the size of the plot of

land the farmer has.

Age: In the model age is the age of the farmer. Age is likely to have a positive and

negative effect on income because the older people are less likely to work as well as

the younger ones and this will affect the output. Age square was used to capture the

diminishing effect of age on the income

Education: The number of formal education years attained by the farmer.

Education affects the farmers' ability to comprehend issues and also the level of

technology adoption. These have an effect on the farmers' production levels and

consequently the income realized.

Off farminc. This is the availability of off farm income for the famer. For farmers

with more sources of off farm income they are likely to have more purchasing power to

access inputs and other production requirements. This will have an impact on the

quantity and quality of the rice they sell. In the end it affects the income realized from

rice sales.

Coop: This value is from the estimated model of cooperative membership.

Land : This is the size of the rice fields the farmer used to produce the rice during

the period being studied. It is expected the farmers with larger fields will have higher

incomes as a result of higher production for their field. ,

Gender: This is the gender of the farmer taking the value of 1 if the farmer is a

male and 0 for female.

The following instruments were used for cooperative membership to deal with the

problem of selection bias.









Experience: This is the period the farmer has been growing rice. The assumption

is that farmers who have been growing rice for a long time has a higher chance of

getting a plot of land in the scheme and participating in the cooperative.

Knew coop members: This is a dummy variable taking a value of one if the

farmer knew any coop members and 0 if not.

Distance: This is the distance to the nearest market.

Expected Relationship

The table below shows the relationship that is expected between the dependent

variable and the independent variables listed in the models explained above. These

expected relationships were derived from theory and other similar empirical studies

reviewed in this study. In some instances the relationships are difficult to determine as

they might either be negative or positive.









Table 3-1. Variables and their expected relationship
Variable Description Variable
Number of buyer farmer encounters Nobuyer
Farmers experience in rice production faexper
Age of the farmer Age
Level of buyer dependency buyerdep
Asset redeployment Assetred
Information about the market marketinfo
Knowledge about the prices Priceinfo
Number of times farmer sells frequency
Gender of the farmer Male
Distance to the market Distance
Quantity produced in the previous season quantity
Education level of the farmer Education
Farmers' membership in a cooperative Coop
Farmers off farm income offfarminc
Land holding size Land


Participation


+/-
+



+
+/-
+

+/-


Income

+/-
+/-









CHAPTER 4
SOCIAL AND ECONOMIC CHARACTERISTICS OF THE FARMERS

Age and Gender of Farmers

A total of 200 rice farmers were sampled in the study with 110 farmers from the

coop and 90 farmers from a target population of non coop members as described in

chapter 3. A total of 101 and 78 farmers respectively were successfully interviewed.

Results from the study show that no gender category is fully dominating in rice

production. For non coop members 50% of the sample were males and 50% were

females while for coop members 54.5% were males and 45.5% were females. For the

coop members, the percentage composition from the sample does not follow the

composition of the total membership of the cooperative which has a higher percentage

of females than males, 53.4% females and 46.6% males. However the results show that

no gender category largely dominates the participation in the coop which is similar to

the total membership composition as shown by the small difference in the percentage of

men and women in both the sample and the total population.

It can be noted that age was somewhat evenly distributed in both samples (coop

members and non-coop members). No age group was dominating as shown by the

results. From the sample 28.7% was between the ages of 18 to 24, 21.8% were in the

age group of 25 to 34 while 19% were in the age group between 45 to 51 years.

Table 4-1 shows the age distribution for the respondents according to their

membership status.

A difference can be noted in the age group of 25 to 34 where for the non cop

members only 12% belong to this group while for the coop members it was 21%. For

the higher ages participation in the coop is lower as shown by the dwindling









percentages in the sample while for non coop members the percentages are going up.

This shows that, currently, farmers tend to participate more in the groups when they

are younger and less when they grow older. Only 3% of the coop members were above

the age of 50. This may be a generational effect, such that as current young members

age, they will maintain their cooperative membership, thus in time, there will be less

difference across age categories in the future.

Table 4-1. Age of farmers
Age Coop membership status
Coop members Non-Coop members
18 to 24 years 28.7% 26.9%
25 to 34 years 21.8% 12.8%
35 to 44 years 26.7% 29.5%
45 to 51 years 19.8% 24.4%
Above 51 years 3% 6.4%



Farmers Education Levels

Education plays an important role in economic development. In addition farmers

who are literate are expected to better understand instructions and comprehend issues.

When farmers attain a certain level of education they adopt new technologies faster

than the less educated farmers. Their level of understanding is expected to improve and

enable them to participate in groups. In this study, farmers with no education level are

those that have not had any kind of formal education.

From the sampled farmers 62.7% of the coop members had some formal

education up to 5 years while 27.4 had never had any kind of formal education. Only 4

percent had more than 8 years of formal education. Similarly for non coop members the

majority in the sample had up to 5 years of education,51.3% and 34.9% never had any

kind of formal education. Only 2.6% had more than 8 years of education.









Table 4-2. Education levels for farmers
Education level Coop membership status
Coop members Non-Coop members
None 27.4% 34.9%
1 to 5 years 62.7% 51.3%
6 to 8 years 11.5% 5.9%
More than 8 years 4% 2.6%

With most of the members having very low education levels or none at all, it

provides a challenge for the cooperative in terms of management. This is so considering

that management positions in the cooperative are occupied by the farmers themselves,

who are responsible for decision making for the cooperative. These proposed decisions

made in the cooperative are then voted for by the members who have one vote each

regardless of the number of shares one owns or the amount of business one does with

the cooperative. It could be a challenge for members to comprehend issues properly

and vote wisely for the decisions.

Purpose of Producing Rice

In Malawi, maize is the staple food and other crops like rice are usually grown for

cash or for batter trade where they exchange rice for maize. However in some areas

where rice is grown in abundance it becomes a substitute for maize with some

households growing it for food. Most of the farmers along the valleys grow rice because

maize does not do well due to the nature of the soils. Soils in these areas are usually

water logged in the rainy season making it hard to grow maize. Bwanje valley is one of

those areas where most of the land is water logged during the rainy season making

production of maize difficult.

In this sample 57.4% of the coop members indicated that they grow rice mainly for

food but they also keep some to sell while 40.6% indicated that their primary goal for









producing rice is to sell but they also keep some for home consumption. Similarly for

non coop members the majority which is 64.1% indicated that they grow rice primarily

for consumption but they keep some to sell while 33.3% indicated that they grow rice

primarily to sell while keeping some for home consumption. Only about 1% in both

cases indicated that they grow rice just to sell or just for consumption.

Table 4-3. Purpose for rice production
Purpose Coop membership status
Coop members Non-Coop members
Consumption only 1.0% 1.3%
Consumption and sell 57.4% 64.1%
Sell only 1.3% 1.3%
Sell and consumption 40.6% 33.3%


It can be noted from the results that from the coop members a higher percentage

indicated that they grow rice with a primary purpose to sell and only keep some for

home consumption compared to the non coop members who indicated the same. This

could be attributed to the business exposure through various training that the coop

members are exposed to, which is not the case for their non coop counterparts.

However the difference between the two groups is not significant.

Farmers' Source of Income

Most rural Malawians depend on farming as their source of income. The study's

findings concur with statistics that indicate that the majority of rural households depend

on agriculture as their source of income. From the results 84.4% of the total sample

indicated that they depend on rice production as their main source of income, while

15.6% indicated that they depend mostly on other sources. These results are similar to

those of other GoM statistics which indicate that 80% of rural households depend on

farming as their major source of income.









Other income generating activities for most rural households include small

enterprises, piece works, and remittances. Off farm income has an effect on the use of

inputs by the farmers. Availability of off farm income improves the farmers access to

inputs which in turn is expected to improve their yields. It can be noted from the results

that the majority of the farmers did not have any other source of income other than

farming. This figure is higher for farmers who are not coop members than the farmers

who are coop members, 59.0% and 43.6% respectively. The study found that the other

major source of income apart from farming is small enterprise, which includes hawkers,

beer brewing, and small scale bakeries.

Table 4-4. Farmers' sources of off-farm income
Source of Income Coop membership status
Coop members Non-Coop members
Small Business 37.4% 24.4%
Piece Works 18.8% 15.4%
Employment 3.0% 0
Remittances .0 1.3%
None 43.6% 59%


From the sample 34.7% of the coop members indicate that they are involved in

some kind of a small business while for non coop members 24.4% indicated that they

have a small business.

The results indicate that most of the coop members have other sources of income.

In this case it would be expected that they would have a higher production than their

non member counterparts since their access to inputs would be higher. It can also be

noted that the other source of off farm income is piece works. The piece works involves

mainly working on other peoples' farms. From the sample 18.8% of the coop members









indicate that they earn additional income from piece works while for non coop members

it was 15.4% who indicated piece works as their other source of income.

Remittances and employment have very minimal significance as sources of

income for the farmers in this area of study. Only 3% from the coop members indicated

that they are employed and none from the non coop members while only 1.3% from the

non coop members indicated getting remittances as additional source of income and

none from the coop members indicated that they get remittances.

Rice Production and Marketing

Rice Production

Bwanje valley is one of the major rice growing areas in Malawi. Production of rice

in the area is done by small scale farmers. Development of an irrigation scheme

improved production by controlling the supply of irrigation water to the rice fields.

However production is also done in fields outside the irrigation scheme. All Cooperative

members have plots of land in the scheme where they grow their rice and some also

have additional plots of land outside the scheme where they also grow rice.

Rice production for these small scale farmers is still on the lower side compared to

the potential of 6000kgs per hectare (Malawi Government, 2002) which translates to

2450kgs or 48 50kg bags per acre. From the sample production ranged from 2 to 80

50Kg bags of rice which is 100 to 4000Kgs. The average production was 22 bags and

the median was 18 bags. It can be noted from the average and the median that

production is still low especially considering that most of these farmers, 76%, had one

or greater than one acres of land.

It can be noted from the study results that for most of the cooperative member

farmers, production was higher than that of the non member farmers. From the sample









46.2% of the non member farmers had a production of less than 10 bags of rice

compared to only 21 % from the coop member farmers.

Table 4-5. Rice production quantities, reported in 50kg bags
Quantity Coop membership status
Coop members Non-Coop members
Less than 10 21.8% 46.2%
11-20 bags 22.8% 26.9%
21 30 bag 23.8% 10.3%
31 40 bags 17.8% 10.3%
Above 40 bags 13.9% 6.4%


A chi square test of association reveals that there is a strong association

between rice production level and farmers membership to a cooperative at 99% level of

significance. The table below shows results from a cross tabulation between production

and membership.

Table 4-6. Results for cross tabulation between coop membership and production

Less than and equal More than 20 bags
to 20 bags
Cooperative member 45 56
Non Cooperative member 57 21

Pearson Chi2 = 14.607 Pr= 0.000

This can be attributed to the benefits that the coop member farmers enjoy in the

irrigation scheme which include extension services, controlled water supply and inputs

such as fertilizer.

The major problem facing the plots outside the scheme that was indicated as lack

of water control where sometimes there is too much for the crops or the water does not

get to the fields in adequate amounts. From the study it can also be noted that for the

coop members a higher percentage produces between 31 to 40 bags and only 13%









produces more than 40 bags. The productivity however is still lower than the potential

even for the farmers in the irrigation scheme.

Rice Marketing

Kohls and Uhl (1990) defined agricultural marketing as comprising the business

activities associated with the flow of goods and services from agricultural production to

food and fiber consumption. Marketing enables agricultural producers to break out of a

subsistence "straight jacket" and to grow produce for sale. In this way, it gives the

farmers more income so that they constitute a growing market for domestic industry

(Nyirenda, 1988). This was evident in the study as almost all the sampled farmers sold

some of the rice they produced for income. However following the low levels of

production most farmers reported that they only sell their production when cash is

needed as evidenced by 57.4% of the coop member farmers sampled who indicated

that they only sell when the need arises and 62.8% non coop member farmers who

indicated the same.

Availability of markets is one element for improving the marketing of agricultural

products. When farmers are making a decision on where to sell a particular crop they

base their decision not only on the prices they expect to receive in each of the market

but also on additional costs related to transacting in these markets (R. Vakis et al,

2003). The study found that most of the farmers sold their rice at the farm gate. From

the sample 85.9% of the non coop member farmers indicated that they sold their rice at

the farm gate and 73.7% of coop members indicated that they sold their rice at farm

gate. It is interesting to note however that for the coop members the figure is lower than

the non coop members. This indicates that coop members access other markets which









most of the non coop members do not. The table below shows the percentages of

farmers which patronize the different markets.

Table 4-7. Farmers markets
Market Coop membership status
Coop members Non-Coop members
Farm gate 73.7% 85.9%
Local 20.5% 29.3%
Distant 9.1% 6.4%
Cooperative 7.1% 0%


From the results it was found that a higher percentage of coop member farmers

sold their rice at the local market as well as the distant market, 29.3% and 9.1%

respectively compared to 20.5% and 6.4% of the non coop member farmers. Distant

market is the least accessed market. The reason indicated for not accessing the distant

market was mainly the high transportation cost that is associated with selling at the

distant markets. Most farmers indicate that they have little information about the distant

markets and accessing the market would be risky for them.

The study results also show that the cooperative market was only accessed by

the coop member farmers. Only 7.1% of the coop member farmers sold their rice to the

cooperative. It was found out that the cooperative had some financial problems and

hence they were not able to purchase rice from most of their members for two growing

seasons. As a result of this most of the members resorted to identifying other markets to

sell their rice. Failure of the cooperative to provide a market to all its members in this

case provides a challenge to one of the essential objectives of a marketing cooperative

which is to improve the members' access to markets by providing a market for their

produce.









Access to the different markets will have a bearing on the transaction costs that

the farmer faces. For example if a farmer only sells at the farm gate he is likely to avoid

incurring costs like transportation, time spent at the market, however competition is

likely to be low at the farm gate and prices are hence also likely to be low. The choice of

farm gate could be attributed to the nature of the business activity the farmers carry out

and also the production level. The study found that 59.9% of farmers sell their rice when

need arises. This means it would be costly for them to take a few bags to the market for

sell every time need arises. Selling at the farm gate becomes a less costly option for the

farmer.

Major Types of Rice Buyers in the Area

The study identified the types of buyers that are major players in the rice market in

the area. The table below shows the typical buyers to whom the farmers usually sell

their rice.

Table 4-8. Type of major rice buyers in the area
Buyers Coop membership status
Coop members Non-Coop members
End consumers only 3% 3.8%
Mostly consumers but also assemblers 2% 1.3%
Assemblers only 72.3% 69.2%
Mostly assemblers but also consumers 22.8% 25.6%.


From the study results it was found that the majority of the farmers sell their rice to

assemblers. Out of the sample 72.3% of the coop members indicated that they sell their

rice only to assemblers and 69.2% of the non coop member farmers indicated selling

their rice only to assemblers. Some farmers indicated that they sell to assemblers but

they also sell directly to consumers. From the sample 25.6% of the non coop members









and 22.8% of the coop member-farmers indicated that they also sell to consumer but

assemblers are their main buyers.

The assemblers are usually the middle men that come to buy the rice from the

area for resale at markets in the cities of Lilongwe and Blantyre as well as neighboring

districts. However the farmers also indicated that some of the assemblers are local

people from the area who also buy the rice to sell at distant markets.

From the sample only 3.8% of the non coop members and 3% of the coop

members indicated selling only to end consumers while 1.3% of non coop members and

2% of the coop members indicated selling primarily to consumers but they also sell to

assemblers. The farmers who indicated selling their rice to consumers said that they

sold polished rice. They mill the rice and usually sell it at the market in small portion as

demanded by the consumer. These portions could be as little as 250 grams. They also

pointed out that sometimes they do sell around their neighborhood through door to door

selling.

It can be noted from the results that middlemen are still the major players in the

marketing of rice in the area. These middlemen usually sell the rice that they purchase

to the rice milling companies in the city at higher prices than the prices they offer the

farmer. Rice Milling Company is one of the companies which package polished rice.

They indicated that they buy their entire paddy from middlemen who bring it to their door

step. This shows that the famers would be able to sell their rice at higher prices if they

were able to access such markets directly. However it entails a higher investment in

terms of transportation and also farmers' willingness to wait for payment for a specified









period of time which is agreed upon with the buyer. These two factors also contribute to

the farmers' failure to utilize such markets to sell their rice.

The study also looked at the type of relationship the farmers had with the buyer.

Knowing the buyer would improve the farmers' access to the buyer and reduce his

search cost because the farmer would know who to contact when they want to sell their

rice and also the negotiation cost because of the built trust between the buyer and the

farmer. From the study results it was found that most of the farmers sell their rice to total

strangers. From the sample 72.6% indicated that they sell their rice to people they do

not know at all while 17.9% indicated that they somehow know the buyers from past

transactions. Only 5.6% indicated that they know their buyers very well while 3.9%

indicated that when selling to consumers they know them but when selling to

assemblers they do not know them.

This can be attributed to the earlier results which show that the majority of the

farmers sell their rice to assemblers who might not always be the same people. The low

percentage of buyers being known makes a lot of sense because very few farmers from

earlier results said that they sell to end consumers. These end consumers are likely to

be people they know because they mentioned that they sell to consumers around the

same area or in their neighborhood.

Results from a cross tabulation between production sold and knowledge of the

buyer shows that the farmers who sell to buyers they do not know are the ones selling

lower volumes of rice than the farmers who sell to people they somewhat know from

past transactions. The difference was marginally significant at 10% level of confidence.

This could be as a result of the buyer farmer relationship that the farmers who sell more









are likely to establish with the buyers. It could be likely that farmers who sell lower

volumes do not really establish relationships with the buyers and hence sell to anyone

who comes to buy. However there is a need to get more insight into this relationship to

objectively describe the reason behind this. The table below shows the results of the

cross tabulation

Table 4-9. Results for cross tabulation between buyer knowledge and sales volume

Less than and equal More than 15 bags
to 15 bags
Do not know buyer at all 91 39
Somewhat know Buyer 24 8

Pearson Chi2 = 6.2938 Pr = .09

Rice Marketing Constraints

The farmers cited various constraints that they face when marketing their rice. The

major problems include: lack of proper and organized markets, low producer prices,

exploitation by buyers, poor road infrastructure, high transportation costs and lack of

market information among others. Table 4-10 below explains more in terms of

percentages.

Table 4-10. Rice marketing constraints
Market Coop membership status
Coop members Non-Coop
members
Lack of markets 33% 35.1%
Low prices 83% 77.9%
Poor road infrastructure 18% 10.3%
Too few buyers' 33% 50.6%
High transportation cost 19% 14.3%
Buyers cheating farmers 36% 26%
No proper link for buyers and farmers 18% 10.4%
Lack of market information 32% 23.4%









From the study results the major constraints that were cited by the famers were

low producer prices (83% for coop members and 77.9% for non coop members), too

few buyers (50.6% for non coop members and 33% for coop members) and lack of

markets (35.1% for coop members and 33% for non coop members). Low producer

prices could be attributed to fact that most of the farmers sell their rice soon after

harvest when the supply is very high. The farmers indicated that they are desperate for

money at this time of the season because they have a lot of bills to settle which are

usually due at harvesting time. This makes them vulnerable to selling at any price the

buyer offers so that they can get the money that is needed. The middle men who are

usually the buyers take advantage of this and offer very low prices. These prices

however improve as the supply gets low. For those farmers who are able to keep some

of their production to sell later, they benefit from the improved prices. The farmers could

benefit from the improved prices by staggering the sale of their produce so that they sell

some of their produce during the off season period to benefit from the higher prices.

The other important problems were: poor road infrastructure, lack of market

information, farmer exploitation by buyers and high transportation costs. In some areas

the roads were impassable during and soon after the rainy season. This hinders access

to the place by the buyers and most of the farmers cannot manage to transport their rice

to other markets. This then makes the few buyers who have managed to reach the area

to offer very low prices leaving the farmer with very few options either to sell at the

giveaway price or to keep his rice with very few prospects of selling in the near future.

The study also tried to seek ideas from the farmers on what they think could be

the solutions to the challenges indicated above. Suggestions to improve the situation









include: improvement of road infrastructure to improve transportation and access to the

area, government intervention and regulation on producer prices, all farmers in the area

getting more organized to form an association, and government to assist the farmers by

identifying markets or reopen ADMARC2 markets in the areas.









































2 ADMARC markets were closed down in the area following the restructuring of the marketing board









CHAPTER 5
TRANSACTION COSTS AND COOPERATIVE MEMBERSHIP

Transaction Costs Sources

As discussed earlier, transaction costs include information, bargaining, and

monitoring costs. Information costs occur before the exchange takes place and include

aspects such as searching for attributes that could facilitate the transactions, seeking

better prices, and looking for potential buyers. Bargaining or negotiation costs are

incurred during the exchange and include the time to negotiate a contract, reach an

agreement, and make arrangements for payment .Transaction costs are difficult to

measure directly. Certain proxies that are believed to affect the cost of transacting are

used instead. The study identified the aspects of a transaction that are a major source

of transaction costs for the farmers in the area. The transaction costs sources include:

market information (information about prices, other markets), availability of few buyers

in the market, limited other uses for land and poor road infrastructure.

Market Information

Market Information encompasses a number of areas which include information

about goods, quantities, qualities, markets, prices, buyers both for the market of

transaction as well as others. The purpose of market information is to assist both

producers and traders in balancing supply and demand so as to limit excessive price

rises and surpluses (Abbot, 1967). Market information is very crucial in agricultural

marketing since it highlights availability of good quality products and prices and it

enables both traders and producers to make informed decisions in regards to marketing

activities. When estimating transaction costs, market information affects search costs.

For the farmers to identify a buyer for their produce, make a choice on where to sell and









at what price, they need to have information in order to make informed decisions. Lack

of market information will hinder farmers from selling in more lucrative markets and

transacting at lower costs.

The study focused on information about availability of other markets and prices in

other markets before the farmer makes a sale. Table 5-1 presents information on the

source of information for the traders.

Table 5-1. Sources of Market Information
Source Coop membership status
Coop members Non-Coop members
Radio 18.8% 33.3%
Family and Friends 22.8% 30.8%
Extension Worker 5% 5.1%
Cooperative 8.9% 1.3%
Buyers 14.9% 2.6%
No Information 29.7% 26.9%

It can be noted from the results that most of the farmers depended on family and

friends as a main source of information. From the sample 30.8% of the non coop

member farmers and 22.8% of the coop member farmers indicated family and friends as

their major source of information. Other sources of information were; extension worker,

5% and 5.1%, cooperative (8.9% and 1.3%), Radio (18.8% and 33.3%) and Buyers

(4.9% and 2.6%) for coop member and non coop member farmers respectively.

It would also be interesting to note that 29.7% and 22.9% of the coop member

farmers and non coop member farmers respectively, indicated that they do not have any

kind of market information. This translates into 28.5% of the total sample indicating that

they do not have any kind of information. This makes market information as one of the

major sources of transaction costs in this area. This lack of market information would

increase the farmers cost of searching for a buyer and increases the chances of not









finding the buyer in advance thereby also increasing the time taken to sell the product in

the market.

A chi square test of association indicates that there is a significant association

between cooperative membership and source of market information. There is sufficient

evidence at 1% level of significance that source of information and coop memberships

are co-dependent. This means that coop membership may have an influence on the

farmers' access to information. The study however found that cooperative is one of the

least indicated sources of information and that non coop members had more information

compared to the coop members as shown by 29% Of coop members who indicated not

having any kind of information compared to 26% for non coop members. The magnitude

of the association cannot be explained by this test because chi square only measures

any evidence of association (Algresti and Finlay, 2008).

A study by Vakis et al in Peru found that farmers selling their produce at farm

gate had more information about prices in other markets than those selling at the local

and distant markets (R.Vakis et al, 2003)..This is similar to the results found by this

study. The earlier findings indicate that the majority of non coop members sell their rice

at farm gate than the coop member farmers and they have more information.

One other interesting aspect of the findings was that buyers are a source of

information. From the total sample, 9.5% indicated that they get information from the

buyers. It would be important to find out the quality of the information the buyer provides

to the farmers. Human beings are said to be opportunistic in nature (Coase, 1937),

which suggests an interesting research question do buyers provide correct and

adequate information to the farmers.









Availability of Few Buyers in the Market

One of the assumptions of perfect competition is many sellers and buyers in the

market. This ensures that prices are set fairly and determined through supply and

demand. Having few buyers may reduce competition and create opportunistic behavior

as buyers try to demand low prices. The study results show that few buyers in the

market was one of the major sources of transaction costs in the study area, evidenced

by 33% and 50.6% of coop and non coop member farmers respectively who indicated

few buyers in the market as one of the challenges. To the farmers few buyers would

increase their negotiation costs. The farmers would have few alternatives which will

make them take longer to negotiate for a better price. Sometimes at the end an

agreement might fail to be reached upon and no sale takes place. This becomes a cost

to the seller who has not made a sale but at the same time lost time negotiating for a

better price.

The study found that negotiating for a better price was indicated as one of the

areas where farmers spend a lot of time. From the sample 73.3% of coop members and

75.3% of non coop members indicated that they have problems negotiating for a better

price with the buyer more so when they are selling to assemblers. It should however be

noted that this could also be attributed to other factors such as age, experience, quality

of product in addition to the number of buyers.

The study also found that there is no significant association between cooperative

membership and number of buyer the farmer encounters in this area. Also, no

significant association was found between challenges in negotiating for price and

cooperative membership.









Poor Road Infrastructure

Transportation is one attribute of transaction costs. Farmers' access to markets

other than farm gate requires that they transport their product to the said market.

Equally buyers' access to farm gate markets requires that the roads be passable to

enable them to get to the farmers. Poor road infrastructure would deter many buyers

from going to buy farmers produce. This would create a non-competitive marketing

environment with the few buyers that will manage to get to the farm gate. With few

buyers, bargaining becomes difficult for the farmers due to limited alternatives for

potential buyers. For the farmers willing to sell their produce at local or distant markets

poor road infrastructure limits their ability to access these markets. It becomes costly to

transport produce in terms of transport cost and time spent to get to the market.

Results from the study indicate that poor road infrastructure was one of the

challenges the farmers meet in marketing their produce. The study found that the

challenge was experienced only in some parts of the study area. From the sample only

17.3% indicated bad roads as a challenge to their marketing. Farmers from these

particular areas indicated that during and soon after rainy season the roads are

impassable thereby limiting access to and from the area. Vehicles travelling to and from

the area wait for a number of days after a rain storm to be able to use the roads, these

delays increase transaction costs.

Asset Specificity Levels: Land Use

When an asset is specific to a particular use it leads to hold up problems. Rice

production requires very wet conditions and few other crops grow well in a similar

environment. The results from the study indicate that most of the farmers in the area

grow rice because the type of soil is suitable for rice production only. The farmers









indicated that during the rainy season the land is usually waterlogged. The only other

crop they think would do well is sugarcane. They however indicated that sugarcane

would not be as lucrative as rice is because of lack of market. From the study 87.2% of

the non coop members indicated using the land only for rice production while for coop

members 99% indicated growing only rice. In addition to the nature of the land, the coop

members are also bound by the irrigation scheme requirement that they grow only rice

during the rainy season. During the dry period they use the land for production of maize.

It can be noted from the results that 12.8% of the non coop members indicated

using their land for other crops. Further inquiry indicated that in some of the areas

outside the scheme the water flow sometimes changes course leaving some areas dry.

This makes it impossible to produce rice, so farmers in these areas grow other crops

such as cotton.

The farmers were also asked how easy it would be for them to use their plot of

land for production of other cash crops. From the sample 74% of the non coop

members and 65% of coop members indicated that it would be very difficult for them to

use the plot of land for production of other cash crops. These results are interesting to

note because from earlier results 99% of coop members indicated growing only rice in

their field from season to season but only 65% mentioned that it would be very difficult

for them to grow a different crop for cash on the field. The expectation was that a much

higher percentage of the coop members would find it very difficult to grow a different

crop on their field because they are bound by scheme requirements to produce rice only

during the rainy season. These results might also mean that 34% of the farmers would

grow a different crop given a chance. However, the farmers indicated that it is difficult









for them to grow a different crop because the only other crop which can do well in the

area is sugarcane which does not have a ready market and would earn very little

compared to rice. The results show the high levels of specificity the land has to the

production of rice.

A chi square test of association indicates that there is no significant association

between land use and cooperative membership. This could be attributed to the fact that

other than the scheme requirement that all farmers in the irrigation scheme should grow

only rice the type of soil for the whole area is suitable for crops that grow well in water

logged conditions. Rice happens to be the crop that suits the area.

It can be noted from the results that there is a high level of asset specificity when

it comes to use of land in the area. This creates a high probability of the farmers facing

hold up problems because they are stuck with production of rice every season

regardless of how the market for rice is behaving. This would create more room for

opportunistic behavior on the part of buyers because the farmers will still grow rice even

if the prices are low because they do not have an alternative crop for income.

Uncertainty Levels

Williamsons (1975) argued that in the imperfect world, where people have limited

information processing capacity and are subject to opportunistic bargaining, high

uncertainty makes it difficult for one party to determine correctness of the other party's

actions in a transaction. Uncertainty is exacerbated by lack of information. Earlier

findings from the study indicated that lack of market information is one of the constraints

the farmers face when marketing their produce.

The table below shows the kind of information about which the farmers were certain at

the time they were carrying out the transaction.









Table 5-2. Information about which farmers are certain at time of sale
Type of information Coop membership status
Coop members Non-Coop members
Number of buyer 6.2% 2.6%
Price to be received 60.8% 55.8%
How long it will take to sell 23.7% 16.9%
Location of sell 87.6% 88.3%
Prices in other markets 16.5% 22.1%
Other available markets 32% 42.2%
Transportation costs 26.8% 22.7%

It can be noted from the results that the uncertainty level is high. Most of the

farmers are only certain on the location of sale. From the sample, 87.6% of the coop

members and 88.3 % of non- coop members indicated being certain of the place of sale

at or before they carry out the transaction. These findings correspond to earlier findings

which indicated that most of the farmers sell their rice at farm gate. They could be more

certain about the place of sale because they always sell their rice at the farm gate.

What is more interesting to note is the percentage of farmers who indicated being

certain about price they would receive at the time of sale. From the sample 60.8% of the

coop members and 55.8% of the non coop members indicated that they are certain

about the price they will receive at the time of sell. This was not expected because

when asked about how the prices are determined the majority of the farmers (76.2%

and 84.6% of the coop and non coop members, respectively) indicated that the price is

determined through negotiations between the farmers and the buyers. The expectation

is that the farmers would be certain about the prices if there were standard prices set for

the market. It can however be noted that in the rest of the areas, most of the farmers

are not certain of what the outcome would be at the time of sale as shown by low

percentages of farmers who were certain about the information.









Frequency

Frequency is one of the transaction dimensions that affect the cost of transaction

(Williamson, 1979). Transaction costs are transaction specific and the more

transactions the farmer carries out with a different buyer for each transaction the more

costly it will be to him/her. The table below shows the frequency at which the farmers go

to sell their rice

Table 5-3. Frequency of transactions
Frequency Coop membership status
Coop members Non-Coop members
One trip and sell everything 5.9% 14.1%
Several trips within short period 5.9% 5.1%
Several trips over long period 30.7% 17.9%
Always at farm gate as need arises 57.4% 62.8%


It can be noted from the results that most of the farmers indicate that they only

sell their rice as need arises. From the sample 62.4% and 57.2% of non coop and coop

members respectively indicated selling their rice at farm gate only as need arises while

30.7% and 17.9% of coop and non coop members indicated that they make several

trips to the market over a long period of time. Some of the farmers indicated that they

make several trips to the market over a short period of time as shown by 5.9% of coop

member farmers and 5.1% of non coop member farmer. Only 5.9% of coop members

and 14.1% of non coop members indicated that they make only one trip to the market

and sell everything.

This shows a low level of business activity associated with the small scale

farmers in the area. Coupled with the nature of rice (its long shelf life) the farmers are

able to sell only when they have cash needs at the household. However due to poverty

levels which are high in the rural areas and low levels of production, as shown from









earlier results, it was indicated that it is difficult for the farmers to keep their produce for

much longer to benefit from increased prices resulting from dwindling supplies. High

levels of frequency of transaction will have a notable impact on the cost of transaction

mainly for those farmers who sell their produce at the local and distant markets when

selling to a different buyer all the time. A chi-square test of association shows that there

is some level of association between the frequency level and cooperative membership

at 10% level of significance with coop member farmers having fewer transactions than

non coop member farmers

Cooperative Membership

There are many reasons why farmers organize themselves into groups. Some of

the reasons include improving access to inputs, credit, and markets as well as

identification of new markets, improving crop production and increasing their bargaining

power for better prices. Farmers often can also procure their inputs more easily through

the cooperative rather than working on an individual basis.

According to the International Cooperative Alliance (ICA), a cooperative is

supposed to be an autonomous association of persons united voluntarily to meet their

economic, social and cultural needs and aspirations through a jointly owned and

democratically controlled enterprise. The study however found that the BVRPMC is not

based on the voluntary membership per se. The cooperative was formed on the basis of

the area having an irrigation scheme which was originally government-owned but was

passed on to the farmers after the Japanese International Cooperation Agency (JICA)

rehabilitated it. The farmers were allocated a plot in the irrigation scheme based on their

involvement during the scheme rehabilitation as part of the project's local community









contribution. Survey respondents indicated that it was a requirement that all farmers

who have a plot in the irrigation scheme should become cooperative members.

This finding created a challenge in determining factors that affect a farmer's

decision to participate in the cooperative or not. The farmer's intentions on choice

decision were used to determine if transaction costs would influence a farmer's

participation decision in the cooperative. This will be discussed under the regression

results.

Reasons for Participating in the Cooperative

As just indicated, the irrigation scheme had a requirement for all farmers with

plots of land in the irrigation scheme to be members of the cooperative. Other farmers

indicated additional reasons for participating in the cooperative. The table below shows

reasons why farmers joined the cooperative.

Table 5-4. Reasons why farmers joined the cooperative
Participation Reason Percentage
It was a must since I had a plot in the scheme 58.3%
I just followed my friends 10.2%
I wanted to get support from friends 10.2%
To access extension messages 8.3%
To access markets 6.5%
Source of inputs 6.5%

From the results it can be noted that the scheme requirement was the major

reason why most of the farmers joined the cooperative. 58.3% of the coop member

farmers mentioned that they joined the cooperative because they had a plot of land in

the irrigation scheme. The other reasons some farmers indicated were the social benefit

from the cooperative. 10.2% of the members said they joined the cooperative just to

follow their friends and another 10.2% said they wanted to get support from friends.









It can be said that social benefit is another major reason for the farmer

participation. Further inquiry showed that the farmers who are members of the

cooperative assist each other in time of need. For example when one is sick fellow

members take up the responsibility of looking after the sick member's plot. They assist

their fellow members with necessary agricultural practices such as cultivating the field,

weeding and other activities.

Few farmers indicated reasons such as improving access to markets and inputs

as the reason for participating shown by 6.5% on both reasons and 8.3% for access to

extension messages. This shows that little consideration is given to issues concerning

transaction costs when the farmers are working in a group.

Reason for Not Participating in the Cooperative

Similar to the participation reason the reason for not participating was also

largely influenced by the restriction that was put by the cooperative on membership.

Membership is restricted to those farmers who have a plot of land in the irrigation

scheme. The table below shows reasons mentioned by farmers on why they were not

cooperative members.

Table 5-5. Reasons for not participating
Participation Reason Percentage
I did not get plot of land in the scheme therefore did not 26%
qualify
The cooperative is far 10.4%
I was not interested 6.5%


The study however discovered that in addition to the restriction that was set by

the cooperative some farmers did not have information on the cooperative. Some

farmers only knew of the availability of the irrigation scheme but did not know of the

grouping called the cooperative. These farmers said that they thought the grouping was









only concerned with the irrigation scheme and since they do not have a plot of land in

the irrigation scheme they gave it little attention. The results also confirm the influence

of the membership restriction, as 26% of farmers who indicated that they were not

members because they do not have a plot of land in the irrigation scheme. The majority

of the farmers, 57.1 % who were not members mentioned that they did not have much

information concerning the cooperative. This could have been a result of the

membership restriction that made it seem less important to create more awareness

about the cooperative in the area for the other farmers to have sufficient information

about the cooperative.

Social Network

The study tried to find out if social networks would have an influence on a farmer's

participation in the cooperative. Social network in this case refers to the farmers'

knowledge of or relation to cooperative members prior to joining the cooperative. The

results showed that most of the members have been members since the establishment

of the cooperative and hence they did not have knowledge of anyone who belonged to a

cooperative prior to their joining. This is shown by 70% of the members indicating that

they did not know any member of a cooperative prior to joining. 30% of the sample

indicated knowing some cooperative members prior to joining. These are farmers who

joined the cooperative later mostly after inheriting or renting, on long-term, a plot of land

in the irrigation scheme. As a result most of the people they knew were extended family

members such as brothers, sisters, parents and cousins.

For farmers who were not members of the cooperative, the majority also said

they did not know many cooperative members. This could be attributed to lack of

sufficient knowledge about the cooperative since many of these same respondents









indicated that they know farmers who have plots of land in the irrigation scheme. The

social network could not have had much influence, even if information about the

cooperative was made available and the farmers knew some cooperative members, due

to the restriction that was placed on membership.

Membership Restriction

After establishing that there was a restriction on membership the study

investigated if the members were aware of this restriction and the reasons behind it.

From the results it was discovered that most of the members were aware of the

restriction that was placed on membership. From the sample 89.9% mentioned that they

knew that membership is restricted while 6.1% said that membership is not restricted

and 4% indicated not being aware of the restriction.

The farmers gave varying reasons to why membership was restricted. This creates a

lot of skepticism relating to the information that was communicated to the members as

being the reason for the restriction. The reasons that were pointed out by the members

include;

1. Rules laid down by the scheme

2. Most of the rules in the cooperative relate to the irrigation scheme and those who
do not have fields in the scheme will not feel as part of the group

3. To create uniformity since the irrigation scheme has rules that are followed

4. Farmers who do not have a plot of land in the irrigation scheme would not be
committed to the cooperative.

More information needs to be collected on how the decision to place a restriction

was made and if the farmers were involved in the decision process. There is also need

to find out how valid the reasons that were given are because the restriction and the









forced participation are violating one of the principles of a cooperative, which is

voluntary participation.

Regression Analysis Results

Participation Model

Two models were run to establish the significance and direction of the effect of

the TC proxies on the decision to participate in the cooperative and on income. As

indicated earlier on in the thesis (chapter 3), the participation model used the

respondents intended choice in participation rather than the actual membership. This

adjustment was made following the discovery made during pre-testing that participation

was restricted. "Join" was used as the dependent variable taking the value of 1 if the

respondent indicated that they would join the cooperative and 0 if they indicated that

they would not join the cooperative given a chance to make a voluntary decision. Table

5-6 shows the frequencies on coop and non coop member responses to the join

question.

Table 5-6. Frequencies on join responses
Frequency Coop membership status
Coop members Non-Coop members
I would join 75 69
I would not join 20 6
I don't know 6 3

The results for the probit analysis are shown in Table 5-7. It can be seen that not

all coefficient signs were as expected. For example the sign for MKTINFO was

expected to be negative, such that the more information one has about availability of

other markets the less likely they would want to join the cooperative. BUYERDEP also

did not turn out the way it was expected.









Table 5-7. Participation model results
Join Coefficient Std Errors Marginal Effects Std Errors
Age .07531 (.08872) -.015018 (.01741)
Agesq -.00312 (.00184) -.015018 (.09977)
Exper .09507** (.05619) .01896 (.01113)
Expersq .00297** (.00164) -.000592 (.00032)
Prcinfo .51216 (.21831) -.10215 (.04252)
Buyerdep .09994 (.12075) -.01993 (.02394)
Mktinfo .32796 (.30275) .06247 (.05488)
Assetred .01706 (.10104) .003402 (.02016)
Dist .01046 (.03536) .002086 (.00706)
Typbuyer .38396** (16869) .076586 (.03296)
Educ -.01663 (.05940) -.003317 (.01183)
Member -.79202* (.30164) .14947 (.0518)
Constant .25751
Psuedo R2 .1677
Log Likelihood -64.036
Note: Significance levels: *1% 5%, 10%. Standard errors are in parentheses

The results show a negative relationship between buyer dependency and

participation in the cooperative although it is not significant. The expectation was that

the more the farmer is dependent on the buyer the more likely they are to participate in

the cooperative. The variable member was included in the model after it was discovered

that a high number of members indicated that they would not want to join the

cooperative in a voluntary participation situation. Member is a dummy variable which

took on the value of one for members and zero for non cooperative members.

Four variables were found to be significant in explaining the choice decision the

farmer would make to participate in the cooperative if they were given a chance to make

a voluntary choice. The four variables are Information about price, member, experience

and type of buyer at 1%for member and information about prices, 5% for type of buyer

and 10% level of significant for experience. Price information has a negative relationship

with the decision to join the cooperative. This shows that the more information the

farmer has about prices in other markets the less likely it is for them to join the









cooperative. The marginal effects for this variable shows that additional information

would reduce the probability of choosing to participate in a cooperative by 4.2%. This

could be explained in a way that once the farmer has information about prices in other

markets it could form as a basis for them to negotiate efficiently with the buyers. This

will make them feel empowered to make informed decisions about a sale based on the

fact that they have the information. Opportunism on the side of the buyer would be less

since the farmer has information.

Membership status of the farmers had a negative significant effect on the farmers

decision to participate in the cooperative. The marginal effects shows that the

probability of choosing to participate in the cooperative is 14.94% lower for members

than non members. The respondent were asked to give a reason for their choice. Most

of the members who indicated that they would not join the cooperative indicated that the

cooperative has not met their expectations in terms of the benefits they anticipated and

that it is too involving for them in terms of the contributions and activities to do with the

cooperative. This could explain the lower probability for members since they have

experience cooperative membership and are making a more informed decision. The

other significant variable was the type of buyer. This variable as indicated earlier on

showed a significant influence on the decision the farmer makes to join the cooperative.

The type of buyer variable has a positive relationship with the decision to join the

cooperative. From the way the variable was set (as a score variable) it shows that the

farmers who sell to assemblers are more likely to join the cooperative than the farmers

who sell typically to consumers. This could be attributed to the fact that the farmers

selling to assemblers face the challenge of trying to negotiate with a big buyer who









sometimes is likely to dominate and dictate prices unlike consumers who buy in small

quantities and usually do not have the same bargaining position as the wholesaler.

Given this scenario then the farmers selling to the wholesalers would likely want to form

a collective action to deal with large scale buyers. They would negotiate more effectively

as a group than as individuals.

It is surprising that in this study gender and education were not significant as it is

with most participation studies. However this could be linked with earlier results that

showed that there was no major difference in participation in the cooperative between

men and women. Earlier study results on education (from chapter 4) also showed that

the majority of the farmers were less educated ( less than 5 years of education ). Very

few farmers in the sample had up to eight years of education (as low as 3%). This could

explain why education did not have any significant influence on the decision to join the

cooperative.

Income Model

The basic assertion on income was that since cooperatives members have lower

transaction costs, their incomes on average are higher than other farmers who are not

cooperative members but have similar characteristics. This was tested by checking

whether the variable Coop was significant in the model whose dependent variable was

the logarithm of rice incomes the farmers earned in the previous production season. As

would probably be expected income was found to be determined by a number of

variables including membership in the cooperative

In the model, the primary determinants of income were found to be land, age and

gender and cooperative membership which were significant at 1% for land, cooperative

membership and gender and 5% for age. Agesq and off-farm income were marginally









significant at 10%. From the results the coefficient on coop shows that membership can

increase rice incomes by 76.6%. Considering that the cooperative has not been

providing marketing services to most of its members, the increase can be attributed to

the higher production that is associated with cooperative farmers. This is so because all

cooperative farmers have plots of land in the irrigation scheme where water supply to

the crop is controlled and there are also intensive extension services being provided to

farmers, which probably also helps increase their production. This is also in line with

earlier results which indicate that cooperative farmers have higher production than non

cooperative farmers and hence their incomes are expected to be higher.

Table 5-8. Income Model Results
Logincome Coefficient Std Errors
Age .7758* .2153
Agesq -.1434** .0524
Educ -.0112 .0333
Land .3271 .0929
Gender .4306* .1478
Coop .7664 .2154
Off-farminc .2798*** .1441
Constant 8.3615 .3816
R2 .3190

Note: Significance levels: 1% **5% and ***10%

The analysis shows that membership increases incomes for the small scale rice

farmers in the area. However it is very hard to associate such differences in incomes

with cooperative membership due to the mandatory participation of scheme farmers into

the cooperative. In a way it could be said that the cooperative member farmers, who

also have irrigated lands have higher incomes because they own a plot of land in the

irrigation scheme and not necessarily because they belong to a cooperative.









CHAPTER 6
CONCLUSIONS AND POLICY IMPLICATIONS

Conclusion

The major objective of the study was to assess the role transaction costs and

marketing cooperatives play in small scale farmers marketing of rice in the Bwanje

valley area. In order to achieve this, the study researched if small scale farmers

consider transaction costs when making the decision to participate in a cooperative.

Results from the study provide information to policy makers concerning the current

promotion of agricultural and marketing cooperatives in the country. The study was also

aimed at finding out if there is any association between cooperative membership and

some dimensions of a transaction that affect the cost of transacting as well as the role

cooperative membership has on the incomes the farmers earn from rice.

A number of areas were discovered to be the major sources of transaction costs

for the farmers in the study area. These sources include limited access to market

information (about prices, markets and buyers) which hindered the farmers capability to

make informed sales decisions, poor road infrastructure which limits the farmers access

to markets as well as buyers access to the farmers, and availability of few buyers in the

area which increased the cost of negotiation for the farmers since they were negotiating

in less competitive marketing environments. These factors affected the cost of

transaction for the farmers through search costs and negotiation costs. Most of the

transaction cost proxies that were used were found not to have any significant

association with cooperative membership. It was hard to establish a relationship

between transaction cost proxies and cooperative membership because of the

restriction on membership into the cooperative. It was found that membership into the









cooperative is limited to those farmers who have plots of land in the irrigation scheme.

This posed a major challenge to most of the study's frameworks that were used.

For example, the participation model did not use membership as the dependent

variable. The model, instead, used a choice decision the farmers would make if

membership was voluntary. The results found that only three of the transaction cost

variables had a significant influence on the choice decision to participate in the

cooperative. The significant variables were type of buyer, information on prices and

membership status of the farmer. The challenge remained the membership restriction,

which is considered to have influenced each farmer's decision on the choice made on

whether they would voluntarily participate in the cooperative or not. The study also

found that cooperative membership has a significant positive effect on the incomes of

the farmers in the area. However it was discovered that the cooperative was not buying

rice from its member farmers. This could have had some effect on the outcome of the

results since the cooperative is currently not providing the required services to its

members. The statistically significant effect could be a result of other benefits that

increase the farmers' level of production such as access to inputs on loan, access to

extension services and participation in the irrigation scheme.

The study also identified other factors that would influence farmers' participation in

the cooperative. The factors the farmers indicated include getting support from friends in

the cooperative, accessing extension messages and as a source of inputs. These were

indicated as additional reasons why farmers joined the cooperative in addition to

membership being a requirement for those farmers with plots of land in the irrigation

scheme.









Other interesting findings included learning that buyers are a source of market

information for many of the farmers. It would be beneficial to identify the type and

adequacy of the information the buyers give to the farmers considering that human

beings are said to be opportunistic.

Finally it was difficult to determine the actual role the cooperative is playing in the

marketing of rice for farmers in the area because of the current operational status of the

cooperative and the restriction on membership. The cooperative was not buying rice

from most of its members due to financial problems. Hence, few of the interviewed

farmers benefitted from the marketing services of the cooperative. The farmers only

benefitted through the controlled water supply in the irrigation scheme, which increased

their production, and some inputs which were given to the farmers on loan. This most

likely explains why farmers in the cooperative reported higher production and hence

higher incomes than farmers who were not in the cooperative.

Policy Implications

Based upon the analysis of the empirical findings of this research, the following

policy initiatives have been identified for potential interventions in order to improve some

of the negative factors affecting farmers in the study area:

Market Information Systems: The study results show that availability of market

information plays a significant role in small-scale farmers marketing as shown by its

significance in the farmers' choice decision. However, the study also found that

information that goes to the farmers is limited and most of the farmers depend on

friends and relatives for market information and sometimes on buyers. In order to

improve in this area there is a need to strengthen the market information systems to

ensure the required information about available markets and quantities demanded,









prices in other markets, and available means of transportation, is delivered to the

farmers in a timely manner. This could be achieved through intensifying the information

delivery through the radio during the farmer radio programs and also making use of

extension workers by providing them with current and updated information regularly. In

addition the study results show that the extension workers are the least common source

of marketing information for the farmers, i.e. 5% of respondents indicated extension

worker as their source of marketing information (Study Results). This could be as a

result of lack of knowledge and skills in the marketing field or insufficient numbers of

extension workers in the area.

In order to improve farmers access to information through the extension workers,

there is a need to consider areas of capacity building of the extension workers in the

field of marketing and also improving the farmer extension worker ratio to ensure that

most of the farmers are being reached with the required extension messages in the

areas of production and marketing. This will help improve the farmers' production levels

as well as their marketing skills, both of which should help them increase their on-farm

income.

Financial Credit for Farmers and Cooperatives: Financial limitation is one of the

major problems rice farmers face. For marketing institutions lack of adequate capital to

invest is also one of the major challenges they face. The study found that the

cooperative is failing to provide marketing services to its members due to financial

problems. There is need to improve access to credit facilities for the cooperative to

operate to its full capacity and in turn provide the needed services to its members.

However there is need to improve the capacity of those in management positions in the









cooperative for them to be able to improve their management skills and operate

efficiently. This could in turn assist them to be able to identify the requirements of the

cooperative and lobby for the necessary support.

Contract Farming: One way to improve the marketing of rice for the small scale

farmer in the area would be to initiate contract farming between the cooperative and rice

milling companies that would be interested in buying paddy from these farmers. This will

ensure a ready market for the farmers which would then reduce the farmers search and

negotiation costs since they will be negotiating as a cooperative with the rice milling

companies. The availability of this kind of scheme which could ensure improved

production and the organization of the farmers would be considered as positive steps

for the development of partnerships between the farmers and the buyers through

contract farming.

Cooperative Awareness: There is need to increase awareness in the general

area of collective action in marketing by small-scale farmers through cooperatives,

including the potential benefits and costs associated with it. The study results found that

most of the farmers in the area do not have adequate information concerning

cooperatives. It was found that there was a very low understanding of what a

cooperative entails and what the farmers would benefit from as well as the cost to the

farmer if they joined.

The results also show that cooperative membership is strongly correlated with

higher farmers' incomes, hence the need for the farmers to be provided with more

information concerning the cooperatives and the associated benefits.









Further Research: The study established a number of research objectives which

proved to be difficult to test empirically or that conflicted with the actual conditions in the

target population. These included restrictions on cooperative membership (where some

farmers were required to join, while others were banned from joining), and the

operational status of the cooperative in the area of marketing. These are believed to

have influenced the results of the study significantly. Re-designing the study at this

later stage was difficult due to time and resource constraints. Further research would

therefore be recommended in the area which would include some of the raised issues in

the design of the study.









APPENDIX A
QUESTIONNAIRE FOR THE STUDY

INVESTIGATION OF THE RELATIONSHIP BETWEEN COOPERATIVES AND TRANSACTION
COSTS FOR SMALL-SCALE FARMERS IN MALAWI, AUGUST OCTOBER 2009


DATE OF INTERVIEW:
INTERVIEWER(S)/ENUMERATOR(S):
FARMER'S NAME:
FARMER'S LOCATION:
QUESTIONNAIRE CODE NUMBER:



1. When you grow rice, what are your plans for your production? [Check only one!]

SI plan to sell all of my production.
SI plan to use all of my production for home consumption.
SI plan to use my production primarily for home consumption, but will sell
some if there is a surplus.

SI plan to produce enough with each crop to sell some and keep some for
home consumption.


2. In your opinion what do you think is the most important problem rice farmers
face in marketing their production?










3. What do you think is the solution to the problem that you have mentioned above?









4. In addition to the problem you just mentioned, are there other problems that rice
farmers in your area have when they try to market their production? [Check all
that apply]
Too few buyers
Too little information about prices
Prices are too low
Urban markets are far away from the farms
Commercial traders/buyers treat farmers unfairly/try to cheat them
Roads are bad
Transportation of rice to markets is expensive
Other:
Other2:




5. Approximately, how much land do you use/farm for rice production?
Less than 0.5 hectare; less than one-half of an hectare
0.5 .99 hectares
1 1.99 hectares
Greater than 2 hectares


6. How much rice did you produce last growing season? ( please indicate the
number in terms of 50 kg bags)




7. How much of your total production did you sell?( please indicate the number in
terms of 50 Kg bags)


8. On fields where you grow rice, do you always grow rice from season to season?
O Yes
O No









9. If you stopped growing rice in your rice fields, how easily could you use these
same fields to grow a different food crop for home consumption, with 1 being
very easily and 5 being very difficult? Please check one.


1 2 3 4 5
Very Easily Somewhat Neutral Somewhat Very difficult
Easily difficult


10. If you stopped growing rice in your rice fields, how easily could you use these
same fields to grow a different cash crop for sale in markets, with 1 being very
easily and 5 being very difficult? Please check one.


1 2 3 4 5
Very Easily Somewhat Neutral Somewhat Very difficult
Easily difficult


11. Please tell us your reasons for growing rice on your fields instead of other crops.




12. Please tell us your sources of income from the farm ( please check all that
apply)
Rice production
Other cereal grains
Fruits and vegetables
Livestock production, specify: (e.g., poultry,
pigs, etc.)
Other ?
specify











13. From these sources of income that you have just told us, please tell us which one
is the most important source?




14. Do you have other sources of income in addition to income from farming?
O Employment
1 Business (Small enterprise)
O Remittances
O Pension pay
O None
O Other, Specify_


15. For how many years have you grown rice on your farm?


16. When you decide to sell your rice where do you go to sell (Please check all that
apply)?

O Farm gate
O Local market
E Distant market
E Cooperative
O Other,
specify


17. How far do you have to travel to get to the closest main market where you are
able to sell rice? Kms


18. What do you think are the advantages and disadvantages of selling your rice in
the closest local market?











19. What do you think are the advantages and disadvantages of selling your rice in
the market you use most frequently?



20. What are/would be the advantages and disadvantages of selling your rice in a
district market?


21. When you sell your rice, do you typically
Make one trip to the market and sell all the rice that you plan to sell from that
growing harvest
Make several trips to market within a one to two week period, selling as
much of your rice as you can with each trip
Make several trips to market, but spreading them out over a more extended
period of time, perhaps as much as one or two months
SOther? Specify



22. On an average day when you are selling your rice how many
wholesalers/assembler/buyers do you see in the market?
O One
O 2to5
O 6 to 10
1 More than 10


23. Typically to whom do you sell your rice?
O Only wholesalers/assemblers
O Only to end consumers
O Mostly to wholesaler buyers and assemblers but also to some consumers
O Mostly to consumers but also to wholesale buyers and assemblers
O Other? Specify









24. When selling your rice how often do you refuse offers from
buyers?


25. Please tell us the reason why you refuse the offers for the number of times you
have indicated in the above question.

26. When you sell your rice, how much time does it usually take?
less than 30 minutes after I arrive at market
30 minutes to one hour
two to five hours
every day is different... sometimes I sell quickly, sometimes it takes all day
SOther? Specify

27. How well do you know the buyers of your rice?

very well (they are my neighbors, family, etc.)
casually (I often sell to the same buyers, so I know them from past sales)
not at all
SIt depends...when I sell to consumers, I know them but when I sell to
assemblers/wholesalers, I do not know them
......... ..................................... .................... O their? S specify


28. When you sell your rice, how is the price that you receive determined?
the buyer sets price, no negotiation
_you, as the seller, set the price, no negotiation
negotiation between you and the buyer
farmer coop/association announces price
there is a standard price in the market that everyone uses
other?



29. Can you negotiate a better price if the quality of your rice is better than other rice
being sold? O Yes O No

30. Must you accept a lower price if the quality of your rice is worse than other rice
being sold? O Yes O No









31. When selling to assemblers/wholesalers, do you trust your buyers) to give you a
fair market price for your products?
I completely trust the buyer to be fair
The buyer is usually fair, but I have to bargain with him/her to get a fair price
I worry that the buyer may not be giving me the best price
I do not trust the buyer at all to be fair
SOther? Specify



32. On a scale of 1 to 5 with 1 being complete trust in the buyer and 5 being no trust
at all in the buyer, how much would you say you trust the wholesalers/trades that
buy your rice?

1 2 3 4 5
Very much Somewhat Neutral Not Very much Not at All


33. What was the highest price that you received for your rice this last growing
season? Kwacha / Kg


34. What was the lowest price that you received when selling rice? Kwacha /
Kg



35. What was the most common price that you received Kwacha / Kg


36. Which of the following potential problems do you encounter most frequently
when marketing your produce

O Finding a buyer
1 Negotiating for price
O Enforcing the contract / making sure that buyer sticks to agreed terms
O Finding information about prices
O Finding information about buyers/ other markets

37. When you sell your rice, what information about the sale do you know before
taking your rice to market? (Please Check all that apply)
The number of buyers in the market
The name of the buyer who will buy your product
The price you will receive









The amount of time it will take to negotiate the sale
The location of the sale/market transaction
Prices in other markets
Availability of other markets

_Any price premiums or price reductions related to the quality of the product
Transportation costs to get the product to market


38. We would like you to tell us how certain or uncertain you were about the
information you just indicated. Using this ruler/chart, with 1 indicating most
certain and 7 indicating most uncertain, your opinion about the certainty of each
bit of information that you have before you take your rice to market.


most certain/---------1----2----------------4-----5----6--- -7---
-/most uncertain


RANK RESPONSES
1
2
3
4
5
6
7


39. What is the most important source of information about market s that you use
when making plans to sell your rice? Please check only ONE.
O Radio
O Extension worker
1 Family or friends
O Cooperative
O Other specify











40. How often do you use the market information that you get from this source in
your sales plan (for negotiating a price or market selection) when you are selling
your rice?
Not at all
SOnly sometimes
Most of the time
All the time


41 .Are you a member of a cooperative?
O Yes
O No
O I don't know


If yes then go to question 42; If no then skip to Question 49.

42. For how long have you been a member?


43. What was/were the major reasons) that influenced your decision to join the
cooperative (check all that apply)


To get access to markets
To find a source of inputs
To get support from friends
Just followed my friends
Other; specify:


44. Before you joined the cooperative, did you know relatives or close friends who
were members of the cooperative? O Yes 1 No


45. If yes, what is your exact relationship with these cooperative members?


46. How many cooperative members did you know prior to joining?









47. Is membership to the cooperative restricted to the farmers who have fields in the
scheme?



48. If yes, why was membership restricted to only those who have fields in the
scheme?



49. If you answered no to question 41, why did you not join the cooperative?
O I did not have much information
E Its costly
O I don't qualify
O I was refused
O I don't have any marketing problems
O Other?

50. Do you know any relatives or friends who are members of a
cooperative?

51. If yes, how many cooperative members do you
know?


52. What is your exact relationship with these cooperative members?



53. Given a situation that membership rules were changed and that everyone in this
area is eligible and free to choose whether to become a member of the
cooperative or not regardless of the location of their plot. what would be your
decision given an opportunity to make a choice between joining the cooperative
or not,

O I would not join the cooperative
O I would join the cooperative
O I don't know


54. What are the reasons that would influence your decision indicated in the previous
question?









55. What year were you born?


56. How many years of formal education did you get?

O None
O Less than 5 years
O Up to 8 years10 years
O 'More than 8 Years

57. Gender: Male / Female










Thank you for your time and cooperation in this survey









APPENDIX B
MAP OF MALAWI SHOWING THE STUDY AREA


MOZAMBIQUE


--KEY
-.-.. International boundary
Main road
A Rice irrigation factory
[ Bwanje valley irrigation scheme


0 50 100
Scale (k
Scale (km)


Source: Sand in The Engine: Travails Of An Irrigated Rice Scheme In Bwanje Valley,
Malawi









LIST OF REFERENCES


Abbot J.C. 1967. "The development of marketing institution in Agriculture." In H.M.
Southworth and B.F Johnson (Ed), Development and Marketing. Cornell University
Press.

Alchian, A. A., and H. Demsetz. 1972, December. "Production, information costs, and
economic organization." American Economic Review 62(5), 777S795.

Algersti, A., and B. Finlay. 2008. Statistical Methods for the Social Sciences, 4 th ed.
New Jersey: Prentice Hall Inc.

Barham, B.L., and M. Childress. 1992. "Membership Desertion as an Adjustment
Process on Honduran Agrarian Reform." Economic Development and Cultural Change,
40-3, ABI/INFORM Global, pg. 587

Blundell, R., and Dias, M.C. 2002. "Alternative Approaches to Evaluation in Empirical

Microeconomics." Portuguese Journal of Agricultural economics 1: 91-115.

Chirwa, E W., P. M. Mule and J. Kadzandira. 2006. "Agricultural Marketing
Liberalization and the Plight of the Poor in Malawi": Working Paper No. 2005/08

Coase, Ronald. 1937. "The nature of the firm." Economica 4: 386-405.

Cook, M. L .. 1995. "The Future of U.S. Agricultural Cooperatives: A Neo-lnstitutional

Approach." American Journal of Agricultural Economics, Vol. 77, No. 5, 1153-59.

Cook, M.L., and C. Iliopoulos .2000. "Ill-defined Property Rights in Collective Action: the
case of US Agricultural Cooperatives. Institutions, Contracts and Organizations:"

Perspectives from New Institutional Economics.

Delgado, C.1999. "Sources of growth in smallholder agriculture in sub-Saharan Africa:
The role of vertical integration of smallholders with processors and marketers of high
value-added items". Agrekon 38(Special Issue): 165-189

Drugger, W.M. 1983. "The Transaction Cost Analysis of Oliver E. Williamson: A New

Synthesis." Journal of Economic Issues. Vol. XVII, No. 1: 95-114.

Gabre-Madhin, Eleni Z. 2001. "Market Institutions Transaction Costs and Social Capital
in the Ethiopian Grain Market." IFPRI Research Report 124, International Food Policy
Research Institute, Washington, D.C.

Heckman, J. J., and Vytlacil, E.1998. "Instrumental Variables Methods for the
Correlated Random effects Model." Journal of Human Resources 33: 974-976.









Hobbs, Jill E. 1997. "Measuring the Importance of Transaction Costs in Cattle
Marketing." American Journal of Agricultural Economics 79: 1083-1095.

Holloway, G., C. Nicholson, C. Delgado, S. Staal and S. Ehui. 2000. "Agro
industrialization through institutional innovation: Transaction costs, cooperatives and
milk-market development in the east-African highlands." Journal of Agricultural
Economics 23(3):279-288.

Iliopoulos, C., and M.L. Cook. 1999. "The internal organization of the cooperative firm:
An extension of a new institutional digest." Journal of Cooperatives 14:77-85.

Key N., E. Sadoulet and A. De Janvry. 2000. "Transaction costs and agricultural
household response." American Journal of Agricultural Economics 82(2):245-259.

Kohls, R.L and Uhl, N.U.1980. Marketing of Agricultural Products, 5th edition,

Macmillan, New York.

Mahoney J.T., R.C McNally. 2004."Explaining and Predicting the Choice of
Organizational Form: Integrating Performance Ambiguity and Asset Specificity
Effects."Working Paper 04-0109

Makhura M.T. 2001. "Overcoming transaction cost barriers to market participation of
smallholder farmers in the Northern Province of South Africa." PhD. thesis, University
of Pretoria.

Malawi Government (2002). Agricultural sector priority constraints, policies and
strategies framework for Malawi. Ministry of Agriculture and Irrigation: Lilongwe.

Matangul P.M., M.C. Lyne and G.F. Ortmann. 2001. "Transaction costs and crop
marketing in the communal areas of Impendle and Swayimana, KwaZulu-Natal."
Development Southern Africa 18(3):347-363. September 2001

National Economic Council. 2002. Malawi Government Economic Report, Government
printer: Zomba

Nyirenda, A.S.O .1988. "Agricultural marketing of grain crops in Malawi: Policy,

Practice and Problems." A paper presented at a workshop on Agricultural marketing in

The SADC region: Problem, Prospect and Practice. Lilongwe, Malawi

Omamo S.W. 1998. "Transport costs and smallholder cropping choices: An application
to Siaya District, Kenya." American Journal of Agricultural Economics 80(1): 116-123.

Ouchi, W.G. 1980. Markets Beruacracies and Claus." Administrative Science
Quarterly, 25: 125 141,1980.

Posner R. 1998. Economic analysis of law, 5th ed. New York: Aspen Publishers Inc.,









Rhodes V.J. 1995. "Cooperative enterprise and organization theory: An appraisal."
Journal of Cooperatives 10:87-91.

Royer, J.S. 1999. "Cooperative organizational strategies: A neo-institutional digest."
Journal of Cooperatives 14:44-67.

Scott, W.R. 1998. "Organizations: Rational, natural, and open systems." Upper Saddle
River, NJ: Prentice Hall, Inc.

Sexton, R.J., and I. Julie. 1993."The Competitive Role of Cooperatives in Market-
Oriented Economies: A Policy Analysis." Agricultural Cooperatives in Transition, ed.
Csaba Csaki and Yoav Kislev, Westview Press, pp. 55-83.

Staal S.J., C. Delgado and C. Nicholson. 1997. "Smallholder dairying under transaction
costs in East Africa." World Development 25:779-794.

Staatz, J.M. 1989."Recent Developments in the Theory of Agricultural Cooperation."
Journal of Agricultural Cooperation 2:74-95. Farmer Cooperative Theory: Recent
Developments. Washington, D.C.: U.S.Department of Agriculture, ACS Res. Rep. 84,
June.

Tschirley, D., M.M. Kavoi, and M.T. Weber. "Improving Kenya's Domestic Horticultural
Production and Marketing system: Current Competitiveness, Forces of Change, and
Challenges for The Future Volume." Horticultural Marketing Working Paper No.
08B/2004

Vakis, R., E. Sadoulet, and A. de Janvry. 2003. "Measuring Transaction Costs from
Observed Behavior: Market Choices in Peru." Department of Agriculture and Resource
Economics Working Paper 962 University of California Berkeley.

Vitaliano, P. 1983. "Cooperative Enterprise: Alternative Conceptual Basis for Analyzing
Complex Institutions." American Journal of Agricultural Economics, 65, 1078-1083.

Wang, N. 2003. "Measuring Transaction Costs: an Incomplete Survey." Working paper
number 2, Ronald Coase Institute.

Williamson, O.E.1975. Markets and Hierarchies: Analysis and Antitrust Implications,
Newyork : The Free Press.

Williamson, O.E. 1979. "Transaction-Cost Economics: The Governance Of Contractual
Relations." Journal of Law and Economics, 22 (2) October 233-261

Williamson, O.E.1981. "The Modern Corporation: Origins, Evolution, Attributes", Journal
of Economic Literature, 19: 1537-1568.

Williamson, Oliver E. 1985. The Economic Institutions of Capitalism, New-York: The
Free Press


100









Williamson, O.E. 1993. "The evolving science of organization." Journal of Institutional
and Theoretical Economics 149:36-63.

Williamson, O.E. (1996). The mechanisms of governance, New York: Oxford University
Press.

Wooldridge, M. J. (2003). "Further Results on Instrumental Variables estimation of
Average treatment Effects in the correlated Random coefficient model." Economics of
Letters 79:185-191.

Wooldridge, M. J. (2006). Introductory Econometrics. 4 th ed. Michigan State University


101









BIOGRAPHICAL SKETCH

Lucy Chamdimba Nyirenda was born on November 28, 1978 in Mangochi,

Malawi. She grew up mostly in the city of Lilongwe, Malawi where she also attained her

secondary education at Bwaila Secondary School up to 1995. She earned her BSc in

Agriculture (Agricultural Economics) from the University of Malawi, Bunda College of

Agriculture in 2000.

After graduating, Lucy worked in various fields which include Microfinance before

joining the Civil Service in the Ministry of Agriculture and Food Security where she has

worked as an Economist as well as an Agribusiness Development Officer.

Upon completion of her master's degree, Lucy shall return to her country Malawi

where she will continue working in the civil service on various assignments to assist in

the development of the warm heart of Africa, Malawi. Lucy is married to Lincoln and

they have two children: Chawezi age 12 and Nancy age 7.


102





PAGE 1

1 COOPERATIVE MARKETING AND TRANSACTION COST ECONOMICS: A STUDY OF BWANJE VALLEY RICE FARMERS IN MALAWI By LUCY NYIRENDA A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE R EQUIREMENTS FOR TH E DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2010

PAGE 2

2 Lucy Nyirenda

PAGE 3

3 To my husband, Lincoln, and my children Chawezi and Nancy, thank you for enduring the hard times we went through and for your understanding and always b eing there for me. It was not easy but we made it through together. To my late Dad and Mom, I am what I am today because of the principles you instilled in me Dad, Mom I made it and I am sure I would have made you proud. Lastly to my brothers and sister, Owen, Phyllis, Sunganani and Geoffrey, thank you for your support throughout my life.

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank USAID through the UILTCB program for their financial support and Ministry of Agriculture and Food Security for letting me part icipate in the program I also would like to thank my brothers and sister and friends for their moral support and encouragement they gave me throughout the study period. My heartfelt thanks go to my committee chair Dr. J.A. Sterns and Dr. Z.F. Gao for tirelessly helping me with my work. Their comments and criticisms were a great contribution to the shaping and perfecting of this thesis. Many thanks also go to the staff of Dedza DAO and especially Mr. Siyasiya and Mr. Mafosha from Mtakataka EPA for their support and input during data collection and also to the farmers of Bwanje Valley for their willingness to provide information which has enabled the accomplishment of my work. I would also like to thank the following colleagues, Mr. M. Banda and Ms J. Msosa from USAID Malawi, Mr. M. Kanjadza, Mr. K. Chaula, Mr. Mlangeni, Mr. Dzim biri, Ms H. Chima who helped me in one way or another preparing for my trip and gatherin g information on my behalf while I was in the United States. Many thanks also go t o my friends, Ms F. Nkana, Mr. P. Soko, Mr. B. Kamwana, Mr. I. Thindwa Mr. J Chiputula and Mr. T. Blare who made my stay in Florida memorable Special thanks go to my husband Lincoln whose love, care, patience, understanding and support gave me the courage to go on even when it was tough. Above all, I would like to thank God almighty for all the blessings he showered upon me and for making it possible for me to finalize my study. Without his will I would not have made it.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF ABBREVIATIONS ................................ ................................ ............................. 8 ABSTRACT ................................ ................................ ................................ ..................... 9 CHAPTER 1 INTRODUCTION, PROBLEM STATEMENT, study OBJECTIVES, TESTABLE ASSERTIONS, ANTICIPATED benefits, AGRICULTURAL production in malawi .. 11 Introduction ................................ ................................ ................................ ............. 11 Anticipated Benefits ................................ ................................ ................................ 13 Problem Statement ................................ ................................ ................................ 13 Study Objectives ................................ ................................ ................................ ..... 14 Testable Assertions ................................ ................................ ................................ 15 Agricultural Production in Malawi ................................ ................................ ............ 15 2 LITERATURE REVIEW ................................ ................................ .......................... 18 Transaction Cost Theory ................................ ................................ ......................... 18 Overview ................................ ................................ ................................ .......... 18 Bounded Rationality and Opportunism ................................ ............................. 19 Asset Specificity ................................ ................................ ............................... 20 Transaction Costs Stu dies ................................ ................................ ...................... 21 Transaction Costs and Organization Theory ................................ ........................... 23 Marketing Cooperatives ................................ ................................ .......................... 25 3 MATERIAL AND METHODS OF STUDY ................................ ............................... 29 Data Collection ................................ ................................ ................................ ....... 29 Study Area ................................ ................................ ................................ .............. 30 Sampling Design ................................ ................................ ................................ ..... 31 Analytical Framework ................................ ................................ .............................. 33 Analytical Approach ................................ ................................ .......................... 33 Descriptive Analysis ................................ ................................ ......................... 33 Analysis of Transaction Costs and Cooperative Membership .......................... 34 Test of association ................................ ................................ ..................... 34 Regression analysis: Participation model ................................ .................. 36 Regression analysis: Income model ................................ .......................... 41 Expected Relationship ................................ ................................ ............................ 44

PAGE 6

6 4 SOCIAL AND ECONOMIC CHARACTERISTICS OF THE FARMERS .................. 46 Age and Gender of Farmers ................................ ................................ ................... 46 Farmers Education Levels ................................ ................................ ...................... 47 Purpose of Producing Rice ................................ ................................ ..................... 48 Source of Income ................................ ................................ ..................... 49 Rice Production and Marketing ................................ ................................ ............... 51 Rice Production ................................ ................................ ................................ 51 Rice Marketing ................................ ................................ ................................ 53 Major Types of Rice Buyers in the Area ................................ ................................ .. 55 Rice Marketing Constraints ................................ ................................ ..................... 58 5 TRANSACTION COSTS AND COOPERATIVE MEMBERSHIP ............................ 61 Transaction Costs Sources ................................ ................................ ..................... 61 Market Informa tion ................................ ................................ ........................... 61 Availability of Few Buyers in the Market ................................ ........................... 64 Poor Road Infrastructure ................................ ................................ .................. 65 Asset Specificity Levels: Land Use ................................ ................................ ......... 65 Uncertainty Levels ................................ ................................ ................................ .. 67 Frequency ................................ ................................ ................................ ............... 69 Cooperative Membership ................................ ................................ ........................ 70 Reasons for Participating in the Cooperative ................................ ................... 71 Reason for Not Participating in the Cooperative ................................ ............... 72 Social Network ................................ ................................ ................................ 73 Membership Restriction ................................ ................................ .................... 74 Regression Analysis Results ................................ ................................ .................. 75 Participation Model ................................ ................................ ........................... 75 Income Model ................................ ................................ ................................ ... 78 6 CONCLUSIONS AND POLICY IMPLICATIONS ................................ ..................... 80 Conclusion ................................ ................................ ................................ .............. 80 Policy Implications ................................ ................................ ................................ .. 82 APPENDIX A QUESTIONNAIRE FOR THE STUDY ................................ ................................ .... 86 B MAP OF MALAWI SHOWING THE STUDY AREA ................................ ................ 97 LIST OF REFERENCES ................................ ................................ ............................... 98 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 102

PAGE 7

7 LIST OF TABLES Table page 4 1 Age of farmers ................................ ................................ ................................ .... 47 4 2 Education levels for farmers ................................ ................................ ............... 48 4 3 Purpose for rice production ................................ ................................ ................. 49 4 4 Farmers farm income ................................ ................................ .. 50 4 5 Rice production quantities, reported in 50kg bags ................................ .............. 52 4 6 Results for cross tabulation between co op membership and production ............ 52 4 7 Farmers markets ................................ ................................ ................................ 54 4 8 Type of major rice buyers in the area ................................ ................................ 55 4 9 Results for cross tabulation between buyer knowledge and sales volume ......... 58 4 10 Rice marketing constraints ................................ ................................ ................. 58 5 1 Sources of Market Information ................................ ................................ ............ 62 5 2 Information about which farmers are certain at time of sale ............................... 68 5 3 Frequency of t ransactions ................................ ................................ .................. 69 5 4 Reasons why farmers joined the cooperative ................................ ..................... 71 5 5 Reasons for not participating ................................ ................................ .............. 72 5 6 Frequencies on join responses ................................ ................................ ........... 75 5 7 Participation model results ................................ ................................ ................. 76 5 8 Income Model Results ................................ ................................ ........................ 79

PAGE 8

8 LIST OF ABBREVIATION S ADMARC Agricultural Development and Marketing Cooperation ADD Agricultural Development Division AEDC Agricultural Extension Development Coordinator EPA Extension Planning Area Coop Cooperative GDP Gross domestic product GoM Government of Malawi ICA International Cooperative Alliance IRB Internal Review Board JICA Japanese International Cooperation Agency MoAFS Ministry of Agriculture and Food Security TC Transaction costs

PAGE 9

9 Abstract of Thesis Pr esented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science COOPERATIVE MARKETING AND TRANS ACTION COST ECONOMICS: A STUDY OF BWANJE VALLEY RICE FARMERS IN MALAW I By Lucy Nyirenda August 2010 Chair: James Sterns Major: Food and Resource Economics The study investigated the role marketing cooperatives and transaction costs play rice from Bwanje Valley, Dedza district in Malawi. Production and marketing cooperatives are currently being promoted in Malawi with a goal to reduce the problems faced by small scale farmers in marketing their produce. The study specifically looked at the consideration that farmers give to transaction costs when making a decision to participate in the marketing cooperatives and consequently the effect that membership has on incomes earned from rice. The study also researched sources of the transaction cost s that th ese farmers face and if the sources are associated with cooperative membership. A probit analysis was used to establish the relationship between transactions costs and rel choice decision on whether to join the cooperative or not. The key conclusion from this analysis is that little consideration is given to transaction costs when the farmers are making a decision about partic ipating in a cooperative.

PAGE 10

10 A regression model was used to determine the relationship between cooperative membership and incomes earned from rice marketing. The results from this analysis suggest that cooperative membership has a significant positive effect on the income earned by the farmers who are cooperative members.

PAGE 11

11 CHAPTER 1 INTRODUCTION, PROBLE M STATEMENT, STUDY O BJECTIVES, TESTABLE ASSERTIONS, ANTICIPA TED BENEFITS, AGRICU LTURAL PRODUCTION IN MALAWI Introduction A gricultural marketing for small scale farmers in Malawi became an issue after several changes in policies were executed, following the Structural Adjustment Program implementation in 1981. Malawi is an agro based country with 74% of the population dependin g on agriculture for food and income. Better, more effective marketing of agricultural produce is a key strategy for improving the incomes of these farmers. The agriculture sector is dual in nature with the small scale and estate sectors. The small scale s ector contributes a higher percentage of the production and employment especially in the rural areas (more than 70% of the agricultural gross domestic product, GDP) compared with the estate sector. These small scale farmers are mostly based in the rural a reas where access to a lot of facilities, like good markets, is limited. Before market liberalization the Agricultural Marketing and Development Corporation (ADMARC) a state owned marketing board, controlled the marketing of all agricultural produce in th e country. Despite the inefficiencies associated with the board, it worked to the benefit of the small scale farmers living in the rural areas by providing a market for their production. The liberalization of agricultural marketing was expected to provid e incentives for the participation of the private sector, with consequences of competitive marketing benefiting smallholder farmers through better marketing arrangements and higher prices. However due to lack of supporting infrastructure like good roads, markets and a proper system to support the liberalization, the results have not produced the intended

PAGE 12

12 benefits. The evidence from rural Malawi does suggest that smallholder farmers, particularly, the po or have benefitted less because of unfair trading prac tices and monopsony power of private traders, and lack of reliable markets for agricultural produce a nd inputs.(E. Chirwa et.al, 2006 ) These small scale farmers now have problems of access to better and reliable rnative markets has eroded their bargaining power for good prices. This coupled with other problems associated with small scale production, like being in the rural areas where access to good roads, information and urban markets is limited, renders marketi ng of produce to be expensive, which contributes to low incomes for these farmers. One o f the proposed solutions for improving marketing for small scale farmers in order to enhance their incomes was the use of institutional arrangements such as cooperative s. These cooperatives are intended in part to minimize the cost of carrying out transactions in marketing. The cooperatives offer an alternative market for produce as well as providing access to inputs such as fertilizers, for its members. Theory states t hat use of institutional arrangements such as cooperatives can minimize transaction costs. Some empirical studies conducted earlier, for example in Canada have also shown that cooperatives can minimize transaction costs. However most of these studies wer e conducted in different scenarios as compared to small scale farmers in Malawi. Small scale farmers in Malawi offer a different perspective of the transaction cost problems because of other additional problems that they face like poor road infrastructure and a general lack of organized forms of marketing, which farmers from

PAGE 13

13 developed countries rarely face. These additional problems are likely to have an effect on the choice to participate in a cooperative. Given this context, this study focuses on the que stion: Do transaction costs, as compared to the other factors, have a significant Anticipated Benefits The G overnment of Malawi (GoM) through the Ministry of Agriculture and Food Securi ty (MoAFS), with support from development partners is promoting the formation of cooperatives as an institutional arrangement for assisting small scale farmers by helping them earn more revenue from their production. It is anticipated that this study will assist policy makers interested in the appropriateness of using cooperative s as a means to improving small scale farmers marketing of agricultural produce and enhancing their incomes. It is also anticipated that the results from the study will benefit t he farmers in Malawi specifically in the area of study, by helping them make a more informed better choice when deciding whether to participate in cooperatives or not, if their goal i s to minimize transaction costs Finally the study will contribute e mpirically to the literature on transaction cost theory and cooperatives as an institutional arrangement for minimizing cost, in relation to small scale farmers in a developing country. Problem Statement Malawi is characterized by widespread poverty with 5 2% of population living on less than one dollar a day. The majority of the population, 80%, lives in the rural areas surviving on subsistence farming with a small surplus to sell for income. National surveys estimate that crop production accounts for 74% o f all rural incomes and agriculture is the most important occupation for 71% of th e rural population (Chirwa,

PAGE 14

14 2006 ). Considering the predominance of agriculture in the lives of many Malawians for food as well as income, marketing strategies being implement ed need to benefit the farmers by lowering the cost of marketing and increasing their returns. As indicated earlier, theory and empirical evidence in developed countries indicate that cooperatives can work to minimize transaction costs incurred by its memb ers. However, small scale farmers in Malawi offer a different perspective of the transaction cost problem due to other additional problems faced in low income countries. Looking at these inherent problems faced by the farmers, and also the promotion of coo peratives by government and development partners as means for transaction costs, which are said to be high for small scale farmers. The focus was on finding out if transaction costs influence the decision to join the cooperative, despite all the inherent problems indicated earlier and the costs associated with membership. This will giv e an indication as to whether cooperatives are an appropriate solution to dealing with high transaction costs problems faced by small scale famers. It will also be beneficial to find out if the cooperative membership has an influence on the incomes of the small holder farmers. The outcome of the study will help to determine if improving an effective means Study Objectives This stud y as sesses what role transaction costs and marketing cooperatives as an institutional arrangement in marketing, play in small scale farmer marketing. Specifically the study will look at the following objectives:

PAGE 15

15 1. Identify the major sources of transaction costs that affect rice farmers (cooperative as well as non cooperative members). 2. Analyze the association between cooperative membership and transaction costs. 3. Determine if transaction costs influence the decision on whether farmers would join the coopera tive or not. 4. Determine if cooperative membership influences incomes earned from selling rice for small scale farmers. 5. Testable Assertions This study aims to contribute to the understanding of the rol e of transaction costs in small scale farmers marketing of agricultural produce, specifically rice. The following assertions are addressed in this thesis: Assertion 1 Since small scale farmers are located in remote areas far away from service providers and major consumers of farm products, the distance to the market, together with the poor infrastructure, poor access to other facilities and information is manifested in high exchange costs due to hold up problems. Assertion 2 Tra nsaction costs prevail for small scale farmers in rural Malawi. This is reflected by the higher levels of asset specificity, information asymmetry and transaction frequency leading to low returns from marketing produce. Assertion 3 Cooperative members in the study area have lower transaction costs, and this cost savings creates an incentive for farmers to join the cooperative, positively influencing their decisions to join. Assertion 4 Since cooperative members have lower transaction costs, the income s of cooperative members are higher than non cooperative members. Agricultural Production in Malawi remains the major source of food and income for the majority of Malaw ian households and also a major foreign exchange earner for the country, contributing over 80% of total foreign exchange into the country. The importance of agriculture in Malawi is evident in

PAGE 16

16 its share to the GDP for the country. Currently agriculture con tribu tes over 35% to the The agricultural sector in Malawi is dual in nature with both a small scale and an estate sector, with the small scale sector being the larger sector. Production for most of t he crops is by the small scale sector. The major crops grown in the country include maize, which is the staple food for the country and usually grown for home consumption, tobacco, which is the major foreign exchange earner for the country, rice, tea, cott on, coffee, and ground nuts. Rice production and marketing in Malawi ; with rice only coming second in most parts of the country Rice is grown in districts that are along the valleys of Lake Malawi and Lake Chiuta. Rice production is mainly by small scale farmers. Most of the rice producing sites were developed into irrigation schemes during the post colonial era. Due to transition problems in management of the schemes, the land was distributed to small scale farmers for the production of rice in the previously government owned schemes. In addition to these irrigation schemes other farmers grow their rice in upland areas where it is usually swampy during the rainy season. Presently the irrigation schemes are being rehabil itated with the help of scale farmers to stop reliance on rain fed production and start production through irrigation. Before market liberalization the marketing of agricultural products was controlled by th e state owned agricultural marketing board. Farmers sold all their surplus produce to the board. After market liberalization private traders constitute the majority of buyers

PAGE 17

17 in the marketing of agricultural produce including rice. These private traders us ually purchase paddy which the y later either sell to rice milling companies or polish it and sell at retail to consumers in the city. The selling of paddy entails very low value addition to the rice on the side of the farmers. This can have an influence on the price that the farmer is offered. Rice sold to consumers by the middlemen is usually sold unpacked. It is either weighed and sold per kilogram or it is sold using plates and charged according to the size of the plate. The promotion of marketing cooper atives by government has opened a new market for farmers marketing rice. These rice marketing cooperatives process rice from their member farmers for sale directly to the consumers through supermarkets. However private traders remain the biggest source of market opportunities for most rice farmers due to low operational scale for most of the cooperatives, which then results in rice.

PAGE 18

18 CHAPTER 2 LITERATURE REVIEW Marketing of produce involves carrying out transactions and transactions are said to have a cost. Different frameworks have been provided to describe the aspects that surround the cost of carrying out a transaction. Similarly literature has provided a number of ways to manage or minimize the cost of transacting, one of which is the use of alternative institutional arrangements besides market exchange. The paper reviews some of the literature on transaction costs, both theory and empirical, and the ways to minimize th ese costs for small scale farmers. Transaction Cost T heory Overview The transaction cost paradigm was pioneered by Coase in 1937. In his article, The Nature of the Firm, Coase argued that market exchange is not without costs. He recognized the role of tran saction costs in the organization of firms, and other contracts. Transaction costs include the costs of information, negotiation, monitoring, co ordination, and enforcement of contracts. He explains that firms emerge to economize on the transaction costs o f market exchange and that the boundary of a firm or the extent of vertical integration will depend on the magnitude of the transaction costs. The work of Williamson ( 1993, and 1996) on the economics of organization and ine of thinking. Williamson combines the concepts of bounded rationality and opportunistic behavior to explain contractual choice and the ownership structure of firms. Opportunistic behavior manifests itself as adverse selection, moral hazard, cheating, an framework, a trade off has to be made between the costs of coordination and hierarchy

PAGE 19

19 within an organization, and the costs of transacting and forming contracts in the market (Drugger, 1983). This trade o ff will depend on the magnitude of the transaction costs. Williamson rests his arguments on the bounded rationality and opportunism assumptions. He states that if these assumptions are not valid then the effects of the variables that affect the transactio n costs will not be valid. Bounded Rationality and Opportunism In transaction cost economics, all players are assumed to behave rationally. Their bounded rationality results, in part, from the fact that they have limited resources, time and energy. Bounded rationality also refers to the fact that people have limited memories and limited cognitive processing power. We cannot assimilate all the information at our disposal, nor can we accurately work out the consequences of the information we do have. Herbert Simon originally defined bounded rationality as the behavior that is intendedly rational but only limitedly (as cited by Williamson 1979 ). Others like Posner describe bounded rationality as not only being that information is costly to acquire and to proc ess, but also refers to the impossibility of thinking through the complex but well struc tured problems (Posner R. A 1998 ) Bounded rationality affects the certainty of a transaction. The transaction cost economics approach focuses on how the characteristic s of a transaction affect the costs of handling it through markets, bureaucracies, and other forms of organization. Williamson identifies the critical dimensions of characterizing a transaction and links these to the institutional governance structure of t ransactions. The principle dimensions describing a transaction are uncertainty, frequency of exchange, and the degree to which investment are transaction specific. Transaction costs include the costs of gathering and processing the information needed to ca rry out a transaction, of reaching

PAGE 20

20 decisions, of negotiating contracts, and of policing and enforcing those contracts. All these transaction costs derive from a combination of bounded rationality (which reflects both imperfect information and a limited cap acity to analyze it) and opportunism, which information about the future, all contracts are said to be necessarily incomplete. If people were never opportunistic, however, inc omplete contracts would not lead to contract enforcement problems; contracts would simply state that if unforeseen contingencies arose, the parties would act in a manner acceptable to all. Asset Specificit y ognizes that transaction costs are more likely to be important when economic agents make relationship specific investments, that is, investments are specific to a particular group of individuals or assets. The party which makes relationship specific invest ments is susceptible to the holdup problem. This leads us to the problem of asset specificity. This is because, once this kind of investment is made, that party is somewhat locked in the relationship: there is no competitive price for such investments once they are made. And furthermore, Williamson claims that every contract is inherently incomplete in that people cannot foresee all the contingencies of the future. The existence of incomplete contracts makes the ex post surplus sharing sometimes unrelated t o ex ante investments. This is similar to what happens in a group setting. As a result, incomplete contract and relationship specific investment leads to under investment, if the two parties are separate firms. Williamson claims that integrating a transact ion into the firm mitigates this opportunistic under investment.

PAGE 21

21 Farmers with specific experience in production of a particular crop, and/or production investments related to a particular crop, will have h uman and physical asset specificity and are liable to hold up problems. A study in Kenya used proxy indicators such as perishability of products and period of production, to analyze the conditions of asset specificity and uncertainty for etermine the expected institutional arrangement for linking producers and exporters/processors (D. Tschirley et al., 2004 ). The study found that the dominant institutional arrangement for coordination was that of long term contracts and vertical integratio n, rather than spot market exchange. Transaction Costs Studies There have been a number of fairly recent applications of transactions cost economics in different fields of the food and agricultural sector. Examples of these studies are Staal, et al. (19 97), Key, et al. (2000) and Hobbs (1997). Very few empirical studies have actually measured transaction costs to date. Staal, et al. (1997) asserts that the limited empirical evidence on the nature and importance of transaction costs is mainly caused by co nceptual and measurement difficulties. For example, when transaction costs are sufficiently high such that they prevent exchanges from occurring, then, by definition, these costs cannot be observed because no transaction took place. The available studies h ave t ended to focus on distance to market as a single indicator of transaction costs (Omamo, 1998 ). One of the earlier studies in agriculture to carry out empirical measurement of transaction costs was the innovative approach by Hobbs (1997). Hobbs carrie d out a study about measuring the Importance of Transaction Costs in Cattle Marketing The

PAGE 22

22 study revealed that some transaction costs variables (such as grade uncertainty, risk of not selling, time spent at the auction ) were significant factors affecting t he choice of either live ring auction or direct to packer sales. A number of studies in Africa have employed a variety of techniques to measure transaction costs in small scale farming systems (M akhura 2001; Staal et al 1997; Matungul, Lyne & Ortmann, 2001). Most of these studies have measured the effect of transaction costs on small scale farmers in relation to selection of marketing channel and commercialization of the farmers. Few studies have considered the effect of transaction costs on decision to participate in a cooperative. Participation in organizational arrangements like a cooperative can be a result of many factors, one of which could be transaction costs. If the f armer considers the transaction costs he incurs to be high t he n he will be exp ected to decide to join the cooperative in order to reduce his transaction costs. This is based on the theory by Williamson that economic agents will choose institutions, organizational forms and transactions that minimize the cost of exchange One way of determining if the theory applies to small scale farmers is opt for the organizational arrangement. Transaction cost theory indicates that vertical integration can low er transaction costs. However most of this research has been in relation to well established organizations. For example a study in Canada for milk producers compared transaction costs of farmers selling through the marketing board vis a vis farmers selling directly to the processors. As will be discussed later in the proposal, this study provides a similar scenario for farmers selling through the marketing cooperative and farmers selling

PAGE 23

23 directly to buyers. The r esults from the study in Canada found that fa rmers selling their milk through the marketing board incur less transaction costs compared to farmers who sell directly to buyers ( Royer et al, 2003 ). There are many differences between these farmers and the small scale farmers in a developing country. Sm all scale farmers in Africa have other challenges, like low literacy levels, poor infrastructure, small scale of production, lack or limited access to production inputs, of which in most of the study scenarios are not major challenges. These differences p rovide a different scenario to the studies conducted in developed countries. Tran saction Costs and Organization T heory Transaction costs theory examines the comparative economic costs of transactions, which are the transfer of goods or services across a te chnologically separable interface (Williamson, 1985). This theory explores the costs associated with the organizational forms used to complete transactions when the exchanging parties are risk neutral, predicting that the more efficient governance structur e i.e., the organizational form that reduces transaction costs more will be chosen The general hypothesis of this strand of the New Institutional Economics is that institutions or organizational forms are transaction cost minimizing arrangements that may change and evolve with changes in the nature and sources of transaction costs. Coase (1937) pioneered this work when he argued that market exchange is not costless. Transaction cost theory is based on three behavioral assumptions: bounded rationality, opportunism, and risk neutrality. This implies that a transaction is associated with contractual risks, for example the opportunistic behavior of one of the contracting parties, or the so called hold s to

PAGE 24

24 (Royer, 1999 ). Asset specificity is important in transaction cost theory (Williamson, 1985) because rising asymmetric specific asset investments make markets less co mpetitive by reducing the number of potential trading partners, locking one partner to the other and offering increasing incentives for renegotiation. As economic incentives to behave opportunistically rise, the preferences of the contracting parties diver ge: while the exchanging firms have an incentive to jointly maximize long term profits, each also wishes to maximize its return in the short term by appropriating as much of the gains as possible whenever contractual changes are required ( Williamson, 1979) The condition of diverging preferences is known as goal incongruence, or a lack of overlapping goals (Ouchi, 1980). Cooperation is difficult to achieve in conditions of goal incongruence because the individuals involved each follow their own differing ob jectives ( Maho ney J.T &R.C McNally, 2004 ) The conceptualization of transaction costs theory assumes goal incongruence is given, although some transaction costs theorists acknowledge that incentive alignment or loyalty may reduce goal incongruence (Alchia n & Demsetz, 1972 ; Williamson, 1975)]. Organization theorists, on the other hand, place particular emphasis on the ability to reduce goal incongruence through means other than incentives because doing so may increase exchange efficiency (Ouchi, 1980). Orga nization theory adds cultural control as a mechanism to reduce goal incongruence. Cultural control focuses on aligning the views of governance structure members through organization culture, which is a fairly stable set of assumptions, beliefs, meanings, a nd values that individuals use to orient their thinking and to guide their actions (Scott, 1998 ).

PAGE 25

25 Transaction costs theory shows that transaction cost problems can be dealt with using the different forms of governance structures. Organizational theory for ms the basis for the concepts underlying the different governance structures. In trying to deal with transaction costs we cannot run away from organizational theories. framework, a trade off has to be made between the costs of coordination and hierarchy within an organization, and the costs of transacting and forming contracts in the market (Williamson, 1985 ). This trade off will depend on the magnitude of the transaction costs Marketing Cooperative s Cooperatives and farmer organizations ar e institutional arrangements, the importance of which has re emerged recently as small scale farmers in developing countries seek ways to organize themselves in the wake of agricultural market liberalization. Rhodes ( 1995 ) defines a cooperative as a busine ss firm owned and operated by a voluntary association of member patrons for mutual benefit. He goes on to describe those members of a cooperative, that as owners they have the additional tive that they can affect services and activities. Opportunism from the side of the processor and marketer is often mentioned as one of the main reasons for farmers to set up a co operative and carry out the processing and marketing of farm products under their control. Farmers can prevent being held up by internalizing the transaction, that is, by integrating forward via the creation of a proprietary co operative firm. Whether farmers will do so, also depends on the type of farm product (perishable or not) and the size of relationship specific vertically via a cooperative firm to avoid opportunistic behavior are greatest where the

PAGE 26

26 proportion of sunk costs to total costs at the time of the transaction is high and the product is highly perishable, making its transfer to alternative markets on short notice very difficult. Fruits, certain vegetables, an 1 9 89). Theory and some empirical studies suggest that the advantages of organizing farmers into groups include, among other factors, a reduction in the transaction costs of accessing input and output markets, as well as improving the negotiating power of smaller farmers vis vis large b uyers or sellers ( Iliopoulos & Cook, 1999) Other studies looked at the characteristics of transactions between farmers and their mode blending market forces with element s of internal organization designed to ( Iliopoulos & Cook, 1999) The history of traditional cooperatives, on the other hand, suggests that cooperatives have not always been successful at serving the needs of its members, and their popularity had waned in the few decades preceding the 1990s. Cooperatives suffered from various organizational problems and a lack of clearly defined property rights assignments resulting in opportunistic behavior (such as free riding, moral hazard, agency problems, etc.), bureaucratic inefficiencies, and under investment in the cooperative (Cook, 1995; Cook & Iliopoulos, 1999 ). Other research building on agency and game theory suggests that traditional cooperative principles undermine optimal resourc es allocation and investment policies ition (Sexton and Julie 1993; Staatz, 1989 ). In other words, major problems of cooperative farming appear related to

PAGE 27

27 membership desertion (Barham and Childre ss, 1992), heterogeneous membership occasioning free riding behavior and limited investments and capital mobilization due to horizon problems. However cooperatives remain one of the most promoted way of dealing with small scale farmers marketing problems i n developing countries. The literature review has shown that transactions are not costless and that there are many ways of looking at the cost of carrying out a transaction. The theory that was transaction costs studies. It has however been established that transaction costs are difficult to measure and most studies look at the cost of transacting in one contractual form or the other to establish the effects of transaction costs. One of the areas that were looked at is the use of vertical integration or institutional arrangements to minimize transaction costs. Most of the studies in this area have looked at the effects of transaction costs in the different forms of organizational arrangements to d etermine the best organizational form to carry out a transaction. For small scale farmers, studies in this area have focused on the found from the studies conducted on whether the transaction costs are considered by the small scale farmers in deciding to participate in a cooperative, a specific institutional arrangement that can potentially lower transaction costs for these farmers Institutions have also been said to be cost minimizing arrangements and that economic agents choose institutions or organizational forms that minimize the cost of transacting. Results from some studies that were carried out, found that organizational forms such as cooperatives lower the costs of transactions for farmers. These lower

PAGE 28

28 costs of transacting could be an incentive for farmers as economic agents to choose to participate in such arrangements in order for them to minimize costs. Since institutions and the institutional framework provide the incentives for production and for people to engage in economic activity, an institutional analysis is required to explain why the cost of transacting is so high in developing countries. The frequent occurrence of market failure and incomplete markets (because of higher transaction costs and information asymmetries) in developing countries cannot be explained by conventional neo classical economics and requires an institutional analysis. Many of the institutions or formal rules of behavior that are take n for granted in developed countries and that facilitate market exchange are absent in low income countries. Therefore, theories of institutional economics provide a useful framework that could help determine the types of institutions needed (either formal or informal) to improve agricultural marketing in developing countries.

PAGE 29

29 CHAPTER 3 MATERIAL AND METHODS OF STUDY Data Collection Data was collected through one to one interviews using a semi structured questionnaire (See Appendix 1). The questionnaire wa s administered to sampled cooperative members and non cooperative member farmers. Data that were collected include s ocial economic characteristics of the farmers, types of markets the farmers access, market information sources and use, type of rice buyers in the markets, period taken to sell rice, frequency at which the farmers sell their rice, and rice prices. The following information was also collected on cooperative membership : why they joined if they are cooperative members and why not if they are non cooperative members, period of membership, knowledge of members before joining the cooperative, their intention on cooperative membership if they were given a choice to join. Prior to data collection an application was made to Internal Review Board (IRB) for approval of the study. An application form together with the questionnaire and consent forms were submitted to the IRB. The consent form was translated into Chichewa which is a local language as was required by the IRB. Data collection proceeded after IRB approved the study. Pre testing was done before the actual data collection. This was done to get feedback on the quality of the questionnaire so that all omitted, irrelevant or misunderstood questions and mistakes could be r ectified. The pre testing was done with approximately 20 farmers in the same study area. Some minor changes were made to the questionnaire following the findings and comments from the pre testing.

PAGE 30

30 The actual data collection covered a period of 3 weeks fro m 26 th October to 11 th November, 2009. Two enumerators were hired to assist with the data collection. These enumerators are Bunda College of Agriculture graduates who had just finished their final year at college. Before going to the field for data collect ion, training of the enumerators was conducted to reduce human error and also to ensure that the questions were clearly understood and properly translated into Chichewa 1 since they were originally in English. Study Area The study was conducted in Malawi. Malawi is in the southern part of Africa. The country is divided into three regions geographically Southern, Central and Northern regions. The regions are further divided into 28 administrative districts In the agricultural sector, the country is divid ed into 8 divisions called Agricultural Development Divisions (ADD). These ADDs are constituted of districts which are controlled by the District Agricultural Offices (DAO). There are a total of 28 DAOs in the country divided among the ADDs. These ADDs ha ve between 2 to 7 DAOs in one ADD. The highest number of DAOs in one ADD is 7 which is Blantyre ADD. Each district is further divided into Extension Planning Areas (EPA) which are further divided into sections. The sections are then organized according to villages in the area. The extension workers are based in the sections and work with the villages that are within the section. This study was conducted within Dedza district, particularly Bw anje valley in Mtakataka EPA. Dedza DAO falls under the Lilongwe A DD which is in the central 1 Chichewa is the local language through which the questionnaire were administered

PAGE 31

31 region of the country. The district was purposively selected because the valley is a rice growing region and also because they have an established rice production and marketing c ooperative. The cooperative was also purposively se lected from the district since it is the only rice production and marketing cooperative in the district. The study target ed Bwanje Valley Rice Production and Marketing Cooperative (BVRPC) This cooperative has been in operation for more than 10 years Refe r to appendix B for a map of Malawi showing the study are a. Sampling Design According to Cochran(1977), a formula for determining a sample size expressed as a percentage is; 1) where t 2 = the standard deviation score that represents the probability level of a variable of falling within a confidence interval when the variable is normally distributed (p)(q) = Variance j 2 = confidence interval We yielded the following results after incorporating our data v ariables into the formula: this is what were represented as follows: (3 2) 384 The probability level and conf idence interval of 1.96 and 0.05 respectively were used as these are the commonly used estimates and normally accord estimation process

PAGE 32

32 efficiency. The variables making up the variance represent the proportion of farmers who are cooperative members and tho se who are non cooperative members., Czaja R. and Jonny B. (1995) recommends that a 50% proportion for each is ideal in a situation where the it is difficult to source the specific estimates for these proportions. The study needed to collect a sample size of about 384 to be able to represent the target population of our study. However, due to budgetary constraints, time and other factors, the study managed to collect a sample size of 190 This is still a significant figure ge sample and it was randomly collected. A two stage sampling was employed to come up with the sampling units. The selection of Lilongwe ADD and Dedza district was the first stage. Dedza DAO has only one rice production and marketing cooperative hence the cooperative was purposively selected. The second stage was the selection of villages in the EPA where the cooperative is based. The following villages were selected Mchanja, Fole, Mthembanji, Bwanamakowa, Chatewa, Madziansatsi, Kafulama, Ndongwe. It was fo und out later that Ndongwe and Chanja villages belong to Golomoti EPA. They were still selected due to their closeness with the irrigation scheme and since most of the farmers from the area were cooperative members who were farming in the irrigation scheme A stratified sampling technique was used where the farmers were stratified into two strata, rice farmers in a cooperative and rice farmers not in a cooperative. U sing a list of f arm families maintained by the Agricultural Extension Development C oordinat ors (AEDC) in the area as a sampling framework, a total of 200 farmers were then randomly sampled from the strata 110 from the cooperative and 90 non cooperative, This represents 3.3% of the total population of farm families in the villages from the ar ea

PAGE 33

33 under study. For coop members a total of 110 were sampled from 2062 cooperative members representing about 5.3% of the total cooperative membership. For the non coop members a total of 90 were sampled from a total population of 3919 farm families who ar e not coop members in the area representing 2.3% of the population. A total of 179 interviews were achieved due to other problems like movement, death and unwillingness to participate in the study of the sampled respondents. Analytical Framework Analyt ical Approach SPSS 15.0 and Stata were used for data analysis. Both qualitative and quantitative analytical approaches were employed in analyzing the data. The two approaches were used in order to widen the explanation base for the results that were found. The data have been analyzed in two parts. The first part includes the descriptive analysis, cross tabulations and contingency tables. The Chi square is used to test the presence of significant association between cooperative membership and proxy variables that are used to determine the extent of transaction costs. The second part is a regression analysis of survey data. Descriptive Analysis Frequencies, means, and percentages were computed to determine and rank the constraints farmers face in marketing the ir rice. The descriptive statistics were also used to summarize and categorize the information that was collected. Information on marketing constraints, information sources and information use were analyzed to determine the sources of transaction costs the farmers face in the area.

PAGE 34

34 Analysis of Transaction Costs and Cooperative Membership Test of association Tra nsaction costs, especially for small scale farmers, are difficult to measure quantitatively. This research focused on assessing and documenting speci fic dimensions of a transaction (asset specificity, transaction frequency and information asymm etry). As noted in the literature review ( Chapter 2 ), proxies for information and search costs such as availability and access to market information, bargainin g and negotiation costs such as number of available buyers, or ho w long it takes for farmers to sell their produce, have been used to determine the extent of transaction costs farmers incur while marketing their produce. When used in these studies, these p roxies provided keys to understanding organizational effects on transaction costs for individual farmers. For this thesis, the analysis is comparative, assessing the cost of conducting transactions in one organizational or contractual form relative to anot her. With this concept in mind, the goal of the tests of association was not to measure the absolute amount of transaction costs, but the relative ranking of transaction costs associated with different organizational or contractual choices. Furthermore, a s reported in numerous project reports and related unpublished empirical studies, transaction costs are not directly measured ( Wang N., 2003) Certain proxies, such as uncertainty, transaction frequency, asset specificity, opportunism, and so on, are used instead, which are believed to critically affect the cost of transactions. A statistically significant relationship between the chosen proxy and organizational governance suffices to make the point clear that economizing on transaction costs is the unifyin g logic behind various contractual arrangements of production. (Wang N., 2003). Hence, these studies are

PAGE 35

35 able to move around the tricky question of quantifying the absolute level of transaction costs. For this thesis the following variables were used to a pproximate the extent of information costs and bargaining costs. These variables are considered to have an effect on transaction costs Information and search Knowledge about buyer Knowledge about distant market Time to sell in market Knew price in marke t of transaction Knowledge of prices in the other markets Frequency of sale Bargaining and negotiation Average price received (K/kg) Number of available buyers Farm experience (years) Land redeployment to other crops Other control variables Farm Gender of the farmer Membership in cooperative Land holding size The study concentrate d on the search and negotiation costs to establish the relationship between the transaction costs and cooperative membersh ip due t o the nature of the data that was expected to be collected. The inability of the target population to access well established markets and other limitations result in sales without properly prepared contracts with buyers. This makes it difficult to identify enf orcement and monitoring costs and determine the relationship. Proxies will be used

PAGE 36

36 in the to establish the existence of transaction costs among the farmers due to the difficulties associated with the measurement of transaction costs. Regressio n analysis: Participation model The regression analysis will be used to determine the significance and direction of the effect of the proxy variables on the choice decision to join a cooperative or not and the effect of membership on income. During the pr e test it was discovered that cooperative membership for Bwanje Valley Rice Production and Marketing Cooperative is restricted to those farmers who have plots of land in the irrigation scheme. With the restricted participation it would be difficult to esta blish the effect of the TC variables on the decision to join the cooperative. In order to find the effect on ones decision regarding cooperative participation, the study considered asking the farmers their intention on choice given a scenario where members hip is not restricted. The choice that they indicated they would make was used in finding out the effect of the TC variables on the choice decision of participating in a cooperative. The TC variables that were used in the study are those that are said to affect the three principle dimensions describing a transaction as identified by Williamson, 1979 The dimension s include uncertainty, frequency of exchange, and the degree to which investment are transaction specific. The assumption is that a change in the variables affecting these three dimensions will have an effect on the cost of the transaction. The variables have been discussed below in relation to the transaction characteristics; Asset specificity : This will be determined by the farmer s level of ex perience in the production, ease of redeployment of the resource ( rice field ) and probability of being held up by the buyer ( dependency on the buyer). Asset specificity has an impact

PAGE 37

37 on the negotiation cost and higher levels of asset specificity lead to high negotiation costs. Uncertainty: Lack of sufficient information contributes to bounded rationality and uncertainty about the transaction. Finding a buyer for the produce and assessing the fairness of his or her offers requires availability of suf fici ent information. Search cost will be determined by level of knowledge or the amount of information (about the markets, buyers and prices) that the farmer has at the time of sale and source o f information Frequency of transaction: Transaction costs are tra nsaction specific. Farmers who carry out many transactions each with a different buyer are likely to incur more costs because for each of the transactions they carry out they will need to search for a buyer. The cost of carrying out a transaction is also i ncreased because there is a highe r probability of opportunistic behavior if there is no buyer seller relationship. For such farmer the higher the number of transactions, the more likely the farmer is a member of the cooperative, as membership will help red uce the transaction cost incurred for each transaction. Other variables that were expected to affect the choice decision, to be a coope rative member or not, were used as control variables in order to reduce the omitt ed variable bias. The following variable s were used as control variables e ducation level of farmer, gender of farmer, age of farmer and distance to the nearest market. A regression was run to establish the relationship between transaction costs and the choice of cooperative membership. Econ ometrically the specification pr oblem followed a latent regression model y = X + (3 3)

PAGE 38

38 Where y the latent variable is unobserved (i.e., the importance of tra nsaction costs in the decision process). What we observe is a dummy variable y defined by y = 1 if y > 0 y = 0 otherwise Here X is the vector of transaction costs proxies and other variables that influence cooperative membership choice decision and 1 is the error term following normal distribution. Probit method was used to estimate this equation. The likelihood function of this model can be written as: y i X i ( X)] (3 4) The marginal effects of this model can be written as: = (3 5) To test the hypothesis, a probit analysis was run on the model indicated below. Join = 0 + 1 AGE + 2 A GESQR + 3 F AEXPER+ 4 B UYDE P + 5 A SSETRED + 6 BUYINFO + 7 M KTINFO + 8 P RICINFO + 9 EDUC + 1 0 FREQ + 1 1 GNDER + 1 2 DIST+ 1 3 TYPBUYER + 1 4 MEMBER + FAEXPSQ + 1 (3 6) In the model, Join was the de pendent variable which took on a value of 1 if the respondent indicated that they would join the cooperative and 0 o therwise. The following are the in dependent variables that were used in the model AGE = Age of the respondent AGESQR = Age of respondent squared FAEXPER = Farmers experience in rice production FAEXPSQ = Farmer experience squared BUYDEP = Farme r level of dependency on buyers

PAGE 39

39 ASSETRED = Ability to use land for production of other crops BUYINFO = If farmer has information about buyer before sale 1= Has Information 0 = other wise MKTINFO = If farmer has information about other mar kets 1 = has information 0 = otherwise PRCINFO = If farmer has information on prices in other markets 1 = has information 0 = Otherwise FREQ = Number of times farmers goes to sell rice EDUC = Farmers level of education DIST = Distance to the nearest main market GNDER = Dummy for gender 1 = If farmer is a male 0 = If farmer is female TYPBUYER = The type of buyer the farmer typically sells to (Score variable; 1 = Wholesaler, 2= Wholesaler and Retailer, 3= Retailer, 4= Reta iler and Wholesaler) Description of variables These affect the level of asset specificity. The TC framework states that substantial asset specificity increases the costs of safeguarding inter firm agreements because of the prospects of opportunistic behavior of the trading partners (Williamson 1985). Such

PAGE 40

40 costs are associated with haggling verification problems, and other bargaining difficulties that may arise because of inappropriate o r ambiguous performance measures, hidden information, or moral hazards. The variable TYPBYR which stands for type of buyer was included in the model after establishing that it had some influence on the negotiation abilities of the farmers. This was establi shed during the discussion held with the farmers on their ability to negotiate. Knowledge about the buyer/s, knowledge about availability of markets, knowledge about prices: In order to carry out a market transaction it is necessary to discover who it is that o ne wishes to deal with. Therefore information becomes vital for the transaction to take place. These variables have an effect on information asymmetry. This can impact on the level of uncertainty and opportunistic behavior which will have a bearing on the search cost. The expectation is that when faced with high costs the farmer is likely to become a member of a cooperative. Education: It is e xpected that farmers with more years of education will be able to understand the benefits of membership to a cooperative and they are more likely to join it. Age and Agesq: Age is the ag e of the respondent and Agesq the age square which was used to capture the diminishing effect of a respondents age Expericence : Experience refers to the period the farmer ha s been growing rice Distance: Distance to nearest reliable produce market is expected to have a positive relationship with participation in the cooperative. The further away the nearest reliable produce market is the more likely it will be for farmers to become cooperative members in order to access a markets through the cooperative.

PAGE 41

41 Regression a nalysis: Income m odel The second model that was run was the income model. I t was expected that the incomes of the cooperative members will be higher than the inco mes of non cooperative members holding all other factors constant following the low transaction costs that are said to be associated with cooperative membership Cooperative membership in this case is considered to be an endogenous variable and as a resu lt it will be included as an instrumental variable in the income equation. In an Instrumental Variable setting, we need to find a variable which should be highly correlated with the treatment but not with the error term or the other explanatory variables i.e. where u is an error term (Wooldridge 2003) Wooldridge (2003) showed that (IV) can consistently estimate the population average treatment effects under weaker assumptions than those needed for two st age least squares plug in estimators proposed by Heckman (1998). Blundell (2002) also indicated that given cross sectional data, instrumental variables can be a reliable tool for evaluating program impacts as long as it satisfies the required assumptions. Under the required assumptions an (IV) provide s the required randomness in the assignment rule. Thus the relationship between the instrument and the outcome for different participation groups identifies the impact of treatment avoiding selection problems. Selection problems arise because ones decision to join a program may be determined by other factors for example education. Higher education might influence their incomes positively and entailing that the high incomes would be as a result of the high educat ion and not necessarily the program participation. The problem could be written as follows;

PAGE 42

42 Y 2 = 1 x + d 1 z member +e 1 (3 7) z = 2 w + e 2 (3 8) Where Y 2 will be the logarithm of rice incomes while z is a dummy for membership into the cooperative and X contains age, educat ion and gender of the farmer l and holding size, m embership to cooperative, production, off farm income, In the model Z will be the coop membership while w will contain Distance Knew coop members an d experience, these variables will be used as instruments for cooperative membership. It is assumed that these variables are correlated with cooperative membership but are not regressors in the income model and the are not correlated with the error term. Following this the empirical model for income can be written as; Log income = 1 + 2 quantity + 3 age + 3 agesqr + 4 educ + 5 off farminc + 6 coop + 7 land + 8 male+ (3 9) Where: Dependent variable is t he log income This is the log of rice incomes the farmer has attained in the previous growing season. The income was determined by multiplying the quantity the farmer sold and the average price at which the farmer sold his rice. Log of income was used in order to reduce the fluctuations in the income data. The following are the independent variables: Quantity: The quantity the farmer produced du ring the period. Since the study is focusing on rice incomes then the quan tity the farmer is able to produce will positively affect the income they realize from the rice sales. This variable was later dropped

PAGE 43

43 because it was highly correlated with the land variable which is the size of the plot of land the farmer has. Age: In the model age is the age of the farmer Age is likely to have a positive and negative effect on income because the older people are less likely to work as well as the younger ones and t his will affect the output. Age square was used to capture the diminishing effect of age on the income Education: T he number of formal education year s attained by the farmer of consequently the income realized. Off farminc. This is the availability of off farm income for the famer For farmers with more sources of off farm income they are likely t o have more purchasing power to access inputs and other production requirements. This will have an impact on the quantity and quality of the rice they sell. In the end it affects the income realized from rice sales. Coop : This value is from the estimated model of cooperative membership Land : This is the size of the rice fields the farmer used to produce the rice during the period being studied. It is expected the farmers with larger fields will have higher incomes as a result of higher production for the ir field. Gender: This is the gender of the farmer taking the value of 1 if the farmer is a male and 0 for female. The following instruments were used for cooperative membership to deal with the problem of selection bias.

PAGE 44

44 Experience: This is the period the farmer has been growing rice. The assumption is that farmers who have been growing rice for a long time has a higher chance of getting a plot of land in the scheme and participating in the cooperative. Knew coop members : This is a dummy variable takin g a value of one if the farmer knew any coop members and 0 if not. Distance: This is the distance to the nearest market. Expected Relationship The table below shows the relationship that is expected between the dependent variable and the independent varia bles listed in the models explained above These expected relationships were derived from theory and other similar empirical studies reviewed in this study. In some instances the relationships are difficult to determine as they might either be negative or positive.

PAGE 45

45 Table 3 1. Variables and their expected relationship Variable Description Variable Participation Income Number of buyer farmer encounters No buyer Farmers experience in rice production faexper +/ Age of the far mer Age +/ +/ Level of buyer dependency buyerdep + Asset redeployment Assetred Information about the market marketinfo Knowledge about the prices Priceinfo Number of times farmer sells frequency + Gender of the far mer Male +/ +/ Distance to the market Distance + Quantity produced in the previous season quantity + Education level of the farmer Education +/ + Coop + Farmers off farm income offfarminc + Land holding size Land +

PAGE 46

46 CHAPTER 4 SOCIAL AND ECONOMIC CHARACTERISTICS OF T HE FARMERS Age and G ender of F armers A total of 200 rice farmers were sampled in the study with 110 farmers from the coop and 90 farmers from a target population of n on coop members as described in chapter 3. A total of 101 and 78 farmers respectively were successfully interviewed. Results from the study show that no gender category is fully dominating in rice production. For non coop members 50% of the sample were mal es and 50% were females while for coop members 54.5% were males and 45.5% were females. For the coop members, the percentage composition from the sample does not follow the composition of the total membership of the cooperative which has a higher percentag e of females than males, 53.4% females and 46.6% males. However the results show that no gender category largely dominates the participation in the coop which is similar to the total membership composition as shown by the small difference in the percentage of men and women in both the sample and the total population. It can be noted that age was somewhat evenly distributed in both samples (coop members and non coop members). No age group was dominating as shown by the results. From the sample 28.7% was betw een the ages of 18 to 24, 21.8% were in the age group of 25 to 34 while 19% were in the age group between 45 to 51 years Table 4 1 shows the age distribution for the respondents according to their membership status. A difference can be noted in the age gr oup of 25 to 34 where for the non cop members only 12% belong to this group while for the coop members it was 21%. For the higher ages participation in the coop is lower as shown by the dwindling

PAGE 47

47 percentages in the sample while for non coop members the per centages are going up. This shows that currently, farmers tend to participate more in the groups when they are younger and less when they grow older. Only 3% of the coop members were above the age of 50. This may be a generational effect, such that as current young members age, they will maintain their cooperative membership, thus in time, there will be less difference across age categories in the future. Table 4 1. Age of farmers Farmers Education L evels Education plays an important role in economic development. In addition farmers who are literate are e xpected to better understand instructions and comprehend issues. When farmers attain a certain level of education they adopt new technologies faster than the less educated farmers. Their level of understanding is expected to improve and enable them to part icipate in groups. In this study farmers with no education level are those that have not had any kind of formal education. From the sampled farmers 62.7% of the coop members had some formal educati on up to 5 years while 27.4 had never had any kind of fo rmal education. Only 4 percent had more than 8 years of formal education. Similarly for non coop members the majority in the sample had up to 5 years of education,51.3% and 34.9% never had any kind of formal education. Only 2 .6 % had more than 8 years of ed ucation. Age Coop membership status Coop members Non Coop members 18 to 24 years 28.7% 26.9% 25 to 34 years 21.8% 12.8% 35 to 44 years 26.7% 29.5% 45 to 51 years 19.8% 24.4% Above 51 years 3% 6.4%

PAGE 48

48 Table 4 2. Education levels for farmers Education level Coop membership status Coop members Non Coop members None 27.4% 34.9% 1 to 5 years 62.7% 51.3% 6 to 8 years 11.5% 5.9% More than 8 years 4% 2.6% With most of the members having very low education levels or none at all, it provides a challenge for the cooperative in terms of management. This is so considering that management positions in the cooperative are occupied by the farmers themselves, who are responsible for decision making fo r the cooperative. These proposed decisions made in the cooperative are then voted for by the members who have one vote each regardless of the number of shares one owns or the amount of business one does with the cooperative. It could be a challenge for me mbers to comprehend issues properly and vote wisely for the decisions. Purpose of P roducing R ice In Malawi, maize is the staple food and other crops like rice are usually grown for cash or for batter trade where they exchange rice fo r maize. However in s ome areas where rice is grown in abundance it becomes a substitute for maize with some households growing it for food. Most of the farmers along the valleys grow rice because maize does not do well due to the nature of the soils. Soils in these areas are u sually water logged in the rainy season making it hard to grow maize. Bwanje valley is one of those areas where most of the land is water logged during the rainy season making production of maize difficult. In this sample 57.4% of the coop members indicat ed that they grow rice mainly for food but they also keep some to sell while 40 .6% indicated that their primary goal for

PAGE 49

49 producing rice is to sell but they also keep some for home consumption. Similarly for non coop members the majority which is 64.1 % indi cated that they grow rice primarily for consumption but they keep some to sell while 33.3 % indicated that they grow rice primarily to sell while keeping some for home consumption. Only about 1% in both cases indicated that they grow rice just to sell or ju st for consumption. Table 4 3. Purpose for rice production Purpose Coop membership status Coop members Non Coop members Consumption only 1.0% 1.3% Consumption and sell 57.4% 64.1% Sell only 1.3% 1.3% Sell and consumption 40.6% 33.3% It can be not ed from the results that from the coop members a higher percentage indicated that they grow rice with a primary purpose to sell and only keep some for home consumption compared to the non coop members who indicated the same. This could be attributed to the business exposure through various trainings that the coop members are exposed to, which is not the case for their non coop counterparts. However the difference between the two groups is not significant. Farmer ncome Most rural Malawians depe findings concur with statistics that indicate that the majority of rural households depend on agriculture as their source of income. From the results 84.4% of the total sample indicated that they depend on rice production as their main source of income, while 15.6% indicated that they depend mostly on other sources. These results are similar to those of other GoM statistics which indicate that 80% of rural households depend on farming as their major sour ce of income.

PAGE 50

50 Other income generating activities for most rural households include small enterprises, piece works, and remittances. Off farm income has an effect on the use of inputs by the farmers. Availability of off farm income improves the farmers acc ess to inputs which in turn is expected to improve their yields. It can be noted from the results that the majority of the farmers did not have any other source of income other than farming. This figure is higher for farmers who are not coop members than t he farmers who are coop members, 59.0% and 43.6% respectively. The study found that the other major source of income apart from farming is small enterprise, which includes hawkers, beer brewing, and small scale bakeries. Table 4 4. f farm income From the sample 34.7% of the coop members indicate tha t they are involved in some kind of a small business while for non coop members 24.4% indicated that they have a small business. The results indicate that most of the coop members have other sources of income. In this case it would be expected that they wo uld have a higher production than their non member counterparts since their access to inputs would be higher. It can also be noted that the other source of off farm income is piece works. The piece works involves om the sample 18.8% of the coop members Source of Income Coop membership status Coop members Non Coop members Small Business 37.4% 24.4% Piece Works 18.8% 15.4% Employment 3.0% 0 Remittances .0 1.3% None 43.6% 59%

PAGE 51

51 indicate that they earn additional income from piece works while for non coop members it was 15.4% who indicated piece works as their other source of income. Remittances and employment have very minimal significance as sources of income for the farmers in this area of study. Only 3% from the coop members indicated that they are employed and none from the non coop members while only 1.3% from the non coop members indicated getting remittances as additional source of i ncome and none from the coop members indicated that they get remittances. Rice Production and Marketing Rice P roduction Bwanje valley is one of the major rice growing areas in Malawi. Production of rice in the area is done by small scale farmers. Developm ent of an irrigation scheme improved production by controlling the supply of irrigation water to the rice fields. However production is also done in fields outside the irrigation scheme. All Cooperative members have plots of land in the scheme where they g row their rice and some also have additional plots of land outside the scheme where they also grow rice. Rice production for these small scale farmers is still on the lower side compared to the potential of 6000kgs per hectare (Malawi Government, 2002) wh ich translates to 2450kgs or 48 50kg bags per acre. From the sample production ranged from 2 to 80 50Kg bags of rice which is 100 to 4000Kgs. The average production was 22 bags and the median was 18 bags. It can be noted from the average and the median tha t production is still low especially considering that most of these farmers, 76%, had one or greater than one acres of land. It can be noted from the study results that for most of the cooperative member farmers, production was higher than that of the non member farmers. From the sample

PAGE 52

52 46.2% of the non member farmers had a production of less than 10 bags of rice compared to only 21 % from the coop member farmers. Table 4 5 Rice production quantities, reported in 50kg bags Quantity Coop membership status Coop members Non Coop members Less than 10 21.8% 46.2% 11 20 bags 22.8% 26.9% 21 30 bag 23.8% 10.3% 31 40 bags 17.8% 10.3% Above 40 bags 13.9% 6.4% A chi square test of association reveals that there is a strong association between rice produ ction level and farmers membership to a cooperative at 99% level of significance. The table below shows results from a cross tabulation between production and membership. Table 4 6. Results for cross tabulation between coop membership and production Pearson Chi2 = 14.607 Pr = 0.000 This can be attributed to the benefits that the coop member farmers enjoy in the irrigation scheme which incl ude extension services, controlled water supply and inputs such as fertilizer. The major problem facing the plots outside the scheme that was indicated as lack of water control where sometimes there is too much for the crops or the water does not get to t he fields in adequate amounts. From the study it can also be noted that for the coop members a higher percentage produces between 31 to 40 bags and only 13% L ess than and equal to 20 bags More than 20 bags Cooperative member 45 56 Non Cooperative member 57 21

PAGE 53

53 produces more than 40 bags. The productivity however is still lower than the potential even for the farmers in the irrigation scheme. Rice Marketing Kohls and Uhl (1990) defined agricultural marketing as comprising the business activities associated with the flow of goods and services from agricultural production to food and fiber consumption. Marketin g enables agricultural producers to break out of a farmers more income so that they constitute a growing market for domestic industry (Nyirenda, 1988). This was evident i n the study as almost all the sampled farmers sold some of the rice they produced for income. However following the low levels of production most farmers reported that they only sell their production when cash is needed as evidenced by 57.4% of the coop me mber farmers sampled who indicated that they only sell when the need arises and 62.8% non coop member farmers who indicated the same. Availability of markets is one element for improving the marketing of agricultural products. When farmers are making a dec ision on where to sell a particular crop they base their decision not only on the prices they expect to receive in each of the market but also on additional costs related to transacting in these markets (R. Vakis et al, 2003). The study found that most of the farmers sold their rice at the farm gate. From the sample 85.9% of the non coop member farmers indicated that they sold their rice at the farm gate and 73.7% of coop members indicated that they sold their rice at farm gate. It is interesting to note h owever that for the coop members the figure is lower than the non coop members. This indicates that coop members access other markets which

PAGE 54

54 most of the non coop members do not. The table below shows the percentages of farmers which patronize the different markets. Ta ble 4 7. Farmers markets Market Coop membership status Coop members Non Coop members Farm gate 73.7% 85.9% Local 20.5% 29.3% Distant 9.1% 6.4% Cooperative 7.1% 0% From the results it was found that a higher percentage of coop member farmers sold their rice at the local market as well as the distant market, 29.3% and 9.1% respectively compared to 20.5% and 6.4% of the non coop member farmers. Distant market is the least accessed market. The reason indicated for not accessing the dista nt market was mainly the high transportation cost that is associated with selling at the distant markets. Most farmers indicate that they have little information about the distant markets and accessing the market would be risky for them. The study resul ts also show that the cooperative market was only accessed by the coop member farmers. Only 7.1% of the coop member farmers sold their rice to the cooperative. It was found out that the cooperative had some financial problems and hence they were not able t o purchase rice from most of their members for two growing seasons. As a result of this most of the members resorted to identifying other markets to sell their rice. Failure of the cooperative to provide a market to all its members in this case provides a challenge to one of the essential objectives of a marketing cooperative produce.

PAGE 55

55 Access to the different markets will have a bearing on the transaction costs that the farme r faces. For example if a farmer only sells at the farm gate he is likely to avoid incurring costs like transportation, time spent at the market, however competition is likely to be low at the farm gate and prices are hence also likely to be low. The choi ce of farm gate could be attributed to the nature of the business activity the farmers carry out and also the production level. The study found that 59.9% of farmers sell their rice when need arise s. This means it would be costly for them to take a few bag s to the market for sell every time need arises. Selling at the farm gate becomes a less costly option for the farmer. Major T ypes of Rice Buyers in the A rea The study identified the types of buyers that are major players in the rice market in the area. Th e table below shows the typical buyers to whom the farmers usually sell their rice. Table 4 8. Type of major rice buyers in the area From the study results it was found that the majority of the farmers sell their rice to assemblers. Out of the sample 72.3% of the coop members indicated that they sell their rice only to assemblers and 69.2% of the non coop member farmers indicated selling their rice only to assemblers. Some farmers indicated that they sell to assemblers but they also sell directly to consumers. From the sample 25.6% of the non coop members Buyers Coop membership status Coop members Non Coop members End consumers only 3% 3.8% Mostly consumers but also asse mblers 2% 1.3% Assemblers only 72.3% 69.2% Mostly assemblers but also consumers 22.8% 25.6%.

PAGE 56

56 and 22.8% of the coop member farmers indicated that they also sell to consumer but assemblers are their main buyers. The assemblers are usually the middle men that come to buy the rice from the area for resale at markets in the cities of Lilo ngwe and Blantyre as well as neighboring districts. However the farmers also indicated that some of the assemblers are local people from the area who also buy the rice to sell at distant markets. From the sample only 3.8% of the non coop members and 3% of the coop members indicated selling only to end consumers while 1.3% of non coop members and 2% of the coop members indicated selling primarily to consumers but they also sell to assemblers. The farmers who indicated selling their rice to consumers said tha t they sold polished rice. They mill the rice and usually sell it at the market in small portion as demanded by the consumer. These portions could be as little as 250 grams. They also pointed out that sometimes they do sell around their neighborhood throug h door to door selling. It can be noted from the results that middlemen are still the major players in the marketing of rice in the area. These middlemen usually sell the rice that they purchase to the rice milling companies in the city at higher prices t han the prices they offer the farmer. Rice Milling Company is one of the companies which package polished rice. They indicated that they buy their entire paddy from middlemen who bring it to their door step. This shows that the famers would be able to sell their rice at higher prices if they were able to access such markets directly. However it entails a higher investment in

PAGE 57

57 period of time which is agreed upon with the buyer. These two factors also contribute to The study also looked at the type of relationship the farmers had with the buyer. and reduce his search cost because the farmer would know who to contact when the y want to sell their rice and also the negotiation cost because of the built trust between the buyer and the farmer. From the study results it was found that most of the farme rs sell their rice to total strangers. From the sample 72.6% indicated that they sell their rice to people they do not know at all while 17.9% indicated that they somehow know the buyers from past transactions. Only 5.6% indicated that they know their buye rs very well while 3.9% indicated that when selling to consumers they know them but when selling to assemblers they do not know them. This can be attributed to the earlier results which show that the majority of the farmers sell their rice to assemblers w ho might not always be the same people. The low percentage of buyers being known makes a lot of sense because very few farmers from earlier results said that they sell to end consumers. These end consumers are likely to be people they know because they men tioned that they sell to consumers around the same area or in their neighborhood. Results from a cross tabulation between production sold and knowledge of the buyer shows that the farmers who sell to buyers they do not know are the ones selling lower volum es of rice than the farmers who sell to people they somewhat know from past transactions. The difference was marginally significant at 10% level of confidence. This could be as a result of the buyer farmer relationship that the farmers who sell more

PAGE 58

58 are li kely to establish with the buyers. It could be likely that farmers who sell lower volumes do not really establish relationships with the buyers and hence sell to anyone who comes to buy. However there is a need to get more insight into this relationship to objectively describe the reason behind this. The table below shows the results of the cross tabulation Table 4 9. Results for cross tabulation between buyer knowledge and sales volume Pearson Chi2 = 6.2938 Pr = .09 Rice Marketing C onstraints The farmers cited various constraints that they face when marketing their rice. The major problems include: lack of proper and organized markets, low producer prices, exploitation by buyers, poor road infrastructure, high transportation costs and lack of market information among others. Table 4 10 below explains more in terms of percentages. Table 4 10. Rice marketing constraints Less than and equal to 15 bags More than 15 bags Do not know bu yer at all 91 39 Somewhat know Buyer 24 8 Market Coop membership status Coop members Non Coop members Lack of markets 33% 35.1% Low prices 83% 77.9% Poor road infrastructure 18% 10.3% Too few buyers` 33% 50.6% High transportation cost 19% 14.3% Buyers cheating farmers 36% 26% No proper link for buyers and farm ers 18% 10.4% Lack of market information 32% 23.4%

PAGE 59

59 From the study results the major constraints that were cited by the famers were low producer prices (83% for coop members and 77.9% for non coop members), too few buyers (50.6% for non coop members and 33% for coop members) and lack of markets (35.1% for coop members and 33% for non coop members ). Low producer prices could be attributed to fact that most of the farmers sell their rice soon after harvest when the supply is very high. The farmers indicate d that they are desperate for money at this time of the season because they have a lot of b ills to settle which are usually due at harvesting time. This makes them vulnerable to selling at any price the buyer offers so that they can get the money that is n eeded. The middle men who are usually the buyers take advantage of this and offer very low prices. These prices however improve as the supply gets low. For those farmers who are able to keep some of their production to sell later, they benefit from the imp roved prices. The farmers could benefit from the improved prices by staggering the sale of their produce so that they sell some of their produce during the off season period to benefit from the higher prices. The other important problems were: poor road in frastructure, lack of market information, farmer exploitation by buyers and high transportation costs. In some areas the roads were impassable during and soon after the rainy season. This hinders access to the place by the buyers and most of the farmers ca nnot manage to transport their rice to other markets. This then makes the few buyers who have managed to reach the area to offer very low prices leaving the farmer with very few options either to sell at the giveaway price or to keep his rice with very few prospects of selling in the near future. The study also tried to seek ideas from the farmers on what they think could be the solutions to the challenges indicated above. Suggestions to improve the situation

PAGE 60

60 include: improvement of road infrastructure to i mprove transportation and access to the area, government intervention and regulation on producer prices, all farmers in the area getting more organized to form an association, and government to assist the farmers by identifying markets or reopen ADMARC 2 ma rkets in the areas. 2 ADMARC ma rkets were closed down in the area following the restructuring of the marketing board

PAGE 61

61 CHAPTER 5 TRANSACTION COSTS AN D COOPERATIVE MEMBER SHIP Transaction Costs S ources As discussed earlier, transaction costs include information, bargaining, and monitoring costs. Information costs occur before the exchange takes place and include aspects such as searching for attributes that could facilitate the transactions, seeking better prices, and looking for potential buyers. Bargaining or negotiation costs are incurred during the exchange and include the time to negotiate a contract reach an agreement, and make arrangements for payment Transaction co sts are difficult to measure directly. Certain proxies that are believed t o affect the cost of transacting are used instead. The study identified the aspects of a transaction that are a major source of transaction costs for the farmers in the area. The transaction costs sources include: market information (information about p rices, other markets), availability of few buyer s in the market, lim ited other uses for land and poor road infrast ructure. Market I nformation Market Informati on encompasses a number of areas which include information about goods, quantities qualities, markets, prices, buyer s both for the market of transaction as well as others. The purpose of market information is t o assist both producers and traders in balancing supply and demand so as to limit excessive price rises and surpluses (Abbot, 1967). Market information is very crucial in agricultural marketing since it highlights availability of good quality products and prices and it enables both traders and producers to make informed decisions in regards to marketing activities. When estimating transaction costs, market information affects search costs. For the farmers to identify a buyer for their produce, make a choice on where to sell and

PAGE 62

62 at what price, they need to have information in order to make informed decisions. Lack of market information will hinder farmers from selling in more lucrative markets and transacting at lower costs. The study focused on information about availability of other markets and prices in other markets before the farmer makes a sale. Table 5 1 presents information on the source of information for the traders Table 5 1. Sources of Market Information It can be noted from the results that most of the farmers depended on family and frie nds as a main source of information. From the sample 30.8% of the non coop member farmers and 22.8% of the coop member farmers indicated family and friends as their major source of information. Other sources of information were; extension worker, 5% and 5. 1%, cooperative (8.9% and 1.3%), Radio (18.8% and 33.3%) and Buyers (4.9% and 2.6%) for coop member and non coop member farmers respectively. It would also be interesting to note that 29.7% and 22.9% of the coop member farmers and non coop member farmers respectively, indicated that they do not have any kind of market information. This translates into 28.5% of the total sample indicating that they do not have any kind of information. This makes market information as one of the major sources of transaction costs in this area. This lack of market information would increase the farmers cost of searching for a buyer and increases the chances of not Source Coop membership status Coop mem bers Non Coop members Radio 18.8% 33.3% Family and Friends 22.8% 30.8% Extension Worker 5% 5.1% Cooperative 8.9% 1.3% Buyers 14.9% 2.6% No Information 29.7% 26.9%

PAGE 63

63 finding the b uyer in advance thereby also increasing the time taken to sell the product in the market. A chi squar e test of association indicates that there is a significant association between cooperative membership and source of market information. There is sufficient evidence at 1% level of significance that source of information and coop memberships are co depende nt. This means that coop membership may have an influence on the least indicated sources of information and that non coop members had more information compared to the co op members as shown by 29% 0f coop members who indicated not having any kind of information compared to 26% for non coop members The magnitude of the association cannot be explained by this test because chi square only measures any evidence of association (Algresti and Finlay, 2008). A study by Vakis et al in Peru found that farmers selling their produce at farm gate had more information about prices in other markets than those selling at the local and distant markets (R.Vakis et al, 2003). This is simila r to the results found by this study. The earlier findings indicate that the majority of non coop members sell their rice at farm gate than the coop member farmers and they have more information. One other interesting aspect of the findings was that buyer s are a source of information. From the total sample, 9.5% indicated that they get i nformation from the buyers. It would be important to find out the quality of the information the buyer provides to the farmers. Human beings are said to be opportunistic in nature ( Coase, 1937), which suggests an interesting research question do buyers provide correct and adequate information to the farmers.

PAGE 64

64 Availability of Few Buyers in the Market One of the assumptions of perfect competition is many sellers a nd buyers in the market. This ensures that prices are set fairly and determined through supply and demand. Having few buyers may reduce competition and create opportunistic behavior as buyers try to demand low prices. The study results show that few buyers in the m arket was one of the major sources of transaction costs in the study area, evidenced by 33% and 50.6% of coop and non coop member farmers respectively who indicated few buyers in the market as one of the challenges. To the farmers few buyers would increas e their negotiation costs. The farmers would have few alternatives which will make them take longer to negotiate for a better price. Sometimes at the end an agreement might fail to be reached upon and no sale takes place. This becomes a cost to the selle r who has not made a sale but at the same time lost time negotiating for a better price. The study found that negotiating for a better price was indicated as one of the areas where farmers spend a lot of time. From the sample 73.3% of coop members and 75 .3% of non coop members indicated that they have problems negotiating for a better price with the buyer more so when they are selling to assemblers. It should however be noted that this could also be attributed to other factors such as age, experience, qua lity of product in addition to the number of buyers. The study also found that there is no significant association between cooperative membership and number of buyer the farmer encounters in this area. Also, no significant association was found between cha llenges in negotiating for price and cooperative membership.

PAGE 65

65 Poor Road I nfrastructure other than farm gate requires that they transport their product to the said market. Equa enable them to get to the farmers. Poor road infrastructure would deter many buyers from going to buy farmers produce. This would create a non competitive marketing environment with the few buyers that will manage to get to the farm gate. With few buyers, bargaining becomes difficult for the farmers due to limited alternatives for potential buyers. For the farmers willing to sell their produce at local or distant markets poor ro ad infrastructure limits their ability to access these markets. It becomes costly to transport produce in terms of transport cost and time spent to get to the market. Results from the study indicate that poor road infrastructure was one of the challenges t he farmers meet in marketing their produce. The study found that the challenge was experienced only in some parts of the study area. From the sample only 17.3% indicated bad roads as a challenge to their marketing. Farmers from these particular areas indic ated that during and soon after rainy season the roads are impassable thereby limiting access to and from the area. Vehicles travelling to and from the area wait for a number of days after a rain storm to be able to use the roads these delays increase tra nsaction costs. Asset S pecificity Levels : Land Use When an asset is specific to a particular use it leads to hold up problems. Rice production requires very wet conditions and few other crops grow well in a similar environment. The results from the study indicate that most of the farmers in the area grow rice because the type of soil is suitable for rice production only. The farmers

PAGE 66

66 indicated that during the rainy season the land is usually waterlogged. The only other crop they think would do well is sugar cane. They however indicated that sugarcane would not be as lucrative as rice is because of lack of market. From the study 87.2% of the non coop members indicated using the land only for rice production while for coop members 99% indicated growing only ric e. In addition to the nature of the land, the coop members are also bound by the irrigation scheme requirement that they grow only rice during the rainy season. During the dry period they use the land for production of maize. It can be noted from the r esults that 12.8% of the non coop members indicated using their land for other crops. Further inquiry indicated that in some of the areas outside the scheme the water flow sometimes changes course leaving some areas dry. This makes it impossible to produce rice so farmers in these areas grow other crops such as cotton. The farmers were also asked how easy it would be for them to use their plot of land for production of other cash crops. From the sample 74% of the non coop members and 65% of coop members in dicated that it would be very difficult for them to use the plot of land for production of other cash crops. These results are interesting to note because from earlier results 99% of coop members indicated growing only rice in their field from season to se ason but only 65% mentioned that it would be very difficult for them to grow a different crop for cash on the field. The expectation was that a much higher percentage of the coop members would find it very difficult to grow a different crop on their field because they are bound by scheme requirements to produce rice only during the rainy season. These results might also mean that 34% of the farmers would grow a different crop given a chance. However, the farmers indicated that it is difficult

PAGE 67

67 for them to gr ow a different crop because the only other crop which can do well in the area is sugarcane which does not have a ready market and would earn very little compared to rice. The results show the high levels of specificity the land has to the production of ric e. A chi square test of association indicates that there is no significant association between land use and cooperative membership. This could be attributed to the fact that other than the scheme requirement that all farmers in the irrigation scheme shoul d grow only rice the type of soil for the whole area is suitable for crops that grow well in water logged conditions. Rice happens to be the crop that suits the area. It can be noted from the results that there is a high level of asset specificity when it comes to use of land in the area. This creates a high probability of the farmers facing hold up problems because they are stuck with production of rice every season regardless of how the market for rice is behaving. This would create more room for opportu nistic behavior on the part of buyers because the farmers will still grow rice even if the prices are low because they do not have an alternative crop for income. Uncertainty Levels Williamsons (1975) argued that in the imperfect world, where people have l imited information processing capacity and are subject to opportunistic bargaining, high actions in a transaction. Uncertainty is exacerbated by lack of information. Earlier findings from the study indicated that lack of market information is one of the constraints the farmers face when marketing their produce. The table below shows the kind of information about which the farmers were certain at the time they were car rying out the transaction.

PAGE 68

68 Table 5 2. Information about which farmers are certain at time of sale It can be noted from the results that the uncertainty level is high. Most of the farmers are on ly certain on the location of sale. From the sample, 87.6% of the coop members and 88.3 % of non coop members indicated being certain of the place of sale at or before they carry out the transaction. These findings correspond to earlier findings which ind icated that most of the farmers sell their rice at farm gate. They could be more certain about the place of sale because they always sell their rice at the farm gate. What is more interesting to note is the percentage of farmers who indicated being certai n about price they would receive at the time of sale. From the sample 60.8% of the coop members and 55.8% of the non coop members indicated that they are certain about the price they will receive at the time of sell. This was not expected because when aske d about how the prices are determine d the majority of the farmers ( 76.2% and 84.6% of the coop and non coop members, respectively ) indicated that the price is determined through negotiations between the farmers and the buyers. The expectation is that the f armers would be certain about the prices if there were standard prices set for the market. It can however be noted that in the r est of the areas, most of the farmers are not certain of what the outcome would be at the time of sale as shown by low percenta ges of farmers who were certain about the information. Type of information Coop membership status Coop members Non Coop members Number of buyer 6.2% 2.6% Price to be received 60.8% 55.8% How long it will ta ke to sell 23.7% 16.9% Location of sell 87.6% 88.3% Prices in other markets 16.5% 22.1% Other available markets 32% 42.2% Transportation costs 26.8% 22.7%

PAGE 69

69 Frequency Frequency is one of the transaction dimensions that affect the cost of transaction (Williamson, 1979 ). Transaction costs are transaction specific and the more transactions the farmer carries out with a different buyer for each transaction the more costly it will be to him/her. The table below shows the frequency at which the farmers go to sell their rice Table 5 3. Frequency of transactions Frequency Coop membership status Coop members Non Coop members One trip and sell everything 5.9% 14.1% Several trips within short period 5.9% 5.1% Several trips over long period 30.7% 17.9% Always at farm gate as need arises 57.4% 62.8% It can be noted from the results that most of the farmers indi cate that they only sell their rice as need arises. From the sample 62.4% and 57.2% of non coop and coop members respectively indicated selling their rice at farm gate only as need arises while 30.7% and 17.9% of coop and non coop members indicated that th ey make several trips to the market over a long period of time. Some of the farmers indicated that they make several trips to the market over a short period of time as shown by 5.9% of coop member farmers and 5.1% of non coop member farmer. Only 5.9% of c oop members and 14.1% of non coop members indicated that they make only one trip to the market and sell everything. This shows a low level of business activity associated with the small scale farmers in the area. Coupled with the nature of rice (its long shelf life) the farmers are able to sell only when they have cash needs at the household. However due to poverty levels which are high in the rural areas and low levels of production, as shown from

PAGE 70

70 earlier results, it was indicated tha t it is difficult f or the farmers to keep their produce for much longer to benefit from increased prices resulting from dwindling supplies. High levels of frequency of transaction will have a notable impact on the cost of transaction mainly for those farmers who sell their p roduce at the local and distant markets when selling to a different buyer all the time. A chi square test of association shows that there is some level of association between the frequency level and cooperative membership at 10% level of significance with coop member farmers having fewer transactions than non coop member farmers Cooperative M embership There are many reasons why farmers organize themselves into groups. Some of the reasons include improving access to inputs, credit, and markets as well as id entification of new markets, improving crop production and increasing their bargaining power for better prices. Farmers often can also procure their inputs more easily through the cooperative rather than working on an individual basis. According to the Int ernational Cooperative Alliance (ICA ), a cooperative is supposed to be an autonomous association of persons united voluntarily to meet their economic, social and cultural needs and aspirations through a jointly owned and democratically controlled enterpris e. The study however found that the BVRPMC is not based on the voluntary membership per se. The cooperative was formed on the basis of the area having an irrigation scheme which was originally government owned but was passed on to the farmers after the Jap anese International Cooperation Agency (JICA) rehabilitated it. The farmers were allocated a pl ot in the irrigation scheme based on their involvement during the scheme rehabilitation as part of

PAGE 71

71 contribution. Survey respondent s indicated that it was a requirement that all farmers who have a plot in the irrigation scheme should become cooperative members. decision to participate in the cooperative or intentions on choice participation decision in the cooperative. This will be discussed under the regression results. Reasons for Participating in the Coopera tive As just indicated, the irrigation scheme had a requirement for all farmers with plots of land in the irrigation scheme to be members of the cooperative. Other farmers indicated additional reasons for participating in the cooperative. The table below shows reasons why farmers joined the cooperative. Table 5 4. Reasons why farmers joined the cooperative Participation Reason Percentage It was a must since I had a plot in the scheme 58.3% I just followed my friends 10.2% I wanted to get support from f riends 10.2% To access extension messages 8.3% To access markets 6.5% Source of inputs 6.5% From the results it can be noted that the scheme requirement was the major reason why most of the farmers joined the cooperative. 58.3% of the coop member farm ers mentioned that they joined the cooperative because they had a plot of land in the irrigation scheme. The other reasons some farmers indicated were the social benefit from the cooperative. 10.2% of the members said they joined the cooperative just to follow their friends and another 10.2% said they wanted to get support from friends.

PAGE 72

72 It can be said that social benefit is another major reason for the farmer participation. Further inquiry showed that the farmers who are members of the cooperative assist each other in time of need. For example when one is sick fellow their fellow members with necessary agricultural practices such as cultivating the field, weeding and ot her activities. Few farmers indicated reasons such as improving access to markets and inputs as the reason for participating shown by 6.5% on both reasons and 8.3% for access to extension messages. This shows that little consideration is given to issues concerning transaction costs when the farmers are working in a group. Reason for Not Participating in the Cooperative Similar to the participation reason the reason for not participating was also largely influenced by the restriction that was put by the c ooperative on membership. Membership is restricted to those farmers who have a plot of land in the irrigation scheme. The table below shows reasons mentioned by farmers on why they were not cooperative members. Table 5 5. Reasons for not participating Pa rticipation Reason Percentage I did not get plot of land in the scheme therefore did not qualify 26% The cooperative is far 10.4% I was not interested 6.5% The study however discovered that in addition to the restriction that was set by the cooperativ e some farmers did not have information on the cooperative. Some farmers only knew of the availability of the irrigation scheme but did not know of the grouping called the cooperative. These farmers said that they thought the grouping was

PAGE 73

73 only concerned wi th the irrigation scheme and since they do not have a plot of land in the irrigation scheme they gave it little attention. The results also confirm the influence of the membership restriction, as 26% of farmers who indicated that they were not members beca use they do not have a plot of land in the irrigation scheme. The majority of the farmers, 57.1% who were not members mentioned that they did not have much information concerning the cooperative. This could have been a result of the membership restriction that made it seem less important to create more awareness about the cooperative in the area for the other farmers to have sufficient information about the cooperative. Social Network The study tried to find out if social networks would have an influence o knowledge of or relation to cooperative members prior to joining the cooperative. The results showed that most of the members have been members since the esta blishment of the cooperative and hence they did not have knowledge of anyone who belonged to a cooperative prior to their joining. This is shown by 70% of the members indicating that they did not know any member of a cooperative prior to joining. 30% of t he sample indicated knowing some cooperative members prior to joining These are farmers who joined the cooperative later mostly after inheriting or renting, on long term, a plot of land in the irrigation scheme. As a result most of the people they knew we re extended family members such as brothers, sisters, parents and cousins. For farmers who were not members of the cooperative, the majority also said they did not know many cooperative members. This could be attributed to lack of sufficient knowledge abo ut the cooperative since many of these same respondents

PAGE 74

74 indicated that they know farmers who have plots of land in the irrigation scheme. The social network could not have had much influence even if information about the cooperative was made available and the farmers knew some cooperative members, due to the restriction that was placed on membership Membership Restriction After establishing that there was a restriction on membership the study investigated if the members were aware of this restriction and the reasons behind it. From the results it was discovered that most of the members were aware of the restriction that was placed on membership. From the sample 89.9% mentioned that they knew that membership is restricted while 6.1% said that membership is not restricted and 4% indicated not being aware of the restriction. The farmers gave varying reasons to why membership was restricted. This creates a lot of skepticism relating to the information that was communicated to the members as being the reason f or the restriction. The reasons that were pointed out by the members include; 1. Rules laid down by the scheme 2. M ost of the rules in the cooperative relate to the irrigation scheme and those who do not have fields in the scheme wil l not feel as part of the g roup 3. T o create uniformity since the irrigation scheme has rules that are followed 4. Farmers who do not have a plot of land in the irrigation scheme would not be committed to the cooperative. More information needs to be collected on how the decision to pla ce a restriction was made and if the farmers were involved in the decision process. There is also need to find out how valid the reasons that were given are because the restriction and the

PAGE 75

75 forced participation are violating one of the principles of a coope rative, which is voluntary participation. Regression Analysis Results Participation Model Two models were run to establish the significance and direction of the effect of the TC proxies on the decision to participate in the cooperative and on income. As indicated earlier on in the thesis (chapter 3), the participation model used the respondents intended choice in participation rather than the actual membership. This adjustment was made following the discovery made during pre testing tha t participation wa s restricted. respondent indicated that they would join the cooperative and 0 if they indicated that they would not join the cooperative given a chance to make a voluntary decision. Tab le 5 6 shows the frequencies on coop and non coop member responses to the join question. Table 5 6. Frequencies on join responses Frequency Coop membership status Coop members Non Coop members I would join 75 69 I would not join 20 6 6 3 The results for the probit analysis are shown in Table 5 7. It can be seen that not all coefficient signs were as expected. For example the sign for MKTINFO was expected to be negative, such that the more information one has about availability of othe r markets the less likely they would want to join the cooperative. BUYERDEP also did not turn out the way it was expected.

PAGE 76

76 Table 5 7. Participation model results Join Coefficient Std Errors Marginal Effects Std Errors Age .07531 (.08872) .015018 (.0174 1) Agesq .00312 (.00184) .015018 (.09977) Exper .09507** (.05619) .01896 (.01113) Expersq .00297** (.00164) .000592 (.00032) Prcinfo .51216 (.21831) .10215 (.04252) Buyerdep .09994 (.12075) .01993 (.02394) Mktinfo .32796 (.30275) .06247 (.05488 ) Assetred .01706 (.10104) .003402 (.02016) Dist .01046 (.03536) .002086 (.00706) Typbuyer .38396 ** (16869) .076586 (.03296) Educ .01663 (.05940) .003317 (.01183) Member .79202* (.30164) .14947 (.0518) Constant 25751 Psuedo R 2 1677 Log L ikelihood 64.036 Note: Significance levels: 1% ** 5%, *** 10%. Standard errors are in parentheses The results show a negative relationship between buyer dependency and participation in the cooperative although it is not significant. The expectati on was that the more the farmer is dependent on the buyer the more likely they are to participate in the cooperative. The variable member was included in the model aft er it was discovered that a high number of members indicated that they would not want to join the cooperative in a voluntary participation situation. Member is a dummy va riable which took on the value of one for members and zero for non cooperative members. Four variables were found to be significant in explaining the choice decision the farm er would make to participate in the cooperative if they were given a chance to make a voluntary choice. The four variables are Information about price, member, experience and type of buyer at 1%for member and information about prices, 5% for type of buyer and 10% level of significant for experience. Price information has a negative relationship with the decision to join the cooperative. This shows that the more information the farmer has about prices in other markets the less likely it is for them to join the

PAGE 77

77 cooperative. The marginal effects for this variable shows that additional information would reduce the probability of choosing to participate in a cooperative by 4.2%. This could be explained in a way that once the farmer has information about prices i n other markets it could form as a basis for them to negotiate efficiently with the buyers. This will make them feel empowered to make informed decisions about a sale based on the fact that they have the information. Opportunism on the side of the buyer w ould be less since the farmer has information. Membership status of the farmers had a negative significant effect on the farmers decision to participate in the cooperative. The marginal effects shows that the probability of choosing to participate in the c ooperative is 14.94% lower for members than non members. The respondent were asked to give a reason for their choice. Most of the members who indicated that they would not join the cooperative indicated that the cooperative has not met their expectations in terms of the benefits they anticipated and that it is too involving for them in terms of the contributions and activities to do with the cooperative. This could explain the lower probability for members since they have experience cooperative membership and are making a more informed decision The other significant variable was the type of buyer. This variable as indicated earlier on showed a significant influence on the decision the farmer makes to join the cooperative. The type of buyer variable has a p ositive relationship with the decision to join the cooperative. From the way the variable was set (as a score variable) it shows that the farmers who sell to assemblers are more likely to join the cooperative than the farmers who sell typically to consumer s. This could be attributed to the fact that the farmers selling to assemblers face the challenge of trying to negotiate with a big buyer who

PAGE 78

78 sometimes is likely to dominate and dictate prices unlike consumers who buy in small quantities and usually do not have the same bargaining position as the wholesaler. Given this scenario then the farmers selling to the wholesalers would likely want to form a collective action to deal with large scale buyers. T hey would negotiate more effectively as a group than as in dividuals. It is surprising that in this study gender and education were not significant as it is with most participation studies. However this could be linked with earlier results that showed that there was no major difference in participation in the coop erative between men and women. Earlier study results on education (from chapter 4) also showed that the majority of the farmers were less educated ( less than 5 years of education ). Very few farmers in the sample had up to eight years of education (as low as 3%). This could explain why education did not have any significant influence on the decision to join the cooperative. Income Model The basic assertion on income was that since cooperatives members have lower transaction costs, their incomes on average are higher than other farmers who are not cooperative members but have similar characteristics. This was tested by checking whether the variable Coop was significant in the model whose dependent variable was the logarithm of rice incomes the farmers earne d in the previous production season. As would probably be expected income was found to be determined by a number of variables including membership in the cooperative In the model, the primary determinants of income were found to be land, age and gender an d cooperative membership which were significant at 1% for land, cooperative membership and gender and 5% for age. Agesq and off farm income were marginally

PAGE 79

79 significant at 10%. From the results the coefficient on coop shows that membership can increase rice incomes by 76.6%. Considering that the cooperative has not been providing marketing services to most of its members, the increase can be attributed to the higher production that is associated with cooperative farmers. This is so because all cooperative fa rmers have plots of land in the irrigation scheme where water supply to the crop is controlled and there are also intensive extension services being provided to farmers which probably also helps increase their production. This is also in line with earlier results which indicate that cooperative farmers have higher production than non cooperative farmers and hence their incomes are expected to be highe r. Table 5 8 Income Model Results Logincome Coeffici ent Std Errors Age .7758* .2153 Agesq .1434** .0524 Educ .0112 .0333 Land .3271 .0929 Gender .4306* .1478 Coop .7664 .2154 Off farminc .2798*** .1441 Constant 8.3615 .3816 R2 .3190 Note: Significance levels: 1% **5% and ***10% The analysis shows that membership increases incomes for the small scale rice farmers in the area. However it is very hard to associate such differences in incomes with cooperative membership due to the mandatory participation of sch eme farmers into the cooperative. In a way it could be said that the cooperative member farmers, who also have irrigated lands have higher incomes because they own a plot of land in the irrigation scheme and not necessarily because they belong to a coopera tive.

PAGE 80

80 CHAPTER 6 CONCLUSIONS AND POLICY IMPLICATIONS Conclusion The major objective of the study was to assess the role transaction costs and marketing cooperatives play in small scale farmers marketing of rice in the Bwanje valley area. In order to achiev e this the study researched if small scale farmers consider transaction costs when making the decision to participate in a cooperative. Results from the study provide information to policy makers concerning the current promotio n of agricultural and marke ting cooperatives in the country. The study was also aimed at finding out if there is any association between cooperative membership and some dimensions of a transaction that affect the cost of transacting as well as the role cooperative membership has on the incomes the farmers earn from rice. A number of areas were discovered to be the major sources of transaction costs for the farmers in the study area. These sources include limited access to market information ( about prices, markets and buyers) which h indered the farmers capability to make informed sale s decisions, poor road infrastructure which limits the farmers access to markets as well as buyers access to the farmers, and availability of few buyers in the area which increased the cost of negotiation for the fa rmers since they were negotiating in less competitive marketing environments. These factors affected the cost of transaction for the farmers through search costs and negotiation costs Most of the transaction cost proxies that were used were fou nd not to have any significant association with cooperative membership. It was hard to establish a relationship between transaction cost proxies and cooperative membership because of the restriction on membership into the cooperative. It was found that mem bership into the

PAGE 81

81 cooperative is limited to those farmers who have plots of land in the ir rigation scheme. For example, the participation model did not use members hip as the depe ndent variable. The model, instead, used a choice decision the farmers would make if membership was voluntary. The results found that only three of the transaction cost variables had a significant influence on the choice decision to participate in the coop erative. The significant variables were type of buyer information on prices and membership status of the farmer. The challenge remained the membership restriction which is considered to have influenced each farmer s decision on the choice made on whether they would voluntarily participate in the cooperative or not The study also found that cooperative membership has a significant positive effect on the incomes of the farmers in the area However it was discovered that the cooperative was not buying rice from its member farmers. This could have had some effect on the outcome of the results since the cooperative is currently not providing the required services to its members. The statistically significant effect could be a result of other benefits that incr ease the level of production such as access to inputs on loan, access to extension services and participation in the irrigation scheme. participation in the cooperative. The fac tors the farmers indicated include getting support from friends in the cooperative, accessing extension messages and as a source of inputs. These were indicated as additional reasons why farmers joined the cooperative in addition to membership being a requ irement for those farmers with plots of land in the irrigation scheme.

PAGE 82

82 Other interesting findings included learning that buyers are a source of market information for many of the farmers. It would be beneficial to identify the type and adequacy of the info rmation the buyers give to the farmers considering that human being s are said to be opportunistic. Finally it was difficult to determine the actual role the cooperative is playing in the marketing of rice for farmers in the area because of the current ope rational status of the cooperative and the restriction on membership. The cooperative was not buying rice from most of its members due to financial problems. Hence few of the interviewed farmers benefitted from the marketing services of the cooperative The farmers only benefitted through the controlled water supply in the irrigation scheme which increased their production and some inputs which were given to the farmers on loan This most likely explains why farmers in the cooperative reported higher pr oduction and hence higher incomes than farmers who were not in the cooperative. Policy Implications Based upon the analysis of the empirical findings of this research, the following policy initiatives have been identified for potential interventions in or der to improve some of the negative factors affecting farmers in the study area : Market Information Systems: The study results show that availability of market information plays a significant role in small scale farmers marketing as shown by its significan ce in the choice decision. However the study also found that information that goes to the farmers is limited and most of the farmers depend on friends and relatives for market information and sometimes on buyers. In order to improve in this area t here is a need to strengthen the market information systems to ensure the required information about available markets and quantities demanded,

PAGE 83

83 prices in other markets, and ava ilable means of transportation, is delivered to the farmers in a timely manner This could be achieved through intensifying the information delivery through the radio during the farmer radio programs and also making use of extension workers by providing them with current and updated information regularly. In addition the study resul ts show that the extension workers are the least common source of marketing information for the farmers i.e. 5 % of respondents indicated extension worker as their so urce of marketing information (Study R esults). This could be as a result of lack of knowle dge and skills in the marketing field or insufficient numbers of extension workers in the area. In order to improve farmers access to information through the extension workers, there is a need to consider areas of capacity building of the extension worker s in the field of marketing and also improving the farmer extension worker ratio to ensure that most of the farmers are being reached with the required extension messages in the ion levels as well as their marketing skills both of which should help them increase their on farm income. Financial Credit for Farmers and Cooperatives : Financial limitation is one of the major problems rice farmers face. For marketing institutions lack of adequate capital to invest is also one of the major challenges they face. The study found that the cooperative is failing to provide marketing services to its members due to financial problems. There is need to improve access to credit facilities for th e cooperative to operate to its full capacity and in turn provide the needed services to its members. However there is need to improve the capacity of those in management positions in the

PAGE 84

84 cooperative for them to be able to improve their management skills a nd operate efficiently. This could in turn assist them to be able to identify the requirements of the cooperative and lobby for the necessary support. Contract Farming : One way to improve the marketing of rice for the small scale farmer in the area would b e to initiate contract farming between the cooperative and rice milling companies that would be interested in buying paddy from these farmers This will ensure a ready market for the farmers which would then reduce the farmers search and negotiation costs since they will be negotiating as a cooperative with the rice milling companies. The availability of th is kind of scheme which could ensure improved production and the organization of the farmers would be considered as positive steps for the development of partnerships between the farmers and the buyers through contract farming. Cooperative Awareness: There is need to increase awareness in the general area of collective action in marketing by small scale farmers through cooperatives, including the potential benefits and costs associated with it. The study results found that most of the farmers in the area do not have adequate information concerning cooperatives. It was found that there was a very low understanding of what a cooperative entails and what the f armers would benefit from as well as the cost to the farmer if they joined. The results also show that cooperative membership is strongly correlated with information concernin g the cooperatives and the associated benefits.

PAGE 85

85 Further Research: The study established a number of research objectives which proved to be difficult to test empirically or that conflicted with the actual conditions in the target population. These included restrictions on cooperative membership (where some farmers were required to join, while others were banned from joining), and the operational status of the cooperative in the area of marketing. These are believed to have influenced the results of the stud y significantly. Re designing the study at this later stage was difficult due to time and resource constraints. Further research would therefore be recommended in the area which would include some of the raised issues in the design of the study.

PAGE 86

86 APPENDIX A QUESTIONNAIRE FOR TH E STUDY I NVESTIGATION OF THE R ELATIONSHIP B ETWEEN C OOPERATIVES AND T RANSACTION C OSTS FOR SMALL SCALE F ARMERS IN M ALAWI A UGUST O CTOBER 2009 DATE OF I NTERVIEW : _______________ I NTERVIEWER ( S )/E NUMERATOR ( S ): ________________________ _____________ F ARMER S N AME : _________________________________________________ F ARMER S L OCATION : _____________________________________________ Q UESTIONNAIRE C ODE N UMBER : _____________________ 1. When y ou grow rice, what are your plans for your production? [Check only one!] ___ I plan to sell all of my production. ___ I plan to use all of my production for home consumption. ___ I plan to use my production primarily for home consumption, but will sell some if there is a surplus. ___ I plan to produce enough with each crop to sell some and keep some for home consumption. 2. In your opinion what do you think is the most important problem rice farmers face in marketing their production? 3. What do you think is the solution to the problem that you have mention ed above?

PAGE 87

87 4. In addition to the problem you just mentioned, are there other problems that rice farmers in your area have when they try to market their production? [Check all that apply] ___ Too few buyers ___ Too little information about prices ___ Prices are too low ___ Urban markets are far away from the farms ___ Commercial traders/buyers treat farmers unfairly/try to cheat them ___ Roads are bad ___ Transportation of rice to markets is expensive ___ Other:_______________________ ________________________ _______ ___ Other2: _____________________________________________________ 5. Approximately, how much land do you use/farm for rice production? ___ Less than 0.5 hectare; less than one half of an hectare ___ 0.5 .99 hectares ___ 1 1.99 hectares ___ Greate r than 2 hectares 6. How much rice did you produce last growing season? ( please indicate the number in terms of 50 kg bags) _____________________________________ 7. How much of your total production did you sell? ( please indicate the number in terms of 50 Kg bags) ___________________________________ 8. On fields where you grow rice, do you always grow rice from season to season? Yes No

PAGE 88

88 9. If you stopped growing rice in your rice fields, how easily could you use these same fields to grow a different food crop for ho me consumption, with 1 being very easily and 5 being very difficult? Please check one. 1 Very Easily 2 Somewhat Easily 3 Neutral 4 Somewhat d ifficult 5 Very difficult 10. If you stopped growing rice in your rice fields, how easily could you use these same fields to grow a different cash crop for sale in markets, with 1 being very easily and 5 being very difficult? Please check one. 1 Very Easily 2 Somewhat Easily 3 Neutral 4 Somewhat difficult 5 Very difficult 11. Please tell us your reasons for growing r ice on your fields instead of other crops. 12. Please tell us your sources of income from the farm ( please check all that apply) ___ Rice production ___ Other cereal grains ___ Fruits and vegetables ___ Livestock production, specify: _____________________ (e.g., poultry, pigs, etc.) ___ Other ? specify__________________________________________ _____ __________

PAGE 89

89 13. From these sources of income that you have just told us, please tell us which one is the most important source? _______________________ 14. Do you hav e other sources of income in addition to income from farming? Employment Business (Small enterprise) Remittances Pension pay None Other, Specify_______________ 15. For how many years have you grown rice on your farm?____________________ 16. When you decide to se ll your rice where do you go to sell (Please check all that apply) ? Farm gate Local market Distant market Cooperative Other, specify___________________________________________________________ 17. How far do you have to travel to get to the clos est main market where you are able to sell rice? ____________Kms 18. What do you think are the advantages and disadvantages of selling your rice in the closest local market?

PAGE 90

90 19. What do you think are the advantages and disadvantages of selling your rice in th e market you use most frequently? 20. What are/would be the advantages and disadvantages of selling your rice in a district market? 21. When you sell your rice, do you typically ___ make one trip to the market and sell all the rice that you plan to sell from th at growing harvest ___ make several trips to market within a one to two week period, selling as much of your rice as you can with each trip ___ make several trips to market, but spreading them out over a more extended period of time, perhaps as much as on e or two months ___ Other? Specify ____________________________________________ 22. On an average day when you are selling your rice how many wholesalers/assembler/buyers do you see in the market? One 2 to 5 6 to 10 More than 10 23. Typically to whom do you sel l your rice? Only wholesalers/assemblers Only to end consumers Mostly to wholesaler buyers and assemblers but also to some consumers Mostly to consumers but also to wholesale buyers and assemblers Other? Specify____________________________________ _________ ____

PAGE 91

91 24. When selling your rice how often do you refuse offers from buyers?__________________________ 25. Please tell us the reason why you refuse the offers for the number of times you have indicated in the above question. 26. When you sell your rice, how much ti me does it usually take? ___less than 30 minutes after I arrive at market ___30 minutes to one hour ___two to five hours ___ Other? Specify__________________________________ _____________ 27. How well do you know the buyers of your rice? ___very well (they are my neighbors, family, etc.) ___casually (I often sell to the same buyers, so I know them from past sales) ___not at all but when I sell to assemblers/wholesalers, I do not know them ___ ................................ ................................ ........... Other? Specify 28. When you sell your rice, how is the price that you receive determined? ___the buyer sets price, no negotiation ___you, as the seller, set the price, no negotiation ___neg otiation between you and the buyer ___farmer coop/association announces price ___there is a standard price in the market that everyone uses ___other? 29. Can you negotiate a better price if the quality of your rice is better than other rice being sold? Yes No 30. Must you accept a lower price if the quality of your rice is worse than other rice being sold? Yes No

PAGE 92

92 31. When selling to assemblers/wholesalers, do you trust your buyer(s) to give you a fair market price for your products? __I completely trust the buyer to be fair ___The buyer is usually fair, but I have to bargain with him/her to get a fair price ___I worry that the buyer may not be giving me the best price ___I do not trust the buyer at all to be fair ___ Other? Specify 32. On a scale of 1 to 5 with 1 being complete trust in the buyer and 5 being no trust at all in the buyer, how much would you say you trust the wholesalers/trades that buy your rice? 1 Very much 2 Somewhat 3 Neutral 4 Not Very much 5 Not at All 33. What was the highest price that you received for your rice this last growing season? Kwacha / Kg____________ 34. What was the lowest price that you received when selling rice? Kwacha / Kg____________ 35. What was the most common price that you received Kwacha / Kg ____________ 36. Which of the following potential problems do you encounter most frequently when marketing your produce Finding a buyer Negotiating for price Enforcing the contract / making sure that buyer sticks to agreed terms Finding information about pri ces Finding information about buyers/ other markets 37. When you sell your rice, what information about the sale do you know before taking your rice to market ? ( Please Check all that apply) ___ The number of buyers in the market ___ The name of the buyer who will buy your product ___ The price you will receive

PAGE 93

93 ___ The amount of time it will take to negotiate the sale ___ The location of the sale/market transaction ___ Prices in other markets ___ Availability o f other markets ___ Any price premiums or price reduction s related to the quality of the product ___ Transportation costs to get the product to market 38. We would like you to tell us how certain or uncertain you were about the information you just indicated. Using this ruler/chart, with 1 indicating m ost certain and 7 indicating most uncertain, your opinion about the certainty of each bit of information that you have before you take your rice to market. most certain/ ---------1 ---------2 ---------3 ---------4 ---------5 ---------6 ---------7 -------/most uncertain RANK RESPONSES 1 2 3 4 5 6 7 39. What is the most important source of information about market s that you use when making plans to sell your rice? Please check only ONE Radio Extension worker Family or friends Cooperative Other specify________________________________

PAGE 94

94 40. How often do you use the market information that you get from this source in your sales plan (for negotiating a price or market selection) when you are selling your rice? ___Not at all ___ Only sometimes ___Most of the time ___All the time 41. Are you a member of a cooperative? Yes No If yes then go to question 42; If no then skip to Question 49. 42. For how long have you been a member? ______________ 43. What was/were the major reason(s ) that influenced your decision to join the cooperative (check all that apply) To get access to markets To find a source of inputs To get support from friends Just followed my friends Other; specify:_________________ 44. Before you joined the cooperative, di d you know relatives or close friends who were members of the cooperative? Yes No 45. If yes, what is your exact relationship with these cooperative members? ______________________ 46. How many cooperative members did you know prior to joining? __ _____________________________

PAGE 95

95 47. Is membership to the cooperative restricted to the farmers who have fields in the scheme?_________________ 48. If yes, why was membership restricted to only those who have fields in the scheme? 49. If you answered no to question 41 why did you not join the cooperative? I did not have much information Its costly I was refused Other? _________________ 50. Do you know any relatives or friends who are members of a cooperative?________ ______ 51. If yes, how many cooperative members do you know?________________________ 52. What is your exact relationship with these cooperative members? ______________________ 53. Given a situation that membership rules were changed and that everyone in this area i s eligible and free to choose whether to become a member of the cooperative or not regardless of the location of their plot what would be your decision given an opportunity to make a choice between joining the cooperative or not, I would not join the cooperative I would join the cooperative 54. What are the reasons that would influence your decision indicated in the previous question? _______________

PAGE 96

96 55. What year were you born? _______ 56. How many years of formal education did you get? None Less than 5 years Up to 8 years 10 years `More than 8 Years 57. Gender: Male / Female Thank you for your time and cooperation in this survey

PAGE 97

97 A PPENDIX B MAP OF MALAWI SHOWIN G THE STUDY AREA Source: Sand in The Engine: Travails Of An Irrigated Rice Scheme In Bwanje Valley, Malawi

PAGE 98

98 LIST OF REFERENCES Southworth and B.F Johnson (Ed), Development and Marketing. Cornell University Press. Alchian, A. A., and H. Demsetz. 1972, December American Economic Review 62(5), 777S795. Algersti, A., and B. Finlay. 2008. Statistical Methods for the Social Sciences 4 th ed. New Jersey: Prentice Hall Inc. B arham, B.L., and M Childress. 1992. Membership Desertion as an Adjustment Process on Honduran Agrarian Reform Economic Development and Cultural Change 40 3, ABI/INFORM Global, pg. 587 Blundell, R., and Dias, M.C 2002 "Alternative Approaches to Evalu ation in Empirical Microeconomics." Portuguese Journal of Agricultural economics 1 : 91 115. Chirwa, E W., P. M. Mule and J. Kadzandira 2006. Agricultural Marketing Liberalization and the Plight of the Poor in Malawi : Working Paper No. 2005/08 Coase, Ron Economica 4: 386 405. Cook, M. L 1995. The Future of U.S. Agricultural Cooperatives: A Neo Institutional American Journal of Agricultural Economics Vol. 77, No. 5, 1153 59. Cook, M.L., and C. Iliopoulos defined Property Rights in Collective Action: the case of US Agricultural Cooperatives. Institutions, Contracts and Organizations: Perspectives from New Institutional Economics Delgado, C .1999 Sources of growth in smallholder agriculture in sub Saharan Africa: The role of vertical integration of smallholders with processors and marketers of high value added items Agrekon 38(Special Issue) :165 189 Drugger, W.M. 1983. The Transaction Cost Analysis of Oliver E. Williamson: A New Synthesis. Journal of Economic Issues Vol. XVII, No. 1: 95 114. Gabre Madhin, Eleni Z. 2001. Market Institutions Transaction Costs and Social Capital in the Ethiopian Grain Market IFPRI Research Report 124, International Food Policy Research Institute Washingt on, D.C. H eckman, J. J., and Vytlacil, E. 1998. "Instrumental Varia bles Methods for the Correlated Random effects Model." Journal of Human Resources 33 : 974 976

PAGE 99

99 Hobbs, Jill E. 1997. the Importance of Transaction Costs in Cattle Marketing Ameri can Journal of Agricultural Economics 79: 1083 1095. Holloway G., C. Nicholson C Delgado, S Staal and S. Ehui. 2000 Agro industrialization through institutional innovation: Transaction costs, cooperatives and milk market development in the east Afric an highlands Journal of Agricultural Economics 23(3):279 288. Iliopoulos C ., and M L Cook 1999 The internal organization of the cooperative firm: An extension of a new institutional digest Journal of Cooperatives 14:77 85. Key N E Sadoulet and A. De Janvry 2000 Transaction costs and agricultural household response American Journal of Agricultural Economics 82(2):245 259. Kohls, R.L and Uhl, N.U .1980. Marketing of Agricultural Products 5th edition, Macmillan, New York. Mahoney J.T., R.C M Organizational Form: Integrating Performance Ambiguity and Asset Specificity 0109 Makhura M T. 2001 Overcoming transaction cost barriers to market participation of smallholde r farmers in the Northern Province of South Africa PhD thesis, U niversity of Pretoria Malawi Government (2002). Agricultural sector pri ority constraints, policies and strategies framework for Malawi. Ministry of Agriculture and Irrigation: Lilongwe Mata ng ul P M M C Lyne and G F Ortmann 2001 Transaction costs and crop marketing in the communal areas of Impendle and Swayimana, KwaZulu Natal. Development Southern Africa 18(3):347 363. September 2001 National Economic Council. 2002 Malawi Govern ment Economic Report Government printer: Zomba Nyirenda A.S.O .1988 Agricultural marketing of grain crops in Malawi: Policy, Practice and Problems. A paper presented at a workshop on Agricultural marketing in The SADC region: Problem, Prospect and Pr actice. Lilongwe, Malawi Omamo S .W. 1998 Transport costs and smallholder cropping choices: An application to Siaya District, Kenya. American Journal of Agricultural Economics 80(1):116 123. Administr ative Science Quarterly 25: 125 141,1980. Posner R. 1998. Economic analysis of law 5th ed New York: Aspen Publishers Inc.,

PAGE 100

100 Rhodes V J 1995 Cooperative enterpr ise and organization theory: An appraisal. Journal of Cooperatives 10:87 91. Royer J S 1999. Cooperative organizational strategies: A neo institutional digest. Journal of Cooperatives 14:44 67. Scott, W.R. 1998. Organizations: Rational, natural, and open systems Upper Saddle River, NJ: Prentice Hall, Inc. Sexton, R. J., and I Julie. 19 93."The Competitive Role of Cooperatives in Market Oriented Economies: A Policy Analysis." Agricultural Cooperatives in Transition ed. Csaba Csaki and Yoav Kislev, Westview Press, pp. 55 83. Staal S J C Delgado and C. Nicholson 1997. Smallholder dai rying under transaction costs in East Africa. World Development 25:779 794. Staatz, J. M. 1989. "Recent Developments in the Theory of Agricultural Cooperation." Journal of Agricultural Cooperation 2 :74 95. Farmer Cooperative Theory: Recent Developments. Was hington, D.C.: U S D epartment of Agriculture, ACS Res. Rep. 84, June Tschirley, D., M M Kavoi, and M.T. Weber. Improving Kenya's Dom estic Horticultural Production a nd Marketing system: C urrent Competitiveness, Forces of Change, and Challenges f or The F uture Volume Horticultural Marketing Working Paper No. 08B/2004 Department of Agriculture and Resource Economics Working Paper 9 62 University of California Berkeley. Vitaliano, P. Complex Institutions American Journal of Agricultural Economics, 65, 1078 1083. Wang, N. 2003. Measuring Transaction Costs: a n Incomplete Survey Working paper number 2 Ronald Coase Institute. Williamson, O.E.1975. Markets and Hierarchies: Analysis and Antitrust Implications Newyork : The Free Press Williamson, O.E. 1979 Transaction Cost Economics: The Governance Of Contr actual Relations. Journal of Law a nd Economics 22 (2) October 233 261 Williamson, O.E. Journal of Economic Literature 19: 1537 1568. Williamson, Oliver E. 1985. The Economic Institutions of Capitalism New York: The Free Press

PAGE 101

101 Williamson, O.E. 1993. The evolving science of organization Journal of Institutional and Theoretical Economics 149:36 63. Williamson O E. (1996). The mechanisms of governance New York: Oxford Universit y Press Wo oldridge, M. J. (2003). "Further Results on Instrumental Variables estimation of Average treatment Effects in the correlated Random coefficient model." Economics of Letters 79 :185 191. Wooldridge, M. J. (2006). Introductory Econometrics 4 th ed. Michigan State University

PAGE 102

102 BIOGRAPHICAL SKETCH Lucy Chamdimba Nyirenda was born on November 28, 1978 in Mangochi Malawi. She grew up mostly in the city of Lilongwe Malawi where she also attained her secondary education at Bwaila Secondary School up to 1995. She earned her BSc in Agriculture (Agricultural Economics) from the University of Malawi, Bunda College of Agriculture in 2000. After graduating Lucy worked in various fields which include Microfinance before join ing the Civil Service in the Ministry o f Agriculture and Food Security where she has worked as an Economist as well as an Agribusiness D evelopment O fficer. where she will continue working in the civil service on various assign ments to assist in the development of the warm heart of Africa, Malawi. Lucy is married to Lincoln and they have two children: Chawezi age 12 and Nancy age 7.