• TABLE OF CONTENTS
HIDE
 Front Cover
 Front Matter
 Title Page
 Table of Contents
 List of Tables
 List of Illustrations
 Foreword
 Acknowledgement
 Summary
 Introduction
 The study area
 Land and capital inputs
 Labor use and allocation
 Relationships between labor, capital,...
 Farm output and resource productivities...
 Policy implications and conclu...
 Appendix 1. Some methodological...
 Appendix 2. Supplementary...
 Bibliography
 Back Cover






Title: Food production in a land-surplus, labor-scarce economy : the Zairian Basin
CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00085382/00001
 Material Information
Title: Food production in a land-surplus, labor-scarce economy : the Zairian Basin
Alternate Title: IFPRI research report ; 74
Physical Description: Book
Language: English
Creator: Tshibaka, Tshikala B.,
Publisher: International Food Policy Research Institute
Place of Publication: Washington, D. C.
Publication Date: June, 1989
Copyright Date: 1989
 Record Information
Bibliographic ID: UF00085382
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: 19846028 - OCLC

Table of Contents
    Front Cover
        Front Cover 1
    Front Matter
        Page 1
    Title Page
        Page 2
        Page 3
    Table of Contents
        Page 4
    List of Tables
        Page 5
    List of Illustrations
        Page 6
    Foreword
        Page 7
    Acknowledgement
        Page 8
    Summary
        Page 9
        Page 10
    Introduction
        Page 11
        Page 12
        Page 13
        Page 14
    The study area
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
    Land and capital inputs
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
    Labor use and allocation
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
    Relationships between labor, capital, and cultivated area
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
    Farm output and resource productivities in the study area
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
    Policy implications and conclusions
        Page 57
        Page 58
        Page 59
    Appendix 1. Some methodological notes
        Page 60
        Page 61
        Page 62
    Appendix 2. Supplementary tables
        Page 63
        Page 64
        Page 65
        Page 66
    Bibliography
        Page 67
        Page 68
        Page 69
        Page 70
    Back Cover
        Page 71
        Page 72
Full Text
RESEARCH REPORT
j74.

I'F


FOOD PRODUCTION
IN A LAND-SURPLUS,
LABOR-SCARCE ECONOMY:
THE ZAIRIAN BASIN

Tshikala B. Tshibaka


Jun 1989
INTERNATI 'SllONL
FOSOD
POLICY4
RESEARCH
iNSTITUTE















The International Food Policy Research
Institute was established in 1975 to identify
and analyze alternative national and inter-
national strategies and policies for meeting
food needs in the world, with particular em-
phasis on low-income countries and on the
poorer groups in those countries. While the
research effort is geared to the precise ob-
jective of contributing to the reduction of
hunger and malnutrition, the factors involved
are many and wide-ranging, requiring analy-
sis of underlying processes and extending
beyond a narrowly defined food sector. The
Institute's research program reflects world-
wide interaction with policymakers, adminis-
trators, and others concerned with increasing
food production and with improving the
equity of its distribution. Research results
are published and distributed to officials and
others concerned with national and inter-
national food and agricultural policy.
The Institute receives support as a consti-
tuent of the Consultative Group on Interna-
tional Agricultural Research from a number
of donors including Australia, Belgium,
Canada, the People's Republic of China, the
Ford Foundation, France, the Federal Re-
public of Germany, India, Italy, Japan, the
Netherlands, Norway, the Philippines, the
Rockefeller Foundation, Switzerland, the
United Kingdom, the United States, and the
World Bank. In addition, a number of other
governments and institutions contribute
funding to special research projects.


Board of Trustees

Dick de Zeeuw
Chairman, Netherlands
Eliseu Roberto de Andrade Alves
Vice Chairman, Brazil
Yahia Bakour
Syria
Anna Ferro-Luzzi
Italy
Yujiro Hayami
Japan
Gerald Karl Helleiner
Canada
Dharma Kumar
India
Anne de Lattre
France
James R. McWilliam
Australia
Harris Mutio Mule
Kenya
Sukadji Ranuwihardjo
Indonesia
Theodore W. Schultz
U.S.A.
Leopoldo Solis
Mexico
M. Syeduzzaman
Bangladesh
Charles Valy Tuho
C6te d'Ivoire
John W. Mellor, Director
Ex Officio, U.S.A.

















FOOD PRODUCTION
IN A LAND-SURPLUS,
LABOR-SCARCE ECONOMY:
THE ZAIRIAN BASIN

Tshikala B. Tshibaka















Research Report 74
International Food Policy Research Institute
June 1989




















Copyright 1989 International Food Policy
Research Institute.
All rights reserved. Sections of this report may
be reproduced without the express permission
of but with acknowledgment to the International
Food Policy Research Institute.


Library of Congress Cataloging-
in-Publication Data

Tshibaka, Tshikala B.
Food production in a land-surplus, labor-
scarce economy : The Zairian Basin / Tshikala
B. Tshibaka.
p. cm.-(Research report/International Food
Policy Research Institute ; 74)
"June 1989."
Bibliography: pp. 67-70.
ISBN 0-89629-076-X
1. Agriculture-Economic aspects-Zaire.
2. Agriculture-Economic aspects-Congo River
Watershed. 3. Agricultural laborers-Zaire-
Supply and demand. 4. Agricultural laborers-
Congo River Watershed-Supply and demand.
I. International Food Policy Research Institute.
II. Title. III. Series: Research report (Interna-
tional Food Policy Research Institute) ; 74.
HD2135.5T754 1989 89-11245
338.1'9'6751-dc20 CIP













CONTENTS


Foreword
1. Summary 9
2. Introduction 11
3. The Study Area 15
4. Land and Capital Inputs 21
5. Labor Use and Allocation 29
6. Relationships Between Labor,
Capital, and Cultivated Area 40
7. Farm Output and Resource Pro-
ductivities in the Study Area 51
8. Policy Implications and Conclu-
sions 57
Appendix 1: Some Methodological
Notes 60
Appendix 2: Supplementary Tables 63
Bibliography 67













TABLES


1. Structure of the sample popula-
tion in the study area
2. Practicability of roads and nav-
igability of rivers in the study
area, 1982/83 survey
3. Community recognition of prac-
tices of right of fallow and lease
of land to nonmembers in study
area
4. Length of fallow period, by house-
holds surveyed, 1982/83
5. Cultivated land/man ratio and
allocation of cultivated area
among crops
6. Quantity of seeds used per hec-
tare by households that produce
rice and maize, 1982/83
7. Allocation of capital tools among
crops, 1982/83
8. Allocation of capital input (seeds
and tools) among crops, 1982/83
9. Official calendar of farm opera-
tions in the study area
10. Farmers' agricultural calendar in
the study area, crop year 1982/
83
11. Intensity of labor use, by month,
1982/83
12. Household labor allocations to
rice, maize, cassava, and plan-
tain
13. Allocation of household produc-
tive labor time between farming
and nonfarm economic activities,
1982/83


14. Determinants of labor allocated
17 to farming, 1982/83
15. Intensity of labor use, by crop,
1982/83
19 16. Cultivated area and intensity of
labor inputs, 1982/83
17. Determinants of the intensity of
labor input
22
18. Intensity of capital input use by
23 crop
19. Cultivated area and intensity of
capital input
24 20. Regression results for determin-
ants of the intensity of capital
input
26 21. Cultivated area and change in
crop mix
27 22. Regression results for cultivated
area, farmgate prices, and share
of cash-crop area
28
23. Regression results for the rela-
tionship between cultivated
area, labor, and input of capital
(tools)
30 24. Regression results for food crop
production functions
S25. Resource use and productivities
in the study area
26. Allocation and productivity of
34 labor, 1982/83
27. Collinson's conversion factors
for man-equivalents
28. Number of man-equivalents per
35 household in the study area













ILLUSTRATIONS


29. Agricultural calendar for study
villages, 1982/83
30. Monthly intensity of labor allo-
cated to farm and nonfarm activ-
ities per member household,
1982/83
31. Preplanting farmgate prices in
the study area, 1982/83
32. Regression results for food crop
production functions using capi-
tal as a deflator for rice and labor
for other crops


1. Map of Zaire
2. Intensity of male and female labor
use, farming and nonfarming ac-
tivities, by month, 1982/83













FOREWORD



Once a net food exporter, Sub-Saharan Africa is now the only major area of the
world where the per capital food supply has been declining for the last two decades.
Despite massive food imports and aid, the deterioration of the food sector, coupled
with rapid population growth, has caused many Africans to face food shortages and
other hardships.
The agricultural policies followed in many parts of Sub-Saharan Africa have played
an important role in moving the region from a net food exporter to a net food importer.
This report on food production in Zaire is the first in a series of studies being undertaken
by the Food Production Policy Program at IFPRI on factors impeding food production
in the smallholder sector of the region.
In this research report, Tshikala B. Tshibaka describes the state of the small-farm
sector in the Zairian Basin and identifies different constraints affecting productivity
and hence the volume of food output this sector produces. This study not only adds
to our knowledge of the Zairian economy, but it also constitutes a useful addition to
the information bank that policymakers need in order to formulate consistent agricultural
policies.
The Zairian Basin represents an important agroecological region in Africa, that, as
is clear from the report, has immense problems that should be studied if the many
people in the region are to participate actively in the development process. We partic-
ularly hope that this research, appropriately modified, will serve as a prototype for
further analyses of the conditions for improvement.
This study, conducted in collaboration with the Institut Facultaire des Sciences
Agronomiques (IFA), Yangambi, Zaire, was partially funded by the International Devel-
opment Research Centre of Canada (IDRC). IFPRI and IFA are grateful to IDRC for its
encouragement and consistent support of this and other research work on Sub-Saharan
Africa.

John W. Mellor
Washington, D.C.
June 1989













ACKNOWLEDGMENTS


This study, initially entitled "Economics of Rice Production in the Zairian Basin,"
is the product of a joint effort between the International Food Policy Research Institute
(IFPRI), Washington, D.C., and the Institut Facultaire des Sciences Agronomiques (IFA),
Yangambi, Zaire. It was generously funded by the International Development Research
Centre (IDRC), Canada, for which IFPRI and IFA wish to express their deep appreciation.
The author especially thanks John W. Mellor for his helpful advice and comments
during this study. While the limitations of space make it impossible to acknowledge
by name all those who helped during the writing of this report, the author feels
especially indebted to Eric Tollens of the University of Leuven, Belgium, and Dayanatha
Jha of IFPRI for their useful comments both on earlier and final drafts of this report.
The contributions of Raisuddin Ahmed, Joachim von Braun, Sudhir Wanmali, Christopher
Delgado, and Behjat Hojjati are also recognized.
Finally, the author strongly appreciates the help of his former graduate assistants,
Muimane-Muende Kalala and Muhiya Aunge, at the Institut Facultaire des Sciences
Agronomiques, who helped during the earlier stages of this study.









1


SUMMARY

Most agricultural projects initiated by the government of Zaire in the small-farm
sector during the last two decades to improve a worsening food supply situation have
failed mainly as a result of limited knowledge about constraints affecting the develop-
ment of the sector. This study was undertaken to fill this gap. The study area is located
in the Zairian Basin, which is part of a tropical rain forest zone. The zone includes
large parts of Equateur, Haut-Zaire, Sud-Kivu, Maniema, Kasai Oriental, Kasai Occidental,
and Bandundu regions, and a small part of the Bas-Zaire region. The Basin also extends
beyond the Zairian boundaries to include a part of the Congo.
The Basin is ecologically more suited to perennial (tree) crops than to annual (food)
crops. Most robusta coffee, cocoa, natural rubber, and palm oil trees are grown on
plantations. The area is the major zone in the country for growing tree crops, as well
as rice and plantain. Small-farm households primarily grow staple food crops, mainly
rice, maize, cassava, and plantain.
Land is readily available in this sparsely populated zone. The mode of farming used
by small farmers in the production of food crops is the slash-and-burn fallow system.
Because of the fragility of the soils, a long fallow period is called for, preferably after
every cropping season. Consequently, labor is required each crop year to open new
plots of cropland by slashing, burning, and clearing. The poor state of roads and other
infrastructure severely limits marketing possibilities for crops.
The study points up two basic facts: the amount of resources, particularly labor,
allocated to farming in the Zairian Basin is small, even by African standards, and the
productivity of these resources is also low. During the 1982/83 crop year, an average
of 271 man-hours was devoted to agriculture per man-equivalent (which takes into
account differences in age and sex) out of 776 man-hours devoted to economic activities.
Farming's share of productive labor time was only about 34.9 percent. The remaining
work time was allocated to hunting, fishing, gathering, handicrafts, and commerce.
The study also reveals that only a small amount of capital was allocated to farming,
and this was limited to a few hand tools and rice and maize seeds produced on-farm.
On average, each household spent about 75.40 zaires (Z) on tools and Z25.34 on seeds
during the 1982/83 crop year (Z5.75 equaled US$1.00 in 1982/83). Because so little
labor and capital (tools) were allocated to land preparation, the amount of area cultivated
was also severely limited. A 1.00 percent increase in labor devoted to land preparation
would increase cultivated area by 0.75 percent, and a 1.00 percent increase in tools
would increase it by 0.25 percent. As it was, cultivated area per household was only
0.67 hectare, about 0.22 hectare per man-equivalent.
The analysis also indicates that, despite the small area cultivated, the amount of
labor input per hectare was high compared with other parts of Sub-Saharan Africa and
Asia-about 1,230 man-hours per hectare during the 1982/83 crop year. Overall, a 1.00
percent decline in cultivated area would increase labor use per hectare by 0.42 percent.
In the study area, labor was more productive on cassava and plantain than on rice
and maize. The average product of labor was about 0.58 kilogram of cereal-equivalents
per man-hour for rice and 1.22 kilograms for maize in 1982/83. For cassava, the
productivity of labor was 6.62 kilograms and for plantain, it was 3.17 kilograms of








cereal-equivalents per man-hour. Overall, the productivity of labor input was estimated
at about 3.08 kilograms in cereal-equivalents per man-hour, and labor input was the
major contributor to overall food crop output in the Zairian Basin. Labor contributed
78 percent to food crop output and capital contributed 22 percent.
The study indicates that, given the current agricultural resource base, the produc-
tivity of labor input could be significantly increased if each household in the area were
to allocate more labor to farming, especially to land preparation, because labor produc-
tivity increases with the size of cultivated area. Domestic terms of trade between crop
and noncrop products and the availability of capital are the key variables affecting the
amount of labor allocated to farming in the Zairian Basin. The elasticity of labor input
devoted to farming per man-equivalent with respect to the domestic terms of trade of
farm relative to nonfarm products was 0.38 and that with respect to capital input was 0.28.
Policies to improve domestic prices for farm products and the amount of capital
input should include development of infrastructure such as roads; transportation, dis-
tribution, and marketing facilities; and rural capital markets. In addition, investments
in new technology such as improved seeds and fertilizers offer some opportunity for
improvement, particularly in the production of cereals in the area.









2


INTRODUCTION

Background and Objectives of the Study
Since the 1970s, per capital food and agricultural output in Sub-Saharan Africa have
been declining as a result of the low land and labor productivities that characterize
African smallholder agriculture. Economic studies have mostly approached the develop-
ment problems of smallholder agriculture in terms of land-labor relationships, and in
so doing, they have underestimated the role of capital in the development process
(Cleave 1974; de Wilde 1967). These studies have at least partly influenced agricultural
policies and led many policymakers to believe that food and agricultural output can be
increased mainly through land expansion. As a result, the introduction of capital inputs
(such as fertilizer, high-yielding seeds, pesticides, irrigation, and equipment) and im-
proved agricultural techniques in the small-farm sector, as well as the development of
physical infrastructure and rural institutions, have received limited attention. This
largely explains why the productivity of land and labor is currently so low.
In comparison, in Asia, where land expansion is no longer a feasible solution to
growth in food and agricultural output as a result of high population pressure, researchers
and policymakers reached a consensus that a fundamental structural change in the
land-labor-capital relationship had to be initiated if a respectable growth in food and
agricultural output was to be achieved. In the process, this consensus led to a policy
framework that promotes extensive use by smallholders of capital inputs, which encom-
pass both land-augmenting and labor-consuming technologies. This technological change
has led to a substantial upward shift in the productivity of both land and labor in Asia
(Hayami and Ruttan 1985; Mellor 1976).
Evidence pertaining to Sub-Saharan Africa suggests that the expansion of cultivated
area in the region resulted in a modest increment in food output during the 1961-80
period, while the share of yield increase in the growth of food output was negligible
(Paulino 1987). Consequently, the rate of growth of food output has not kept pace
with population growth in the region. This situation has led to massive food imports,
which have drained limited foreign exchange earnings and hence constrained the ability
of the countries in the region to improve and expand the productive capacity of agricul-
ture and of the rest of the economy.
Zaire is one of the countries in Sub-Saharan Africa where the food supply situation
has been deteriorating at an alarming rate for at least two decades. The annual average
growth rate of domestic staple food crop output, which was about 4.0 percent during
1966-70, declined to 1.6 percent during 1971-81. The annual average growth rate of
cereal imports decreased from 11.2 percent during 1966-70 to 10.4 percent during
1971-81, due to restrictive trade and exchange rate policies. The annual average growth
rate of the total staple food crop supply (including imports) declined from 4.1 percent
during 1966-70 to about 1.8 percent during 1971-81, while the annual population
growth rate was about 2.8 percent during the period (Tshibaka 1986).
The widening gap between food supply and population growth has created hardships
for many of the people of Zaire. A 10-year field study on nutrition and health revealed
that malnutrition has been spreading rapidly (World Bank 1980). Average calorie intake
was only 80-90 percent of minimum requirements, indicating that food production








and imports fell short of meeting growing food needs. To reverse the trend, the Zairian
government initiated a number of agricultural projects, including programs to expand
production of rice, maize, cassava, and pulses, and introduction of grain marketing
boards and farmer cooperatives. Most of these projects failed to meet expectations
(Kapambwe 1974; Lititiyo 1977). Limited empirical evidence indicates that they were
formulated and implemented with little knowledge about key constraints affecting the
development of the small-farm sector. For example, in the rain forest zone known as
the Zairian Basin, it was found that the pattern of labor allocation among different
household activities constitutes one of the major constraints to increased farm output.
A survey conducted in 1974/75 on a small sample of 17 households in the Zairian
Basin, specifically in Yalibwa, Turumbu Community, indicated that farming claimed only
11 percent of adult male labor time and 27 percent of adult female time (Tshibaka 1975).
A detailed study of the farming system was needed to reveal not only the key
determinants of labor allocation, but also other major constraints to increased food
output. To this end, this study examines the following basic questions. First, what is
the household resource base for small farmers? Second, how are these resources allo-
cated between farming and nonfarming? Third, what are the key determinants of this
resource allocation, and what are the implications for the farm sector? And fourth,
what is the level of resource productivity, and what actions need to be initiated to
improve it?

Data Source and Limitations
A joint study by the Institut Facultaire des Sciences Agronomiques, Yangambi, Zaire,
and the International Food Policy Research Institute, Washington, D.C., was undertaken
with the support of the International Development Research Centre of Canada to
provide basic data that could help the government of Zaire formulate consistent agricul-
tural policy to enhance the production of rice and other food crops in the Zairian Basin
in particular and in the country as a whole.
The Zairian Basin is part of a tropical rain forest that covers almost 40 percent of
the Zairian land area. Tree crops-robusta coffee, palm oil, rubber, and cocoa-are
mainly grown in this forest zone. The zone is also the major growing area for rice,
cassava, and plantain in the country. Zaire is currently divided into 11 regions (prov-
inces): Kinshasa, Equateur, Haut-Zaire, Maniema, Nord-Kivu, Sud-Kivu, Shaba, Kasai
Oriental, Kasai Occidental, Bas-Zaire, and Bandundu. The rain forest includes a part
of each of these provinces, except Kinshasa and Shaba. Each province (region) is, in
turn, divided into districts subregionss) and each district into zones (counties).
The data used in this research report were derived from a cost route survey con-
ducted from September 1982 to August 1983 by the author in Turumbu, Mongandjo,
and Bomaneh rural communities located in Isangi and Basoko zones, Tshopo District.
This subregion is located in the rain-forest part of Haut-Zaire Province (see Figure 1).'
The study area covered in the survey was divided into six subareas on the basis of
available infrastructure (the road network system), and a sample of 180 households,
30 from each subarea, was drawn. To increase the sample efficiency (to lower the
sampling variance), the 30 households were chosen following a proportionate, stratified


I This type of survey has been widely used in Africa. For detailed information, see: Purdue University
(1980), Zuckerman (1979), Matlon et al. (1979), Shapiro (1973), Norman (1973a), Elliott et al. (1970),
and Winch (1976).








Figure I--Map of Zaire


Districts of Haut-Zaire Province


Study area

D Zairian Basin


Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).








sampling procedure (Rossi et al. 1983; Upton 1973). First, a number of villages were
randomly drawn in each subarea. Second, taking account of the population of selected
individual villages, a given number of households were randomly drawn in each village.
One enumerator and an aide were posted in each subarea for one crop year. Two
research assistants were allocated to the project to help supervise the survey. Selected
households were visited by enumerators four times a week for one crop year and two
techniques of data collection were used: interview and direct measurement. For various
reasons, 48 households did not provide all the required data, and were eliminated.
Hence, at the end of the survey, data from only 132 households were retained. Data
were collected on population, infrastructure, input, output, and prices. Enumerators
measured perimeters and triangles of land under crops using tape measures, compasses,
and pickets, and cultivated area was estimated by the triangulation method.
The most serious limitation of this type of survey has always been the assessment
of time spent on different activities, which is affected by farmers' own judgment and
that of the enumerators. With regard to capital, farmers do not easily recall the monetary
value of their tools and the date of acquisition. These limitations are hard to avoid in
a situation where most farmers are illiterate or do not keep records.
In a system where mixed cropping is the most prevalent mode of production, the
estimation of cultivated area, labor, and capital allocated to individual crops presents
a number of unresolved practical and methodological problems. To circumvent these
difficulties, the researcher is tempted to convert all food crop outputs into cereal
equivalents and to proceed with the computation of input/output relationships without
having an idea of how much area is allocated to individual crops. This approach is
inappropriate for a number of reasons. First, there is a hierarchy of crops in mixed
cropping. There are major and minor crops in the same field. For example, other things
remaining the same, a field sown mainly with rice is different from a field mostly
planted with cassava, both in resource requirements and yields expressed in cereal
equivalents. Hence, an estimate of the share of area allocated to each crop is indispensable
for reflecting the effects of various crop mixes. Second, it would be difficult to answer
a number of crop-specific questions that are of policy interest if crops were lumped
together, such as the productivity of resources used in the production of individual crops.
A statistical method proposed by the Institut National des Statistiques et d'Etudes
Economiques de France was adopted to estimate the cultivated area allocated to indi-
vidual crops. For a description of this method, see Appendix 1. The proposed method,
notwithstanding its limitations, provides estimates that are useful when studying a
mixed-crop farming system (Norman 1972).
The allocation of time spent on nonspecific farm operations such as land preparation
and weeding among individual crops was performed on the basis of area shares of
individual crops. For capital tools-mainly hoes, cutlasses, and hatchets-the labor
time allocated to each of the four crops expressed as a fraction of total labor time
devoted to farming was used as a weight to calculate the depreciation of capital tools
(in zaires) used in the production of each of the crops. This approach is supported by
the fact that most of the tools used in the area are hand-held; therefore, their use is
highly correlated with labor input per crop. For example, tools must be used much
more extensively on 1 hectare of rice than on 1 hectare of plantain. To estimate crop
densities and yields, a yield plot method was used (see Appendix 1).









3


THE STUDY AREA


Ecology of the Study Area
The study area is part of the Zairian Basin, a forest zone in the humid tropics of
Africa. Rainfall averages about 1,800 millimeters a year with bimodal precipitations.
There are two rainy seasons-one between March and May and the other between
August and November-and two dry seasons--one between January and February and
the other between June and July.
The area receives an average of 1,972 hours of sunshine per year, representing 45
percent of the days of possible sunshine, most of it concentrated during the drier periods
of the year. The limited sunshine constitutes an important ecological constraint to crop
production. Temperatures in the area average 25.50 C (Crabbe 1970; Bernard 1945).
Soils are leached. High surface heat and hard, frequent rains lead to rapid breakdown
of organic soil materials (Sanchez 1976). As a result, fertility is low. Once land is
cleared and put under cultivation with annual crops, soils are rapidly depleted. The
forest that covers the Zairian Basin is rich in flora and fauna. Birds, pests, and diseases
constitute a major environmental threat to food crop production in this area.

Small-Farm Agriculture in the Zairian Basin
The small-farm sector, the core of this study, includes subsistence-oriented farmers
who cultivate less than 5 hectares of land each crop season. Inputs are limited to simple
hand tools such as cutlasses, hoes, and hatchets, on-farm crop seeds, and household
labor. Use of hired labor is uncommon in the area. Land is abundantly available to
every household. All the staple food crops-cassava, plantain, maize, and rice-are
produced under a mixed cropping system.
Small-farm production of robusta coffee is negligible and production of cocoa and
natural rubber by small farmers is almost unknown. None of the households selected
in the sampling process produces coffee. A substantial share of households, however,
exploit wild palm oil trees. Since the palm oil tree is a native of tropical Africa's rain
forest, which includes the Zairian Basin, the small farmer's production of palm oil is
considered to be a gathering rather than a farming activity. As for livestock, the rain
forest zone of Sub-Saharan Africa is not suitable, especially for cattle raising, because
of the tsetse. Although a few households keep a small number of goats, pigs, or chickens,
livestock raising as an economic activity is negligible. The study area derives most of
its animal protein requirements from hunting and fishing.
The forest area of the humid tropics of Africa is ecologically more suitable to
perennial (tree) crops than to annual (food) crops (Jurion and Henry 1967). Since the
colonial era, the Zairian Basin has been the major tree-growing area in the country.
The main tree crops produced are robusta coffee, cocoa, natural rubber, and palm oil
for export. These crops are mainly produced on privately owned, large-scale plantations,
using modern equipment and machinery, hired labor (who live on the plantations),
and modern agricultural and managerial techniques. Because roads in the study area
have deteriorated drastically, few plantations are now operating there.








Tree crop agriculture uses the land-the most abundant resource in the country-
most intensively, and it approximately recreates the natural environment by replacing
the natural forest with an artificial one made up of tree crops. As such, its negative
effect on the environment, particularly on soil and forest conservation, is small. In
contrast, small-farm agriculture uses the land extensively under a shifting mode of
cultivation. That is, the same farm fields are planted with a mix of rice, maize, cassava,
and plantain. Maize is harvested three months after sowing, and rice in four, whereas
cassava and plantain occupy the land for a period that extends into the new cropping
season, hence compelling farmers to open new plots of forest or fallow at the beginning
of each cropping season. When cassava and plantain crops are harvested, the fields are
then left fallow. Under this cassava-based system, farm fields are generally cropped for
one season and then left fallow for a number of seasons.
It follows then that, from a long-run ecological standpoint, in the Zairian Basin, the
appropriate small-farm agriculture to promote is a tree crop production system. Once
food markets begin to operate adequately, small farmers could purchase food, but in
the short run food can be produced for home use at much lower factor returns than
in tree crop production. The logical tree crops to promote among small farmers are
palm oil, coffee, and cocoa.
The first step in an effort to enhance regional crop specialization in the country is
to initiate a well-planned, long-term investment program to develop infrastructure such
as roads, river transport, and distribution and marketing channels.

Study Population
The study area, located in Haut-Zaire Province, Tshopo District, includes three
communities: Turumbu people in the administrative zone of Isangi; Basoo people in
Bomaneh, Bomenge Rural Community; and Bangelima people in Mongandjo Community
in the administrative zone of Basoko.
These ethnic groups mainly depend on farming, hunting, gathering, and fishing for
their living. The total population of the area was estimated at 49,103 people in 1981. Its
annual population growth rate was estimated at about 1.2 percent, which was substan-
tially lower than the national average growth rate of 2.8 percent per year (Zaire 1983).
Data are drawn from 53 households in Turumbu Community, 53 households in
Mongandjo, and 26 households in Bomaneh, Bomenge Community (see Table 1). The
sample population of 885 people is 48.1 percent male and 51.9 percent female.
The most productive age group, the 20-55 age bracket, accounts for 16.5 percent
of males and 22.7 percent of females. The 15-19-year-olds, the group that will be most
productive in the near future, includes 4.1 percent of the males and 3.4 percent of
the females. The age groups under 15 and above 55 account for 27.6 percent of the
males and 25.8 percent of the females.2 The sample population of 885 people represents
a total labor force of 400.8 man-equivalents of which 217.6 are males and 183.2 are
females. The factors used to convert people of different ages and sexes to man-equivalents
are given in Appendix 1.3 This gives an average of 3.9 man-equivalents per household
in Turumbu, 2.5 in Mongandjo, 2.4 in Bomaneh (Bomenge), and 3.0 in the study area



2 In Zaire, farmers more than 55 years old are regarded as retired.
3 For more details see Collinson (1972). The total number of man-equivalents per household are given in
Appendix 1, Table 28.








Table 1-Structure of the sample population in the study area

Turumbu Villages Mongandjo Bomenge
Yangambi- Total Total Study
Kisangani Wekoand Yangambi- Turumbu Mongandjo Area
Sex/AgeGroup Road* Yambaw IsangiRoadb Community Bolikango Babendja Community Bomaneh Total
(number of people)
Males
Less than 1 year old 2 3 0 5 1 0 1 6 12
1-4years 12 15 9 36 12 6 18 8 62
5-9years 9 9 12 30 13 12 25 14 69
10-14years 13 8 6 27 3 9 12 9 48
15-19years 14 5 6 25 6 1 7 4 36
20-55years 26 24 27 77 21 21 42 27 146
More than 55 years old 5 10 8 23 12 16 28 2 53
Total males 81 74 68 223 68 65 133 70 426
Females
Less than 1 year old 2 6 4 12 3 3 6 7 25
1-4years 9 5 9 23 6 7 13 14 50
5-9years 12 12 13 37 9 11 20 9 66
10-14years 7 8 5 20 3 8 11 9 40
15-19years 5 9 9 23 4 0 4 3 30
20-55years 34 31 30 95 35 34 69 37 201
More than 55 years old 6 9 4 19 14 14 28 0 47
Total females 75 80 74 229 74 77 151 79 459
Total number of people 156 154 142 452 142 142 284 149 885
Number of households 17 17 19 53 25 28 53 26 132

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut Facultaire des Sciences Agronomiques,
Yangambi, Zaire, and International Food Policy Research Institute, Washington, D.C., 1986 (computer printout).
a Households in this group are located along the Yangambi-Kisangani Road up to 22 kilometers from Yangambi.
b Households in this group are located along the Yangambi-Isangi Road up to 20 kilometers from Yangambi.







as a whole. The 20-55 year age group, relatively more involved in farming, accounts
for 70.0 percent of the overall sample of which 36.4 percent are males and 33.6
percent females.


Basic Infrastructure in the Study Area
The Zairian Basin is the least developed region in the country, which is reflected
in the poor state of the road network and consequently the transportation system.
Most villages in the region are isolated or poorly connected to major markets and
urban centers. There are differences from one group of villages to another in physical
condition of the roads and distances to rural markets or urban centers, making some
villages more accessible than others. Most Turumbu villages are served by the Zaire
River and the Isangi-Yangambi-Kisangani road. Local traders provide the sole means of
transportation on the river, the traditional canoe. The traders are Lokele people, who
play an important role in the distribution and marketing of farm and nonfarm com-
modities produced by Turumbu farmers. Although canoes are used all year round, the
volume of goods transported remains small, and the risk of shipwreck high.
The Isangi-Yangambi-Kisangani road, considered to be a major national road, is
unpaved. Even during relatively dry periods of the year, the use of this road by vehicles
is limited. The 97 kilometers that separate Yangambi and Kisangani may be covered
in three hours in the best of weather conditions. During wet periods, the road is seldom
used by vehicles. The high level of precipitation in the Zairian Basin makes the use of
unpaved roads by vehicles difficult during most of the year. As a result, Turumbu
villages along this axis are poorly connected to Yangambi and Lotokila-the major local
markets in the community-as well as to Kisangani, the major urban center. Farmers
living along this road usually walk 4-23 kilometers, depending on the location of their
villages, to reach a rural marketplace.
A second group of Turumbu villages, even more isolated, includes Weko and Yambaw
in the remote forest. The road connecting these villages to Yangambi has deteriorated
to such an extent that it is no longer used by vehicles at all. Farmers bring their produce
twice a week to a local marketplace (which is 1 kilometer from Weko and 12 kilometers
from Yambaw) for sale to traders, most of whom come from Yangambi by foot or bicycle.
The distance from this local market to Yangambi is about 42 kilometers.
In Mongandjo Community, farm households are served by the Aruwimi River and
two feeder roads. The 42-kilometer road connecting Bolikango to Basoko is the best
maintained in the entire study area. It was in fair condition for seven months of the
survey year, in poor condition for three months, and in very poor condition for two
months. Farmers walk on foot 5-23 kilometers to reach three rural markets: Longote,
Bokote, and Baonde. The first two rural marketplaces are along the Aruwimi River,
while the third one is at the point where the Basoko-Baonde route crosses the Aruwimi
River. Farm and nonfarm products sold there are transported by traditional canoes or
vehicles to Basoko, an important river port and marketplace.
The 190-kilometer road connecting Babendja in Mongandjo Community to Basoko
is the second best road in the area. This road was in fair condition for five months,
and in poor or very poor condition for the rest of the 1982/83 crop year. Babendja
farmers walk 4-13 kilometers to reach a local market, where farm and nonfarm products
are sold and transported to Basoko either by road or by the Aruwimi River. The Aruwimi
River is used all year round by small boats and canoes. The distance by water from
Babendja to Basoko is about 105 kilometers.








The 21-kilometer road connecting Bomaneh to Basoko in Bomenge Community is
no longer practicable for vehicles, and it is now left entirely to those on foot. Only the
Aruwimi River connects Bomaneh farmers to the two marketplaces, Baonde, 23 kilomet-
ers away, and Basoko, 24 kilometers. Canoes and small boats are the means of trans-
portation used by farmers to reach these rural marketplaces.
Table 2 summarizes data on the conditions of the roads and the navigability of
rivers in the area. It shows that Mongandjo and Bomaneh (Bomenge) farmers are in a
somewhat better position than Turumbu farmers. The negative impact of this poor
road network on the development of local marketing channels cannot be overemphasized.
In all, the transportation system in the study area is grossly inadequate. For this and
other reasons, few plantations are now operating in the study area, and none of the
surveyed households were employed by plantations. This low level of infrastructural
development constitutes one of the most important constraints to increased food and
agricultural output in the area.


Distribution, Marketing, and Pricing of Farm Products
The distribution, marketing, and pricing of farm and nonfarm products were regu-
lated under different policy regimes until September 1983, when prices of all farm and
most nonfarm products were liberalized, and restrictions imposed on domestic trade
were lifted (Tshibaka 1986).
The distribution of cassava, plantain, other root and tuber crops, game meat, fish,
handicrafts, and products from gathering have never been controlled by the Zairian
government. The demand and the supply of these products were and continue to be



Table 2-Practicability of roads and navigability of rivers in the study area,
1982/83 survey
Distance from
Primary (Local)
TransportRoute Condition of Road/River to Secondary Community Villages
Destinations Fair Poor Very Poor Markets Served Served
(number of months) (kilometers)
Yangambi-Kisangani 0 5 7 97 Turumbu Villages along the
road Yangambi-Kisangani
road
Yangambi-Weko 0 0 12 42 Turumbu Villages in remote
forest (Weko and
Yambaw)
Yangambi-Isangi 0 5 7 97 Turumbu Villages along the
road Yangambi-Isangi
road
Basako-Baonde road 7 3 2 47 Mongandjo Bolikango
Basoko-Mongandjo 8 4 0 105 Mongandjo Babendja
(Aruwimi River)
Basoko-Bomaneh 8 4 0 24 Bomenge Bomaneh
(Aruwimi River)

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
Note: For rivers, "fair" indicates navigable and "very poor," not navigable.







regulated by domestic market forces. But crops such as rice, maize, groundnuts, cotton,
coffee, tea, palm oil, cocoa, and rubber were subject to government control until
September 1983.
Due to the poor state of the road network in the study area, the distribution,
marketing, and pricing of rice and maize were not totally enforced. But rice millers
who were able to reach the villages applied some political pressures to persuade farmers
to accept the official price for their paddy rice. A substantial portion of paddy, however,
was and still is converted into rice at the household level, using mortar and pestle.
This is the traditional method used in every part of the country where rice is grown.
Although the method produces a low-quality product from a commercial standpoint
with a significant amount of broken rice, the nutritional value of the rice is preserved,
as most vitamins and minerals are retained. This rice product is sold by female farmers
in the local market to any potential trader or consumer at the market price. Regarding
maize, most of it is converted into an alcoholic drink favored in the study area. The
price of this drink is determined by local market conditions. Rice and maize are the
major source of cash income from farming in the study area.
At the local level, three groups of participants carry out most of the distribution
and marketing of farm products. These include farmers, small transporters, and petty
traders, most of whom are women. Cassava products offered for sale by farmers are
fresh roots, dry roots, chickwangue (cassava bread), and leaves (Tshibaka and Lumpungu
1989). Fresh plantain, paddy rice, milled rice, and maize grains are the most common
forms in which plantain, rice, and maize are offered for sale by small farmers. Because
the infrastructure- roads and other transportation facilities-is inadequate in the study
area, leaves, fresh cassava roots, chickwangue, and plantain have a limited marketing
area. Being the most perishable, they are generally sold in the local markets. From this
description, the critical role of infrastructure in distribution and marketing is apparent,
as well as the need to promote the processing of farm produce into products that can
be stored and marketed over long periods of time.








4


LAND AND CAPITAL INPUTS

At the outset, it should be noted that this study is based on an implicit assumption
that all the resources allocated to the small-farm sector are used up in the production
of the four major crops, namely rice, maize, cassava, and plantain. Although minor
crops such as vegetables and fruits are also produced, they are not considered here.

Land and Related Aspects
Land Tenure System
Although the constitution of Zaire stipulates that land and all natural resources
belong to the state, Zairian ethnic groups still consider certain lands their property. In
the study area, all of the household heads interviewed affirmed that all members of
the clan, regardless of sex, religion, or political affiliation, have free access to land and
other natural resources (forests, streams, and rivers) belonging to the clan. These
resources are common property and cannot be alienated.
The traditional land tenure system in the area, however, seems to be tacitly moving
toward private ownership of land. Sixty-four percent of the surveyed household heads
interviewed affirmed that every household retains the right to use their fallow land.
This right is extended to household male descendants. In practice, this amounts to
private ownership of land. One should note that this practice does not imply that land
is scarce in the area. Indeed, the population density does not exceed five persons per
square kilometer.
Furthermore, the survey shows that in Turumbu Community payment in cash or
kind by nonmembers of the clan is accepted where the right to fallow land is recognized.4
Among Bangelima and Basoo in Mongandjo and Bomenge, there is no correlation
between payment and the right to use fallow land. Table 3 presents this data. It should
be stressed that payments are in no circumstances regarded as a sale price of land or
as rent.5 From the traditional point of view, land cannot be bought or sold. It remains
a common property of the clan, and the nonmember is allowed to use it as long as he
lives with the clan or has not fallen under disgrace.
These are only a few aspects of the complex traditional land tenure system related
to farming, which varies from one clan to another within the same ethnic group. A
common feature of all these traditional land tenure systems is that land and other
natural resources are common property.
Regarding the government's role in land use and management, 69.6 percent of farmers
interviewed believed that the government should play a monitoring role. The rest of
the farmers considered government intervention likely to create additional problems.


4 The first household to open a piece of land keeps the right to use that land as long as household members
continue to live in the village, unless they state explicitly that they are no longer interested in it. At that
time, the land returns to the common pool.
5 The payment is made only once, which differs from rent that has to be paid continuously as long as the
land is being used. It is given by the nonmember of the clan in appreciation for the right to use the land,
which the clan has extended.








Table 3-Community recognition of practices of right of fallow and lease of
land to nonmembers in study area
Number of Community Community
Household Accepts Recognizes
Heads Payment Right of
Rural Community Interviewed for Land Fallow
(percent)
Turumbu
Yangambi-Kisangani road 17 65.4 100.0
Weko and Yambaw 17 4.8 9.0
Yangambi-Isangi roadb 19 23.8 60.0
Mongandjo
Bolikango 25 0.0 8.0
Babendja 28 7.1 25.0
Bomenge
Bomaneh 26 0.0 80.8

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
a The households in this group are located up to 22 kilometers from Yangambi.
b The households in this group are located up to 20 kilometers from Yangambi.



Land Quality
Many of the farmers interviewed claimed that the presence of certain tree species
provides a good indication of soil fertility. This technique, although crude, has some
scientific backing. About 79.7 percent of the household heads attested that forest area
to be cleared for farming was selected based on this method.
Once the selection of forest area is made, the traditional head of the clan, in
collaboration with an agricultural agent, delimits and divides the area among various
farm households within the limits of each household's resources. This practice, insti-
tuted by the Colonial Administration to ensure an orderly use of the forest for agricultural
purposes, is still observed today. About 81 percent of the households surveyed had
their farm fields in the portion of forest preselected for farming in 1982/83. The rest
of the households had their farm fields scattered in the forest.

Land Availability and Use
Exact data on land area held by individual clans are not available. All household
heads interviewed agreed that land is not a constraint to production. It is abundant
and available to any member of the clan. For instance, the Turumbu Rural Community,
with a population of 29,532 in 1981, had an area of 4,000 square kilometers. This
represents a density of about seven persons per square kilometer. It should be stressed,
however, that the area of land that can be cropped is limited by the fact that hand
tools have to be coupled with human force to clear the land, an especially energy-consuming
activity in the forest zone. The Zairian Basin is clearly a land-surplus, labor-scarce area.
About 80.5 percent of the households exploit the land for two cropping seasons
before putting it to fallow. Table 4 shows that 41.9 percent of the households leave
their fields to fallow for a period of time ranging from three to six years, while the
remaining 58.1 percent of the households put their fields to fallow for at least seven years.
Several measurable changes in tropical soil characteristics occur during the cultiva-
tion period, including increased acidity, decreased levels of available plant nutrients,








Table 4-Length of fallow period, by households surveyed, 1982/83
Percent of
Surveyed
Fallow Period Households
(years)
Less than 4 4.0
4 5.3
5 17.3
6 15.3
7 14.7
8 11.3
9 6.7
10 10.7
More than 10 12.7
Forest 2.0
Total 100.0

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
a Two percent of the households surveyed had plots in the primary forest.



loss of organic matter, greater compaction, decreased storage of available water, and
decreased infiltration capacity. Marked changes in the faunal population and micro-
biological activity also occur. These detrimental changes undoubtedly influence crop
yields increasingly as cultivation is extended beyond one or two cropping cycles or the
length of the fallow period is reduced. A minimum fallow period of seven years is
necessary to maintain soil fertility in the tropics under a traditional farming system
(Greenland and Okigbo 1983).
A minimum fallow period of 10 years was instituted during the Colonial Adminis-
tration. Most measures initiated during this period were enforced through coercion
rather than persuasion (urion and Henry 1967). This allegation does not imply, how-
ever, that all the measures taken during the period before independence were wrong
or unnecessary.
A positive question one should ask is why did a substantial share of farmers shorten
the fallow period to less than seven years despite the low population density? One
possible explanation may be that reducing the labor input required for clearing by
shortening the fallow period could well be an attempt to increase labor productivity
where opportunity costs for labor are rising, at least outside the farm sector.
The effects of shortening the fallow period on the intensities of labor and capital
use, as well as on the productivities of land, labor, and capital, will be examined
subsequently in order to make an economic judgment on farmers' behavior.

Allocation of Cultivated Area among Crops
The cultivated area per household (0.67 hectare) and the cultivated land per man-
equivalent in the study area (0.22 hectare) are small by any standards (Table 5).
Perreault (1978) found that the cultivated area per household was 0.88 hectare for
Yalenge village, 0.94 hectare for Banakanole, and 0.82 hectare for Yambela in the
Zairian Basin. All these estimates are smaller than those reported in other regions of
the country. Durocher (1983) estimated an average cultivated area of 1.02 hectares
per household in Nande Rural Community, Nord-Kivu, and Ntamulyango (1975) re-








Table 5-Cultivated land/man ratio and allocation of cultivated area among
crops
Man-
Equiva-
Number lents/ Cultivated Cultivated
ofHouse- House- Area/ Area/Man- Allocation ofArea
Rural Community holds hold Household Equivalent Rice Maize Cassava Plantain
(hectare) (percent)
Turumbu 53 3.88 0.66 0.17 27.61 26.92 33.14 12.32
(0.51)
Yangambi-Kisangania 17 4.24 0.85 0.20 9.13 37.65 32.73 20.49
(0.77)
Weko and Yambaw 17 3.99 0.50 0.13 52.15 21.57 23.29 2.99
(0.25)
Yangambi-Isangib 19 3.46 0.63 0.18 22.19 22.11 42.33 13.37
(0.35)
Mongandjo 53 2.52 0.75 0.30 39.19 9.33 51.48 0.00
(0.33)
Bolikango 25 2.68 0.73 0.27 32.13 11.97 55.90 0.00
(0.32)
Babendja 28 2.37 0.77 0.32 45.50 6.97 47.54 0.00
(0.34)
Bomenge
Bomaneh 26 2.38 0.53 0.22 8.75 34.99 54.58 1.68
(0.33)
Study area 132 3.04 0.67 0.22 28.55 21.45 44.73 5.28
(0.42)

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
Note: The numbers in parentheses are standard deviations.
a These villages were located up to 22 kilometers from Yangambi.
b These villages were located up to 20 kilometers from Yangambi.



ported an area of 1.48 hectares per household in Bushumba, Sud-Kivu. These figures
are not given for comparative purposes. The areas mentioned in these two Kivu regions
are in savannah zones. Land preparation requires more energy in the forest zone than
in the savannah. In all, the cultivated area per household and cultivated land-man ratio
remain small despite the availability of land, both in the Kivu regions and in the Zairian
Basin. The key factors affecting the size of cultivated land area and the implications
for farm output and labor productivity will be considered later.
Crop area estimates reported in Table 5 show wide intercommunity variability. In
Turumbu, cassava claims the largest share of the cultivated area. Rice, maize, and
plantain are second, third, and fourth. In Mongandjo, cassava has the largest share of
area followed by rice and maize. In Bomaneh, Bomenge Rural Community, cassava
occupies the first position; maize, the second; rice, the third; and plantain, the fourth.
Overall, cassava covers the most area, followed in decreasing order by rice, maize, and
plantain.
Variability in the shares of area devoted to individual crops among villages within
the communities reflects both local market conditions and the degree of access to
markets in the major centers. The communities with relatively large local markets tend








to allocate more of their cultivated area to cassava, maize, and plantain. The Turumbu
villages located along the Yangambi-Kisangani road allocated 90.8 percent of cultivated
area to cassava, maize, and plantain, and those along the Yangambi-Isangi road 77.8
percent, with rice claiming the rest. These villages have relatively large local markets
(at Yangambi, Lotokila, and Isangi), compared with Weko and Yambaw, which are
located in remote forest areas without substantial local markets. They are connected
to Yangambi by a very poor road of about 42 kilometers. In Weko and Yambaw,
households allocated 52.2 percent of cultivated area to rice alone. Because of its biolog-
ical nature, rice can be stored and sold over a relatively longer period of time than
maize, cassava, and plantain. Therefore, in a remote area without a large local market
and poorly connected to major centers, farmers tend to allocate more of their cultivated
area to crops that can be stored under farm conditions with limited loss. For this reason,
rice has become the major cash crop in poorly connected, remote areas of the Zairian Basin.
This same pattern of land allocation between rice, cassava, and maize also holds
for Mongandjo Community. (Production of plantain in Mongandjo is severely con-
strained by monkeys.) In Bolikango, which is much closer to the major market of
Basoko than Babendja, farm households allocated 67.9 percent of the cultivated area
to cassava and maize as opposed to 54.5 percent in Babendja. Farmers in Bolikango
devoted about 32.1 percent of the cultivated area to rice as opposed to 45.5 percent
in Babendja. In Bomaneh, rice is a new crop that was introduced into the farming
system in 1981/82. A comparison of rice and other crops does not hold for this community.

Capital Input
As stated earlier, capital inputs available to the peasant farmer include a few hand-
held tools and rice and maize seeds produced on-farm.
Quantity of Rice and Maize Seeds Used
The quality of seeds being used in the Zairian Basin has severely degenerated since
Independence in 1960, because until recently, no seed service existed in the country.6
Although data on technically optimal quantities of seeds to be used in a mixed-cropping
system are not available, Table 6 indicates that, for the area as a whole, farmers used
59.6 percent for rice and 79.3 percent for maize of the quantity recommended for use
in pure cropping by the Institut National d'Etude et de la Recherche Agronomique du
Zaire (INERA 1971).
There is a great deal of divergence among communities and also among villages.
For Turumbu Community, the amount of rice and maize seeds used per hectare repre-
sents 50.5 percent and 67.0 percent respectively of the amount recommended for pure
cropping. In Mongandjo Community, 63.0 percent of rice seeds and more than 100.0
percent of maize seeds were sown. In Bomaneh, rice seeds amounted to 67.3 percent
of the quantity recommended for pure stands, and maize seeds, 62.0 percent. On the
basis of these data, it seems likely that farmers who planted maize in Mongandjo in
1982/83 used too much maize seed per hectare for mixed stands, rather than too
little. For the sample as a whole, 60.6 percent of surveyed households produced rice
and 62.1 percent produced maize in 1982/83.



6 A new seed project sponsored by the World Bank was established in 1985 and is now operational. It has
to be given some time, however, before its impact on the small-farm sector can be assessed.









Table 6-Quantity of seeds used per hectare by households that produce rice
and maize, 1982/83
Amount of Seeds
Used in Mixed
Cropping as a
PercentofAmount
Number of House- Amount of Seeds Used Recommended in
holds Producing in Mixed Cropping Pure Cropping
Rural Community Rice Maize Rice Maize Rice Maize
(kilograms/hectare) (percent)
Turumbu 23 38 25.26 20.09 50.52 66.97
(5.26) (6.66)
Yangambi-Kisanganiroada 4 14 29.93 22.11 59.86 73.70
(2.22) (5.99)
WekoandYambaw 12 16 25.47 21.64 50.94 72.13
(5.27) (6.68)
Yangambi-Isangi roadb 7 8 22.23 13.45 44.46 44.83
(4.76) (2.88)
Mongandjo 53 31 31.50 30.48 63.00 100.16
(7.23) (7.26)
Bolikango 25 20 32.71 30.44 65.42 101.47
(8.04) (7.99)
Babendja 28 11 30.41 30.57 60.82 101.90
(6.37) (6.07)
Bomenge
Bomaneh 4 13 33.66 18.60 67.32 62.00
(19.99) (4.12)
Studyarea 80 82 29.81 23.78 59.62 79.27
(8.12) (8.38)

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
Note: Standard deviations are in parentheses.
a These villages were located up to 22 kilometers from Yangambi.
b These villages were located up to 20 kilometers from Yangambi.



Tools
Because most of the tools used were hand-held, their use was highly correlated
with labor time allocated to each crop. Thus, labor shares of individual crops were
used as weights to estimate the depreciation of tools allocated to these crops during
the 1982/83 crop year. The estimated average depreciation of tools (material capital)
per household in the study area as a whole was about Z75.40 for the year (Table 7).7
A crop-based comparison shows that 42.5 percent of the tools in the study area
were used on rice, followed by 40.4 percent on cassava, 15.1 percent on maize, and
2.0 percent on plantain. Table 7 also indicates much variability between and within
communities in the amount spent on capital input (tools) and in allocation among



7 The same tools are used in the production of rice, maize, cassava, and plantain. In dollar terms, the cost
of tools used per household was about US$13.11, using the official 1982/83 exchange rate of Z5.75/
US$1.00.








Table 7-Allocation of capital tools among crops, 1982/83

Numberof Cost of Share of Tools Allocated
Rural Community Households Capital Tools Rice Maize Cassava Plantain
(Z) (percent)
Turumbu 53 49.52 41.52 21.98 31.61 4.88
(58.80)
Yangambi-Kisangani road 17 43.33 27.29 29.60 35.06 8.05
(57.93)
WekoandYambaw 17 47.64 57.47 13.55 26.73 2.25
(60.21)
Yangambi-Isangiroadb 19 56.74 38.25 23.50 33.43 4.82
(60.73)
Mongandjo 53 98.51 49.30 7.36 43.14 0.20
(45.93)
Bolikango 25 84.12 45.84 7.69 46.29 0.18
(51.59)
Babendja 28 111.36 53.14 7.06 39.58 0.22
(36.53)
Bomenge
Bomaneh 26 81.06 18.36 25.62 54.87 1.15
(44.36)
Studyarea 132 75.40 42.52 15.09 40.40 1.99
(55.49)

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
Note: Standard deviations are in parentheses.
a These villages were located up to 22 kilometers from Yangambi.
b These villages were located up to 20 kilometers from Yangambi.



crops. These inter- and intracommunity differences in capital input are also a function
of cultivated land area.

Allocation of Seeds and Tools among Crops
Table 8 summarizes the data on capital input (seeds and tools) used per household,
by crop. Rice received 48.7 percent of total capital input; cassava, 32.8 percent; maize,
17.2 percent; and plantain, 1.3 percent. Hand tools contributed a much larger share
to capital input costs than seeds.
A community comparison indicates that Mongandjo allocated more capital input to
farming than Bomaneh and Turumbu. An average household in Mongandjo spent 1.4
times more on capital than an average household in Bomaneh and 1.8 times more than
in Turumbu.
In sum, this survey shows that capital input is in short supply in the small-farm
sector of Zaire. All farmers interviewed suggested that the lack of an input delivery
system, credit facilities, and adequate roads constitute the main constraints to increases
in food output and income.









Table 8-Allocation of capital input (seeds and tools) among crops, 1982/83

Numberof Total Cost of Share of Total Capital Input Allocated
Rural Community Households Capital Input Rice Maize Cassava Plantain
(Z) (percent)
Turumbu 53 72.39 47.25 25.10 23.97 3.68
(76.06)
Yangambi-Kisangani road' 17 72.39 39.92 29.45 25.65 4.98
(91.00)
Weko and Yambaw 17 66.45 64.20 12.73 20.48 2.59
(76.30)
Yangambi-Isangiroadb 19 71.69 40.38 30.94 25.24 3.44
(64.11)
Mongandjo 53 130.90 59.70 7.73 32.46 0.12
(55.94)
Bolikango 25 117.44 57.55 8.85 33.50 0.09
(63.19)
Babendja 28 142.92 61.27 6.90 31.69 0.14
(46.47)
Bomenge
Bomaneh 26 97.09 20.85 31.32 46.96 0.87
(51.15)
Studyarea 132 100.74 48.73 17.22 32.76 1.29
(68.80)

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
Notes: Cassava cuttings and plantain shoots to be planted are collected from previous farm fields by farmers
themselves. The time spent on this activity is included in the labor input. Cassava cuttings and plantain
shoots are neither bought nor sold in the study area. Tools constitute the only items of capital input used
in the production of these two crops in the area. The prices of rice and maize seeds used to compute
the value of seeds are reported in Appendix 2. The numbers in parentheses are standard deviations.
' These villages were located up to 22 kilometers from Yangambi.
b These villages were located up to 20 kilometers from Yangambi.








5


LABOR USE AND ALLOCATION

This chapter presents the calendar of farm operations, the monthly intensity of
labor use during the 1982/83 crop year, and the allocation of labor time among crops.
In addition, the allocation of labor time between farm and other economic activities
performed in the household is discussed.
In this connection, one should recall that small-farm households in the Zairian Basin
solely rely on labor provided by household members to carry out economic, domestic,
and sociocultural activities. Economic activities are made up of farm and nonfarm activities
(such as hunting, fishing, and gathering). Most of these are food-producing activities.

Calendar of Farm Operations
and Monthly Intensity of Labor Use
A number of farm-level studies have shown that labor shortages, particularly during
peak periods, constitute one of the major constraints to food and agricultural production
in most parts of Sub-Saharan Africa (Eicher and Baker 1982; Delgado 1982; Norman
1973b). For this reason the sequences of farm operations as performed by farmers in
the area are presented, and attempts are made to verify whether labor shortages are
experienced during peak production periods in the Zairian Basin.
The official agricultural calendar of farm operations for the area, based on research
conducted at the Institut National pour l'Etude Agronomique du Congo (INEAC)8 during
the colonial period, is summarized in Table 9. This calendar assumes only one crop
season for maize and rice per year, although with the bimodal pattern of precipitation
that exists in the area, it is agronomically possible to have one major and one minor
crop for both rice and maize per year. From sowing to harvest, three months are
required to produce maize and four months for rice. For cassava and plantain, only
one crop can be produced a year. The official calendar, however, is not followed by
all the farmers in the area.
Table 10, based on Appendix 2, Table 29, presents the agricultural calendar followed
in each study community during the 1982/83 crop year. Because crops are planted in
mixed stands, farm operations such as land preparation and weeding are simultaneously
carried out for all crops. The sowing period, which runs from April to October, is more
critical for rice and maize than for cassava and plantain. Harvesting of each crop is
performed at the end of the cycle for that crop.
In all, farmers in the study area follow an agricultural calendar that is not strictly
the same as the official one. The execution of some farm operations is extended beyond
the recommended periods. Does this late execution of some farm operations imply
that the demand for labor to carry out various household activities is high and concen-
trated in time? In other words, is this late execution of farm operations an indication
of labor shortages in the area? Or is it merely a tendency among small farmers, partic-


8 The institute is now known as the Institut National d'Etude et Recherche Agronomique (INERA).









Table 9-Official calendar of farm operations in the study area

Period When Operations Operations to Be Conducted Operations to be Continued
Are to Be Performed in the New Farm Field on Last Season's Crops

January-February Slashing Weeding and harvesting of cas-
sava and plantain; marketing of
rice, maize, cassava, and plan-
tain.
March-April Burning and clearing of the new Weeding and harvesting of cas-
plot of land; sowing of rice and sava and plantain; marketing of
maize; planting of cassava and rice, cassava, and plantain.
plantain.
May-July Sowing of rice and maize; Harvesting and marketing of
weeding, cassava and plantain.
August-October Harvesting of rice and maize; Harvesting and marketing of
weeding of cassava and plan- cassava and plantain.
tain; selection of plots for the
new season.
November-December Processing and marketing of Harvesting and marketing of
rice; weeding of cassava and cassava. Field will be turned to
plantain. The new crop season fallow.
begins and farmers start clear-
ing the forest.

Source: Zaire, Division R4ginale de L'Agriculture, "Calendrier Agricole Approuvd pour la Sous/Region de la
Tshopo," Kisangani, Zaire, 1980.




Table 10-Farmers' agricultural calendar in the study area, crop year 1982/83

Turumbu Villages
Along Along
Yangambi- Yangambi-
Kisangani Weko and Isanl Mongandjo Bomenge
Farm Operation Roada Yambaw Road" Bolikango Babendja Bomaneh

Slashing January- January- January- January- January- January-
August May May August May May
Burning and clearing February- April- May- April- April- April-
May July July October June August
Sowing and planting April- May- July- May- April- April-
September September August October August August
Weeding May- July- August- August- August- July-
August December October December December December
Harvesting, processing,
and marketing July- September- October- August- August- August-
December December December December December December

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
Notes: This calendar includes only farm operations carried out in the ongoing crop year. Operations related to
last crop season performed mainly by female members are weeding of cassava and plantain and marketing
of rice and maize during the January-February period. Harvesting and marketing of cassava and plantain
continue throughout the year.
SThese villages were located up to 22 kilometers from Yangambi.
b These villages were located up to 20 kilometers from Yangambi.








ularly those living in an area where rainfall is relatively well distributed throughout
the year, to be flexible about the timing of farm operations?
The survey data summarized in Table 11 show that, for the study area as a whole,
the largest amount of labor time allocated to household economic activities in a month
was about 88 hours for males and 85 hours for females during the surveyed crop year,
while the smallest amount of labor time spent was about 71 hours for males and 23
hours for females. In the humid tropics of Africa, 6 hours a day or 150 hours a month
is generally regarded as the maximum a person can productively allocate to labor,
either farming or nonfarming; therefore, it seems that, other things being the same,
more time could have been devoted to both farming and nonfarming activities in the
study area without encountering a labor shortage. The breakdown of labor allocation
for each of the communities included in the study is given in Appendix 2, Table 30.


Seasonal Allocation of Labor Between
Farming and Nonfarming Tasks
Figure 2, which is based on Appendix 2, Table 30, shows that the intensity of labor
use throughout the year exhibits a clear seasonal pattern. Three distinct periods can
be delineated for males and four for females. During January-February, the amount of
time males allocate to farming increases and that spent on nonfarming activities declines.
During the February-April period, the amount of time males allocate to farming is at
its highest period. During April-December, the pattern is reversed, with nonfarming
activities increasing and farming decreasing. For female members, May-July and September-
December are the peak periods for farming activities. During these periods, the level
of nonfarming activities performed by females goes down. During the January-May and
July-September periods, however, the situation reverses in favor of nonfarming activities.


Table 1 1-Intensity of labor use, by month, 1982/83
EconomicActivities
(Farming and Non-
farming) Farming Nonfarming
Month Male Female Male Female Male Female
(hours/month)
January 73 32 24 9 49 23
February 85 32 42 0 43 32
March 73 32 40 1 33 31
April 79 23 39 5 40 18
May 77 27 28 15 49 12
June 78 44 24 30 54 14
July 88 61 23 36 65 25
August 85 85 16 32 69 53
September 80 54 14 25 66 29
October 71 43 14 31 57 12
November 71 58 12 37 59 21
December 80 52 9 36 73 16

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
Notes: These are the average number of hours allocated to farming and nonfarming economic activities by a
household member of either gender regardless of age. They should not be regarded as man-hours.
Nonfarming activities referred to are hunting, fishing, and gathering.











Figure 2-Intensity of male and female labor use, farming and nonfarming
activities, by month, 1982/83


Hours


Male


60



40



20


Feb. Mar. Apr.


May Jun. Jul. Aug. Sep. Oct.


Nov. Dec.


Female


Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov.


Dec.


-- Farming

-- -- Nonfarming


Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).




In short, the demand for labor during farming peak periods is met by substituting
farming work for nonfarming, rather than by reducing leisure.
Because of the division of labor, the allocation of labor time presents different
seasonal patterns for men and women. The time allocated to farming by males is large
from January to May, when land preparation is mostly performed. During this period,
very little time is devoted to farming by females. The reverse occurs during the period
from June to December, when farm operations such as planting, weeding, harvesting,
and farm-level processing are performed.


0L


Jan.


Hours


40 -




20 -


r\
r
I \
I \
I \
~I \
F~
I' \
I \I
I
I : :\


,.i N


-- --
--
5
--
--
C-
C
I
ii
5







Survey data indicate that high energy-consuming farm operations such as forest
slashing, burning, and clearing are almost entirely carried out by men. Less risky, low
energy-consuming operations such as planting, weeding, harvesting, processing, and
marketing are mainly performed by women, with substantial help from male members.
This division of farm operations between men and women is consistent with biological
considerations.
The area does not appear to suffer from the seasonal labor shortages that are
characteristic of subhumid and semiarid areas of Sub-Saharan Africa. The distribution
of rainfall, which is mostly responsible for seasonal labor shortages elsewhere in the
Sub-Saharan region is not a factor here. Key factors that affect the allocation of household
labor time to farming in the Zairian Basin still have to be identified.


Allocation of Agricultural Labor
Allocation Among Food Crops
Labor input allocated by each member of the household to farm and nonfarm
economic activities was recorded four times a week during the one-year field survey.
A number of points regarding the allocation of labor among food crops in different
communities in the study area can be made based on these data (see Table 12).
In Turumbu and Mongandjo, where rice-growing has a long history, a large share
of labor time was allocated to rice during the 1982/83 crop year. Seven times more
labor was expended on rice than on maize in Mongandjo and twice as much in Turumbu.
Because rice was only recently introduced in Bomaneh, it claimed a relatively small
share of the labor allocated to farming, but its labor share was still much larger than
that of plantain. A substantial amount of labor time was spent on cassava in all three
communities. This allocation of labor to cassava, the main staple food crop in the area,
is revealing in terms of concerns about household food security.
Table 12 also shows substantial inter- and intracommunity variability in labor
allocation among crops, which has been ascribed earlier to differences in local market
conditions and the degree of access to major centers. For the study area as a whole,
rice, the major cash crop, claimed slightly more labor than cassava and two-and-a-half
times more labor than maize and plantain combined.
Allocation of Labor to Farming
First, regarding the allocation of labor to farming (Table 13), the total labor time
spent on economic activities per man-equivalent varied between 449 and 1,292 man-
hours per year, with an average of 776 man-hours for the study area as a whole. Since
it has already been established that as many as 1,800 man-hours a year can productively
be allocated to economic activities per man-equivalent in the humid tropics of Africa,
it can be said that the labor input allocated to economic activities in the area is rather
low, representing on average 43.1 percent of potential productive labor time. This
suggests that the area is still at a low level of economic development and that the
economic activities carried out are mainly for subsistence purposes.
Second, the data show that in most villages, farming is second to nonfarming,
except in the Turumbu villages along the Yangambi-Isangi road, where productive labor
time was almost equally allocated to farming and nonfarming. Overall, the share of
farming in total labor time spent on economic activities during the 1982/83 crop year
varied between 27.3 and 55.5 percent among villages, giving an average of about 34.9
percent for the area as a whole.








Table 12-Household labor allocations to rice, maize, cassava, and plantain

Labor
Number of Allocated
Rural Community Households to Farming" Rice Maize Cassava Plantain
(man-hours) (percent)
Turumbu 53 844 41.52 21.98 31.61 4.88
(469)
Yangambi-Kisangani roadb 17 712 27.29 29.60 35.06 8.05
(491)
WekoandYambaw 17 954 57.47 13.55 26.73 2.25
(512)
Yangambi-Isangiroadc 19 864 38.25 23.50 33.43 4.82
(401)
Mongandjo 53 954 49.30 7.36 43.14 0.20
(351)
Bolikango 25 1,086 45.84 7.69 46.29 0.18
(435)
Babenga 28 837 53.14 7.06 39.58 0.22
(198)
Bomenge
Bomaneh 26 519 18.36 25.62 54.87 1.15
(321)
Studyarea 132 824 42.52 15.09 40.40 1.99
(426)

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
Note: Figures in parentheses are standard deviations.
a The hours devoted to farming by each member of the household were converted into man-hour equivalents
using Collinson's conversion factors reported in Appendix 1.
b These villages were located up to 22 kilometers from Yangambi.
cThese villages were located up to 20 kilometers from Yangambi.



Third, during the 1982/83 crop year, an average of 271 man-hours per man-equivalent
were devoted to agriculture. This limited amount of labor input allocated to farming
should not be seen as a consequence of relying on a rainfed agricultural system, where
the distribution of rainfall imposes the pattern of labor use and allocation to farming.
The distribution of rainfall in the Zairian Basin allows for farming all year round. A
comparison with other developing areas indicates that the amount of labor input allo-
cated to farming per man-equivalent in the study area is small even by African standards.
In a review of the literature on labor use in Africa, Eicher and Baker (1982) wrote

Farm level surveys throughout Sub-Saharan Africa have consistently shown that farmers
have low annual labor inputs in agricultural production by international standards. Cleave
(1974) reviewed 50 micro-level studies in countries of both high and low man-land ratios
and found that male adults were working an average of 1,000 hours per year in agri-
cultural production as compared with 2,500 to 3,000 hours per year in Egypt and
many Asian countries. Some researchers have reported even fewer hours spent on
farming activities, as low as 500 to 600 hours per year (Haswell 1953; Norman 1972).

The survey clearly indicates that farming is not the leading economic activity in
the area. Hunting, fishing, and gathering are more important. They claimed about






Table 13-Allocation of household productive labor time between farming and nonfarm economic activities,
1982/83

Labor Labor
Average Allocated Allocated Labor Labor Allocated to
Number of to Economic to Economic Allocated Nonfarm Economic
Numberof Man-Equivalents/ Activities/ Activities/ to Farming/ Activities/
Rural Community Households Household Household Man-Equivalent Man-Equivalent Man-Equivalent
(man-hours) (man-hours) (percent) (man-hours) (percent)
Turumbu 53 3.88 2,279 587 217 36.97 370 63.03
(1.53) (1,239)
Yangambi-Kisanganiroada 17 4.24 2,593 612 168 27.45 444 72.55
(1.69) (1,326)
Weko and Yambaw 17 3.99 2,776 696 239 34.34 457 65.66
(1.49) (1,366)
Yangambi-Isangiroadb 19 3.46 1,552 449 249 55.46 200 44.54
(1.38) (576)
Mongandjo 53 2.52 2,747 1,090 379 34.77 711 65.23
(0.92) (1,250)
Bolikango 25 2.68 2,395 894 405 45.30 489 54.70
(1.05) (928)
Babendja 28 2.37 3,062 1,292 353 27.32 939 72.68
(0.77) (1,423)
Bomenga 26 2.38 1,734 729 218 29.90 511 70.10
(0.73) (716)
Bomaneh 26 2.38 1,734 729 218 29.90 511 70.10
(0.73) (716)
Studyarea 132 3.04 2,359 776 271 34.92 505 65.08
(1,212)

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut Facultaire des Sciences Agronomiques,
Yangambi, Zaire, and International Food Policy Research Institute, Washington, D.C., 1986 (computer printout).
Notes: Household members were converted into man-equivalents to take account of age and sex using conversion factors derived from Collinson (1972) (see
Appendix 1). Economic activities include farming and nonfarming activities (namely hunting, fishing, and gathering). Figures in parentheses are standard
deviations.
a These villages were located up to 22 kilometers from Yangambi.
b These villages wree located up to 20 kilometers from Yangambi.







65.08 percent of the labor time allocated to economic activities during the 1982/83
year. A survey conducted in 1974/75 on a small sample of 17 households in Yalibwa,
Turumbu Community, revealed that the contribution of hunting, fishing, and gathering
activities to household cash income was larger than that of farming. These activities
contributed about 54 percent of cash income as opposed to 36 percent from farm
products. Handicraft and alcohol production shared 2 percent and 6 percent respectively
(Kilumba 1975). Although it can generally be observed that the overall time allocated
to economic activities in the study area is low, an assessment of key determinants of
labor allocation between activities carried out in the household is beyond the scope of
this study, which focuses on farming. A detailed appraisal of household labor use would
warrant an entire study by itself.

Determinants of Labor Allocation to Farming
Having established that labor input per man-equivalent is low, a model is devised to
identify key factors that affect the allocation of labor to farming in order to assess the
implications for labor productivity and to examine policy actions that the government of
Zaire should initiate to address this issue. From this exercise, implications for other key
parameters such as cultivated land area, farm output, and resource productivity are derived.
Model
As stated earlier, the rural small-farm economy includes farming and nonfarming
activities. These two groups of activities compete for the labor time of the household.
One can then assume that the amount of labor time allocated to farming per man-equivalent
is a function of the domestic terms of trade between farming and nonfarming activities.
The amount of capital input available per man-equivalent and the degree of dependency
also affect the amount of labor time allocated to farming per man-equivalent.9 This
formulation also should include a dummy variable for locational and other differences
among study communities and an error term.
Formally, one can express this relationship as follows:

L/W = f(PI/P2, K/W, C/W, D, e), (1)
where
L/W = the amount of labor time in man-hours allocated
to farming per man-equivalent,
P,/P2 = farmgate terms of trade between household
agricultural and nonagricultural activities dur-
ing the preplanting period,
K/W = capital input (in zaires) allocated to farming per
man-equivalent,
C/W = degree of dependency expressed as a ratio of
the number of household consuming units over



9 This chapter deals with resource allocation (of labor) within the small-farm economy. Analysis of the
resource flow between smallholder and other components of the national economy, although important,
is beyond the scope of the present effort.







the number of man-equivalents,10
D = dummy for location, and
e = the error term.
It is expected that
[8(L/W)/8(P1/P2)1 0 ; [8(L/W)/8(K/W)] 0; and [8(L/W)/8(C/W)] 0. (2)
This implies that a positive change in the farmgate terms of trade between farm
and nonfarm activities, capital input per man-equivalent, and degree of dependency
will be associated with a positive change in the amount of labor time allocated to
farming per man-equivalent.

Definition of Explanatory Variables
The main ingredients of the expression depicting the amount of labor input allocated
to agriculture are capital input and the farmgate price of farm relative to nonfarm
commodities produced in the households, and their degree of dependency. In this
study, the preplanting farmgate prices for rice, cassava, maize, plantain, and nonfarm
products (such as game meat and fish) are taken to be the relevant producer prices
that affect labor allocation between household agricultural and nonagricultural activities.
For each sampled household, the average preplanting price for the above crops is
computed, using each crop's share of labor time allocated to farming as a weight. The
price for game meat was considered for Turumbu and Mongandjo Communities, whereas
in Bomaneh (Bomenge Community), the price for fish was taken. In the first two
communities, hunting is the most important household nonfarm activity, whereas in
the last community, fishing is. Appendix 2, Table 31, summarizes the price data.
In this study, the degree of dependency is defined as a ratio of the number of
household consuming units over the number of household man-equivalents in order
to relate the household food requirements to the size of the household labor force.
Regression Results
A least squares method applied to the survey data gives the regression results
summarized in Table 14.
For labor allocated to farming, the regression equation indicates that the fit is fair,
and the F-statistic is highly significant at the 1 percent level. The coefficients for both
the preplanting farmgate terms of trade between farm and nonfarm activities (PI/P2)
and the amount of capital input per man-equivalent (K/W) are positive and significantly
different from zero, implying that these variables are key determinants of the amount
of labor time allocated to farming per man-equivalent (L/W) in the Zairian Basin. The
degree of dependency (C/W), introduced in the analysis on the assumption that large
household size would also generate pressure to produce more foodcrops, does not
explain the amount of labor time allocated to farming. Locational differences among
study communities on labor allocation to agriculture are significant.


10 One consuming unit represents the food consumption of one adult male. The number of household
consuming units expresses the household food requirements, whereas the number of man-equivalents,
sometimes called working units, represents the available household labor force. The conversion factors
proposed by Collinson (1972) to transform household members into consuming units are reported in
Appendix 1.








Table 14-Determinants of labor allocated to farming, 1982/83


ExplanatoryVariables
and Important Statistics


Intercept

Ln (PI/P2)

Ln (K/W)

Ln (C/W)


Dummies for location


Dependent Variable
Ln(L/W)


5.62
(12.42)**
0.38
(2.70)**
0.28
(4.58)**
0.12
(0.66)

-0.39
(-2.35)*
-0.35
(-1.72)
-0.38
(-1.44)
-0.26
(-1.36)
-1.06**
(-4.42)
0.45
14.30


Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
Note: The numbers in parentheses are t-values. Ln(P1/P2) is the terms of trade between agricultural and
nonagricultural activities, Ln(K/W) is capital input allocated to farming per man-equivalent, and Ln(C/W)
is the degree of dependency.
*Significant at the 0.05 level.
**Significant at the 0.01 level.



The elasticity estimate of agricultural labor time per man-equivalent with respect to
the preplanting farmgate terms of trade is 0.38 and that with respect to capital input per
man-equivalent is 0.28. This suggests that a 1.00 percent increase in the preplanting farm-
gate terms of trade between farming and nonfarming activities will be associated with
an increase of about 0.38 percent in labor time allocated to farming per man-equivalent,
whereas a 1.00 percent increase in capital input available per man-equivalent will lead
to about a 0.28 percent increase in labor time allocated to agriculture per man-equivalent.
These findings suggest that appropriate policy actions to improve both the domestic
terms of trade of farm products and the content of capital input (tools, equipment,
seeds, fertilizers, and so on) for smallholders need to be considered if farmers are to
divert more of their labor time to farming. These policies include infrastructure devel-
opment and technology policies.
Given the paramount importance of labor in Zaire, an in-depth analysis of this issue
is fundamental to understanding how the small-farm sector works and its level of
performance. In addition, an assessment of factors that determine the amount of labor
allocated to farming will contribute to understanding at least indirectly the workings







of the nonagricultural sector of the rural economy. This alone would clarify the inter-
action of the two sectors and help the government to formulate appropriate policies.
Although the foregoing section of this chapter has identified some key variables affecting
the allocation of labor to farming, a detailed appraisal of labor use in the rural economy,
with all its related aspects and implications, deserves an entire study, which should
logically succeed this piece of work.








6


RELATIONSHIPS BETWEEN LABOR, CAPITAL,
AND CULTIVATED AREA

This chapter estimates the intensities of labor and capital use, identifies their key
determinants, establishes the relationship between cultivated area, labor, and capital,
and finally relates the above to the observed pattern of labor allocation to farming in
the study area. The estimates of the structural parameters derived from this analysis
are relevant for policymaking.

Intensity of Labor Use
The estimates of the intensity of labor use by crop reported in Table 15 show that
rice is the most labor-consuming crop in the production mix. It is followed by cassava
and then maize, while plantain is the most labor saving. Taking the labor per hectare
allocated to plantain as a numeraire, one can see that rice requires about 2.95 times
more labor than plantain, cassava about 2.21 times, and maize about 1.75 times. For
all the crops combined, an average labor input of about 1,230 man-hours per hectare
was recorded during the 1982/83 crop year, with a range of 525 man-hours per hectare
for plantain and 1,547 man-hours for rice.
This intensity of labor use is high compared with other parts of the developing
world, and this becomes even more disturbing in an area characterized by low population
density like the Zairian Basin. For example, Lassiter (1982) reported 822 man-hours
per hectare in the production of sorghum in Eastern Burkina Faso, and in the Mumbwa
area of Zambia, farmers spent 568 man-hours per hectare in the production of maize
(Dodge 1977). In Maharashtra State, India, 477 man-hours per hectare were allocated
to traditional varieties of sorghum, whereas in Bangladesh, the production of local
varieties of aman rice claimed about 1,067 man-hours per hectare (Mellor and Ranade
1986; Hossain 1988).
This high labor intensity is partly explained by high labor requirements for land
preparation in the forest zone. This situation is further exacerbated by the use of hand
tools and by the hot, humid tropical weather conditions, which increase the difficulty
of farming. To these environmental constraints, the amount of labor time allocated to
land preparation (slashing, burning, and clearing), which determines the size of culti-
vated area, seems to be an additional factor explaining high overall labor use per
hectare. This is understandable in the sense that if a small amount of labor input is
devoted to land preparation, the cultivated area will also be small. Therefore, to produce
the highest possible output, relatively more labor input has to be spent on subsequent
farm operations, including planting, weeding, harvesting, and farm-level processing.
The overall labor use per hectare becomes much higher than it would be otherwise.
Data relating household labor and the size of cultivated area are summarized in
Table 16, showing that during 1982/83, about 81.1 percent of the sampled households
had a cultivated area that did not exceed 1.0 hectare per household. The rest of the
households in the sample, 18.9 percent, had cultivated area greater than 1.00 hectare
per household. For the sample as a whole, the average amount of cultivated area per
household was 0.67 hectare with a minimum area of 0.05 hectare and a maximum








Table 15-Intensity of labor use, by crop, 1982/83
Average Average
Number of Area/ Labor Input/
Crop Households Household Household Labor Intensity
(hectare) (man-hours) (man-hours/hectare)
Rice 80 0.36 557 1,547
Maize 82 0.19 175 921
Cassava 128 0.29 336 1,159
Plantain 52 0.08 42 525
All crops 132 0.67 824 1,230

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).


area of 2.99 hectares per household. If farm households are to be described by the
size of cultivated land area, one would say, like Sen (1964), that the study area is made
up of very small and small farms. As will be shown later in this chapter, this qualification
of farm households based on cultivated area is critical in explaining the patterns of
labor and capital use per hectare across farm households.
It is also important to note that the size of cultivated area per household varies
enormously relative to the number of man-equivalents per household. The minimum
and maximum numbers of man-equivalents per household recorded during the 1982/83
survey were 1.17 and 8.01 respectively with an average of 3.03 man-equivalents per
household for the study area as a whole. In addition, the data show that the participation
of household members in farming increased with the cultivated land area, but the
variation is small relative to variation in cultivated area. This explains the observed
pattern of declining labor input per hectare with increasing cultivated land area. Norman
(1973b) in his study of farming among the Hausa in northern Nigeria found the same
relationship but failed to explore the reasons behind this phenomenon in the Nigerian
context. This inverse relationship between per hectare labor input and size of cultivated
area has been extensively documented in many Asian and Latin American countries
(Booth and Sundhrum 1985).
A close look at Table 16 reveals that the intensity of labor input allocated to land
preparation exhibits an inverse relationship to the size of cultivated area. This finding
is not anticipated: one would expect the per hectare labor input devoted to land
preparation not to be affected by size of cultivated area. One explanation could be a
deterioration in the quality of land preparation as larger areas are cultivated.
In addition to the size of cultivated land area, the intensities of labor allocated to
land preparation and to subsequent farm operations are also expected to be affected
by the length of the fallow period and by the type of crop mix adopted. To capture the
effect of the length of the fallow period on the amount of labor time per hectare
allocated to land preparation, a dummy variable with a value of one is assigned for a
fallow period equal to or greater than seven years; otherwise, zero is assigned. The
use of seven years as a benchmark is supported by agricultural research indicating that
a minimum fallow period of seven years is necessary to maintain soil fertility in the
humid tropics under a traditional farming system (Greenland and Okigbo 1983; Jurion
and Henry 1967).
To account for the effects of the crop mix on the intensity of labor input allocated
to subsequent farm operations, the share of cassava in cultivated land area is introduced








Table 16-Cultivated area and intensity of labor inputs, 1982/83

Labor Input Labor Input Labor Input
Average forAll Farm for Land Other Farm Labor Intensity
Number of Household Area per Operations/ Preparation/ Operations/ All Farm Land Other Farm
Cultivated Area Households Size Household Household Household Household Operations Preparation Operations
(hectares) (percent) (man-equivalents) (hectares) (man-hours) (man-hours/hectare)
0.25orless 12.88 2.83 0.19 442.72 181.71 261.01 2,330.11 956.37 1,373.74
0.26-0.50 24.24 3.01 0.38 617.87 279.08 338.79 1,625.97 734.42 891.55
0.51-0.75 29.55 2.97 0.62 805.23 302.07 503.16 1,298.76 487.21 811.55
0.76-1.00 14.39 3.01 0.86 996.50 407.48 589.02 1,158.72 473.81 684.91
Morethan 1.00 18.94 3.35 1.32 1,247.57 403.85 843.71 945.12 305.95 639.17
Studyarea 100.00 3.04 0.67 824.42 315.44 508.98 1,230.48 470.81 759.67

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut Facultaire des Sciences Agronomiques,
Yangambi, Zaire, and International Food Policy Research Institute, Washington, D.C., 1986 (computer printout).








into the equation. The use of cassava's area share is supported by the fact that small
farming in the study area is a cassava-based system, as indicated by the number of
households producing this crop. Locational and other differences among study com-
munities are taken into account through a dummy variable. And finally, the intensity
of the overall labor input allocated to farming is also related to all these variables.
The regression results depicting these relationships are summarized in Table 17.
The analysis shows that locational differences are significant. The regression coefficients
for the dummy representing the fallow period display the expected sign, but are not
significantly different from zero. This implies that the length of the fallow period does
not significantly affect the intensity of labor input allocated to land preparation or to
farming as a whole.
The type of crop mix does not explain the intensity of labor input allocated to all
the farm operations combined, but it is an important determinant of the intensity of
labor input allocated to planting, weeding, harvesting, and farm-level processing. Be-
cause the share of cassava is used to express the type of crop mix, this finding indicates
that the intensity of labor spent on these farm operations declines as the share of
cassava in the cultivated area increases. Specifically, a 1.00 percent increase in the
share of cassava leads to a decline of about 0.29 percent in the amount of time per
hectare spent on planting, weeding, harvesting, and on-farm processing.
As expected, the coefficient for cultivated area is negative and highly significant in
all cases. The elasticities of labor input allocated to land preparation per hectare and
to other farm operations with respect to cultivated area are -0.58 and -0.26, respec-
tively, while the estimate of the elasticity of the per hectare labor input allocated to
all farm operations combined with respect to the same explanatory variable amounts
to -0.42. This implies that a 1.00 percent increase in cultivated area leads to a 0.42
percent decline in intensity of labor input devoted to farming. These findings confirm
what has been observed elsewhere in Asia and Latin America.


Intensity of Capital Use
The analysis of capital input data by crops indicates that rice is the most capital-
consuming crop in the survey area. By taking the intensity of capital input use for
plantain as a numeraire, it can be seen that I hectare of rice uses 6.29 times the capital
input of plantain, whereas maize uses 3.48 times and cassava 3.42 times more inputs
than plantain (Table 18).
The analysis of the same capital input data in relation to the size of cultivated area
in Table 19 indicates that the amount of capital (tools and seeds combined) per hectare
declines as the size of cultivated area increases. The disaggregation of capital input
into tools and seeds gives a different picture. The amount of tools per hectare allocated
to farming continues to exhibit an inverse relationship with the size of cultivated area,
whereas the amount of seed shows a positive relationship.
In addition to the size of cultivated area, the intensity of capital input is also
determined by the type of crop mix adopted. Locational and other differences among
study communities are also expected to affect the intensity of capital use. These factors
are incorporated in the analysis reported in Table 20. The regression coefficient for
the type of crop mix is negative and significantly different from zero at the 1.00 percent
level except for the intensity of tool use. The elasticities for the intensity of capital
input (tools and seeds) and input of seeds with respect to the type of crop mix are
about -0.50 and -1.65, which implies that a 1.00 percent increase in the share of









Table 17-Determinants of the intensity of labor input

Intensity Intensity
Intensity of Labor of Labor
ofLabor Allocated Allocatedto
Explanatory Varables and Allocated to Land Other Farm
Important Statistics to Farming Preparation Operations
Ln (man-hours/hectare)
Intercept 6.81 5.35 6.00
(17.17)** (51.01)** (38.00)**
Ln cultivated area -0.42 -0.58 -0.26
(-7.04)** (-9.14)** (-3.37)**
Ln type of crop mix -0.07 ... -0.29
(-0.67) (-2.13)*
Dummy for fallow period 0.04 0.04
(0.51) (0.48)
Dummies for location
DI 0.09 0.65 -0.64
(0.74) (4.63)** (-3.88)**
D2 0.56 0.74 0.23
(3.46)** (5.02)** (1.15)
D3 0.40 0.48 0.34
(3.36)** (3.33)** (2.29)*
D4 0.57 0.79 0.46
(4.93)** (5.64)** (3.30)**
D5 0.34 0.38 0.32
(3.17)** (2.96)** (2.35)*
R2 0.42 0.48 0.33
F 12.84 18.43 10.24
n 132 132 132

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
Note: t-values are in parentheses.
*Significant at the 0.05 level.
**Significant at the 0.01 level.


Table 18-Intensity of capital input use by crop

Average
Number of Average Area/ Capital Input/ Intensity of
Crop Households Household Household Capital Use
(hectare) (V)
Rice 80 0.36 77.61 215.58
Maize 82 0.19 22.65 119.21
Cassava 128 0.29 33.97 117.14
Plantain 52 0.08 2.74 34.25
All crops 132 0.67 100.74 150.36

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
Note: The capital inputs here are seeds and tools combined.






Table 19-Cultivated area and intensity of capital input

Capital Input/Household Capital Intensity
Average Capital Input Capital Input
Number of Household Area/ (Tools and (Tools and
Cultivated Area Households Size Household Seeds) Tools Seeds Seeds Tools Seeds
(hectares) (percent) (man- (hectares) (Z) (Z/hectare)
equivalents)
0.25orless 12.88 2.83 0.19 52.70 49.38 3.32 277.36 259.89 17.47
0.25-0.50 24.24 3.01 0.38 68.44 58.16 10.27 180.08 153.05 27.03
0.51-0.75 29.55 2.97 0.62 104.42 82.71 21.71 168.41 133.40 35.01
0.76-1.00 14.39 3.01 0.86 95.50 70.69 24.81 111.05 82.20 28.85
More than 1.00 18.94 3.35 1.32 173.03 107.35 65.58 131.09 81.33 49.76
Studyarea 100.00 3.04 0.67 100.74 75.40 25.34 150.36 37.82 112.54

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut Facultaire des Sciences Agronomiques,
Yangambi, Zaire, and International Food Policy Research Institute, Washington, D.C., 1986 (computer printout).








Table 20-Regression results for determinants of the intensity of capital input

Intensityof
Capital Input Intensity
Explanatory Variables and Allocated to ofToolUse Intensity
Important Statistics Farming for Farming of Seed Use
Ln (-/hectare)
Intercept 4.61 5.26 2.24
(23.11)** (6.41)** (14.18)**
Ln cultivated area -0.42 -0.64 0.30
(-4.32)** (-5.26)** (3.82)**
Ln type of crop mix -0.50 -0.17 -1.65
(-2.94)** (-0.82) (-11.80)**
Dummies for location
DI -1.22 -1.34 -1.36
(-5.89) ** (-5.20)** (-7.86)**
D2 -1.19 -1.06 -1.50
(-4.78)** (-3.39)** (-7.79)**
D3 -0.73 -0.81 -1.07
(-3.76)** (-3.38)** (-6.90)**
D4 -0.06 -0.20 0.61
(-0.35) (-0.93) (4.55)**
Ds 0.13 -0.20 0.24
(0.75) (-0.94) (1.86)
R2 0.35 0.39 0.67
F 11.08 12.56 36.95
n 132 132 123

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
Note: t-values are in parentheses.
*Significant at the 0.05 level.
**Significant at the 0.01 level.



cassava in the cultivated area leads to a 0.50 percent decline in the intensity of total
capital input and a 1.65 percent decline in the amount of seeds used per hectare.
The coefficient for cultivated area is negative and highly significant in equations
where total capital input per hectare and the cost of tools per hectare are used as
dependent variables, but positive and highly significant in the equation where the cost
of seeds sown per hectare is the dependent variable. The estimates of the elasticities
of the intensity of capital input with respect to cultivated area are -0.42 for tools and
seeds combined, -0.64 for tools alone, and 0.30 for seeds alone.
These results call for some explanation. Given that the share of tools in capital
input is very large (74.9 percent on average for the study area as a whole) and given
that these tools are hand held, and thus highly correlated to labor input, the intensity
of capital input (tools and seeds combined) would be expected to decline as the size
of cultivated area increased, as was the case for intensity of labor input. The observed
positive relationship between the intensity of seeds sown and the size of cultivated
area, however, suggests that households cultivating a relatively large area of land allocate
relatively more land to cash crops (rice and maize) than others. In other words, farmers
move from a subsistence to a market-oriented crop mix as the size of cultivated area








increases. A summary of relevant survey data supports this conclusion (Table 21). The
share of cultivated area allocated to the cash crops, rice and maize, increases with the
size of cultivated land area, whereas the subsistence crops, cassava and plantain, exhibit
a declining trend.
An in-depth analysis of this relationship is summarized in Table 22. It shows that
the share of cultivated area allocated to cash crops is linearly related not only to the
size of cultivated land area, but also to the domestic terms of trade between cash and
subsistence crops. The size of cultivated area, as well as the farmgate price of cash
crops relative to subsistence crops, are critical variables explaining the change in the
share of cultivated area allocated to cash crops by small farmers in the Zairian Basin.
A 1.00 percent increase in the size of cultivated area is associated with a 0.20
percent increase in the share of cultivated area devoted to cash crops, while a 1.00
percent increase in the domestic terms of trade between cash and subsistence crops
leads to a 0.12 percent increase in the share of cultivated area allocated to cash crops.

Relationship Between Cultivated Area,
Labor, and Capital
The foregoing sections have shown that expansion of cultivated area would reduce
the intensities of both labor and capital use, which in turn would help increase the
productivity of labor and capital inputs allocated to farming. It should also be stressed
that in a situation where land is abundant relative to other resources and yield-increasing
technologies are unknown, expansion of cultivated area is the main source of food and
agricultural output growth. This depicts most parts of Sub-Saharan Africa and the Zairian
Basin in particular, in sharp contrast with many developing areas of Asia and Latin
America (Hayami and Ruttan 1985; Paulino 1986). The economic history of Zaire since
1960 suggests that the technological environment facing the small-farm sector is not
going to improve in the foreseeable future unless some drastic changes in policy
direction are initiated (Tshibaka 1986): therefore, expansion of cultivated area is likely
to remain the key determinant of production growth.




Table 21-Cultivated area and change in crop mix
Share of Area
Average Area/ AreaAllocated Allocated
Number of Household House- Cash Subsistence Cash Subsistence
Cultivated Area Households Size hold Crops Crops Crops Crops
(hectares) (percent) (man- (hec- (hectare) (percent)
equivalents) tares)
0.25 orless 12.88 2.83 0.19 0.07 0.12 37 63
0.26-0.50 24.24 3.01 0.38 0.19 0.19 50 50
0.51-0.75 29.55 2.97 0.62 0.32 0.30 52 48
0.76-1.00 14.39 3.01 0.86 0.41 0.45 48 52
More than 1.00 18.94 3.35 1.32 0.79 0.53 60 40
Studyarea 100.00 3.04 0.67 0.36 0.31 54 46

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).









Table 22-Regression results for cultivated area, farmgate prices, and share
of cash-crop area


Explanatory Variables and
Important Statistics


Intercept

Cultivated area
Farmgate price of cash crops
relative to subsistence crops
Dummies for location
Di


DependentVariable
Cash-Crop Area Share


0.42
(6.95)**
0.15
(4.51)**
0.01
(5.05)**

-0.17
(-2.48)*
-0.11
(-1.66)
-0.12
(-1.92)
-0.04
(-0.69)


16.87
132


Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
Note: t-values are in parentheses.
*Significant at the 0.05 level.
**Significant at the 0.01 level.





Because land is abundant and freely accessible to each household in the study area,
and given the current state of technology in the area, the expansion of cultivated area
depends on the amount of labor time allocated to land preparation and on the amount
of capital tools available. This can be expressed as follows:


H = f(Lp, Kt, D, e),

H = the cultivated area in hectares,

L = the labor time allocated to land
preparation in man-hours,

Kt = capital tools in zaires,

D = a dummy to account for loca-
tional and other differences
among study villages, and

e = an error term.


where








The regression analysis was performed using Cobb-Douglas and transcendental
functions. Due to serious multicollinearity between labor time allocated to land prep-
aration and capital tools, the original data were transformed assuming a Cobb-Douglas
function with constant returns to scale (Koutsoyiannis 1984). An F-test to compare the
Cobb-Douglas and transcendental functional forms, derived after this transformation,
suggests that the Cobb-Douglas functional form is appropriate. Table 23 displays these
regression results. The fit is good and the F-statistic is significant at the 1.0 percent
level. Locational differences are, overall, negligible. The coefficient for the ratio of labor
to capital is positive and significantly different from zero at the 1.0 percent level.
The estimates of the elasticities of cultivated area with respect to labor time allocated
to land preparation and to capital tools are 0.75 and 0.25, respectively. These elasticity
estimates suggest that labor time devoted to land preparation is the most important
determinant of the size of cultivated area. But one should note that the effects of capital
tools used is also significant. More specifically, a 1.00 percent increase in labor time
and in capital tools allocated to land preparation will be associated with a 0.75 percent
and 0.25 percent increase in cultivated area respectively.
The implications for the intensities of both labor and capital use are obvious. A
1.00 percent increase in labor allocated to land preparation will ultimately lead to a
0.32 percent decline in the intensity of labor input allocated to farming."
Finally, to reduce the intensity of labor and capital use, the size of cultivated area
has to be increased through greater labor time devoted to land preparation. This change
in labor use and allocation will have substantial implications for the productivities of
both labor and capital, which will be discussed in the next chapter. Moreover, increasing
the size of cultivated area assumes that crops to be grown on this incremental area
are those that can be easily marketed. The access to market, which reflects the level
of infrastructural development, becomes critical. This underscores the need to improve
the infrastructure in the area.



















'' The intensities of total agricultural labor (L) and capital (K) are related to the amount of labor time
allocated to land preparation as follows: w = -0~J and p = pJ, where w and p stand for the elasticities of
intensities of labor and capital inputs with respect to labor time allocated to land preparation; -0 and -p
stand for the elasticities of the intensities of labor and capital inputs with respect to the size of cultivated
area; and finally, j$ is the elasticity of cultivated area with respect to labor time allocated to land preparation.









Table 23-Regression results for the relationship between cultivated area,
labor, and input of capital (tools)


Cultivated Area/
Capital Input
(Tools)


Explanatory Variables
and Important Statistics


Intercept

Ln total labor time to land
preparation/tools

Dummies for location
DI


Ln (hectare/Z)


-5.66
(-50.44)**

0.75
(13.60)**

-0.02
(-0.13)
-0.37
(-1.89)
-0.07
(-0.38)
-0.31
(-1.96)*
-0.07
(-0.45)
0.69
49.16


Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
Note: t-values are in parentheses.
*Significant at the 0.05 level.
*Significant at the 0.01 level.








7


FARM OUTPUT AND RESOURCE PRODUCTIVITIES
IN THE STUDY AREA


Farm Output
Production Function
Under given technological conditions, farm output is viewed as a function of culti-
vated area, labor, and capital, as well as other variables such as weather. When a
cross-section study is carried out in a fairly homogeneous zone, these "other" variables
are regarded as given, and the farm output mainly becomes a function of cultivated
area, labor, and capital.
This formulation is theoretically adequate, but since land itself is not a constraint
to food and agricultural production in the Zairian Basin, the volume of farm output can
be regarded as mainly determined by labor and capital inputs. In addition, because
fallowing-a natural way of regenerating soil fertility-is the most important feature
of the farming system in the area, the effect of the length of fallow on the level of farm
output is also examined through a dummy that assumes a value of one for fallows of
seven or more years, and zero for those of less than seven years (Greenland and Okigbo
1983). Furthermore, the effects of factors such as locational differences among study
communities on the level of farm output are also taken into account through a dummy
variable with a value of one for the community of concern and zero otherwise. Formally,
this relationship can be expressed as

Q = f(L, K, Df, DI, e), (4)
where
Q = the crop output in kilograms,
L = the labor input in man-hours,
K = the capital input in zaires,
Df = the dummy for the length of
the fallow period,
D1 = the dummy for locational dif-
ferences, and
e = the error term.

An important policy variable that is included with this relationship is the period
when a new crop-rice-was introduced. Rice was first introduced in Turumbu and
Mongandjo during the period 1840-1911, but it was not introduced in Bomaneh until
1981/82. A dummy variable to account for the history of rice in the study area is
included in the rice production function, with a value of one assumed for Turumbu
and Mongandjo communities and zero for Bomaneh.








For all crops combined, the aggregate output production function includes the type
of crop mix adopted as an additional shift variable. As indicated earlier, the share of
the subsistence crop, cassava, in cultivated area also expresses this variable in the
production function analysis.
Regression Results
Input/output data from the 1982/83 survey were regressed against the model
developed above. Cobb-Douglas and transcendental functions were estimated. Due to
serious multicollinearity between labor and capital inputs, the original data were trans-
formed assuming a constant returns to scale Cobb-Douglas production function. An
F-test comparing the Cobb-Douglas and transcendental forms showed the Cobb-Douglas
to be a more appropriate transformation (Koutsoyiannis 1984).
For crops other than rice, the transformation in which the average product of capital
(Q/K) was regressed against the labor/capital ratio (LK) produced the most robust
regression results. For rice, however, the transformation having the average product
of labor (Q/L) and capital/labor ratio (K/L) as the main argument turned out to be
best. The econometric reason for this is not immediately obvious. The most robust
equation for either crop is given in Table 24. It is striking to see that in either case,
however, the estimates of output elasticities remain the same (see Appendix 2, Table 32).
The analysis shows that capital relative to labor is the most important factor affecting
the production of rice in the area, while labor input has a stronger impact on maize
than capital. For cassava and plantain, labor is the overriding input. For all crops
combined, the analysis suggests that labor is the main determinant of the farm output
level. The labor share of the aggregate crop output is about 78 percent, whereas that
of capital input amounts to about 22 percent. When all is said and done, however, it
should be kept in mind that, even in this predominantly subsistence sector, the effects
of capital cannot be ignored, although they are small relative to those of labor.
The analysis also indicates that the type of crop mix adopted is an important shift
parameter of the overall household crop output, but locational differences among study
communities cannot be ignored, and the length of the fallow period does not seem to
affect the level of farm output. Finally, the date when an innovation-rice-is introduced
in a community is an important variable. The analysis of rice input/output data indirectly
suggests that delaying the introduction of improved inputs and techniques in a given
area could seriously hinder agriculture in that area. Time plays a critical role in the
process of technological change.

Resource Productivities
Average Products of Inputs
At the outset, one should note that the farm output per hectare (all crops combined),
achieved under a cassava-based, mixed-crop, traditional farming system in the Zairian
Basin, is quite substantial (Table 25).
The average products of labor and capital (expressed in cereal equivalents) and
their values are higher in the production of cassava and plantain than in maize and
rice, suggesting that cassava and plantain are more productive and profitable than maize
and rice in the area. Furthermore, it is important to observe that the productivities of
land, labor, and capital vary greatly across commodities within this relatively homoge-
neous area. This result is confirmed by a number of other studies conducted elsewhere
in Sub-Saharan Africa. In the rain forest of C6te d'Ivoire, den Tuinder (1978) found
that the returns to labor across major crops varied by a factor of one to five, and the









Table 24-Regression results for food crop production functions

Explanatory Crop/DependentVariable
Variables and Rice Maize Cassava Plantain AllCrops
Important Statistics LnQ/L LnQ/K LnQ/K LnQ/K LnQ/K


Intercept


LnL/K

LnK/L


Dr

Df

S

Dummies for location
DI


Elasticities with
respect to
L
K


-1.06
(-2.14)*


0.87
(5.38)**
2.14
(5.06)**
0.21
(0.99)



0.08
(0.18)
1.09
(3.39)**
0.58
(1.64)
0.01
(0.05)


0.41
8.81
80


0.13
0.87


0.60
(1.96)
0.63
(4.42)**




-0.22
(-1.03)



0.73
(2.13)*
-0.16
(-0.41)
-0.07
(-0.17)
-0.52
(-1.55)
-0.09
(-0.25)
0.45
8.71
82


0.63
0.37


2.72 2.92
(17.63)** (12.13)**
0.98 0.85
(16.11)** (9.18)**


0.04
(0.34)



-0.58
(-2.96)**
-1.17
(-5.60)**
-0.24
(-1.22)
0.65
(3.60)**
0.61
(3.91)**
0.75
54.43
128


0.98
0.02


0.12
(0.61)



0.41
(1.58)
0.27
(0.66)
-0.10
(-0.38)





0.74
29.17
52


0.85
0.15


Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
Notes: Dr is the dummy for the date of introduction of rice in the study communities, Df the dummy for the
length of the fallow period, and S the cassava share of the cultivated area to represent the type of crop
mix. Output for all crops combined is expressed in kilograms of cereal-equivalents using the following
conversion factors: 1.00 for maize-grains, 0.60 for paddy rice, 0.303 for cassava, and 0.220 for plantain;
t-values are in parentheses. Because rice, compared with other crops, requires relatively more capital
than labor, the transformation of data was performed using labor as a deflator (L). For other crops and
all crops combined, which happen to require more labor than capital compared with rice, the transformation
was made using capital as a deflator (K). The use of capital as a deflator in the rice equation and labor
in other crop equations give the regression coefficients for the labor/capital ratio (Ln LK) in the rice
equation and capital/labor ratios (Ln K/L) in all other equations that are not statistically different from
zero except in the case of maize. All the estimates of the critical parameters, for example, the elasticities,
are the same regardless of the deflator used (see Appendix 2, Table 32).
*Significant at the 0.05 level.
**Significant at the 0.01 level.


0.93
(4.31)**
0.78
(11.24)**





0.05
(0.44)
0.83
(2.57)**

0.11
(0.54)
-0.78
(-3.34)**
-0.25
(-1.31)
0.53
(3.06)
0.34
(2.22)*
0.63
28.14
132


0.78
0.22








Table 25-Resource use and productivities in the study area

Item Rice Maize Cassava Plantain All Crops

Resource use and output
Distribution of crops among households 80 82 128 52 132
Cultivated area (hectare) 0.36 0.19 0.29 0.08 0.67
Labor (man-hours) 557 175 336 42 824
Capital(Z) 77.61 22.65 33.97 2.74 100.74
Output (kilograms of cereal-equivalents) 324.66 213.07 2,225.14 133.20 2,539.31
Share of each crop in total output (percent) 7.75 5.21 84.97 2.07 100.00
Average products of resources and their values
Crop yield (kilograms of cereal-equivalents/
hectare) 901.83 1,121.42 7,672.90 1,665.00 3,790.00
Valueofcropyield(Z/hectare) 5,576.32 11,875.84 24,563.41 9,762.95 11,714.55
Average product of labor (kilograms of
cereal-equivalents/man-hour) 0.58 1.22 6.62 3.17 3.08
Valueofaverageproductoflabor (/man-hour) 3.60 12.92 21.19 18.58 9.52
Average product of capital (kilograms of
cereal-equivalents/Z) 4.18 9.41 65.50 48.61 25.21
Valueofaverageproductofcapital (Za'res/Z1) 25.85 99.65 209.69 285.03 98.30
Marginal products of resources and their values
Marginal product o flabor (kilograms of
cereal-equivalents/man-hour) 0.08 0.77 6.49 2.69 2.40
Value of marginal product of labor (Z/man-
hour) 0.50 8.15 20.76 15.76 9.74
Marginal product of capital (kilograms of
cereal-equivalents/Z) 3.64 3.48 1.31 7.29 5.55
Value of marginal product of capital (Zaires/Z) 22.49 36.85 4.19 42.72 35.57

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
Notes: The outputs of individual crops are converted into cereal-equivalents using the following conversion
factors: 1.00 for maize-grains, 0.60 for paddy rice, 0.303 for cassava, and 0.220 for plantain. The estimates
for individual crops and for all the crops combined are based on weighted sample sizes. In addition, since
most maize output in the study area is used for the production of a local beverage, the value of average
product of maize in production of the drink, which is about Z10.59 per kilogram of maize, is used as
the "implicit" price of maize. The average product of maize in the production of local beverage as about
1.05 liter per kilogram of maize, and the average price of one liter of this drink was 210.09 during the
survey. Farmgate prices used to compute values of marginal and average products of resources are reported
in Appendix 2, Table 31. The marginal products of resources are computed by multiplying the average
products of resources with the elasticities of output.



returns to land even more. The great variability in resource productivity across areas
and within areas among crops in the Sub-Saharan region could be ascribed to differences
in soil types, ecology, agricultural practices, biological nature of individual crops, physical
and human capital, and so forth.
Labor productivity in the Zairian Basin is lower than in other parts of Sub-Saharan
Africa, particularly in the production of cereals (Mellor, Delgado, and Blackie 1987;
Mellor and Ranade 1986; Dodge 1977; Massell and Johnson 1968). This study attempts
to answer the basic question of whether the observed level of labor productivity can
be increased given the current state of technology and resource endowment in the area.

Increasing Labor Productivity
Smallholder agriculture in the Zairian Basin has four basic characteristics: low labor
input per man-equivalent, a low capital-labor ratio, a high labor share of output, and
a low average product of labor. This suggests that the small-farm sector in the Zairian
Basin is still at a low stage of development (Hayami and Ruttan 1985).







The current debate in Sub-Saharan Africa is how to increase the productivity of
labor and consequently agricultural income. Conceptually, increasing labor productivity
involves two groups of related actions. The first group concerns actions that will prompt
movement along the production curve to a point where the value of the marginal
product of labor is equal to the implicit wage rate. At that level, the average product
of labor maximizes the returns to labor input. The second group of actions includes
those that lead to an upward shift in the farm production curve. In sum, the first group
of actions takes an improvement approach, while the second defines a transformation.
A combination of these actions results in a substantial increase in the labor productivity
(average product) of the farm sector.
This section examines whether a movement along the production curve, given the
current resource base in the study area, could result in increased labor productivity,
and if so what actions need to be considered in order to achieve this goal. Put differently,
it analyzes, on the basis of survey data, the effects of an increased allocation of labor
to farming on the productivity of labor.
Grouping the sample households on the basis of cultivated land area, as reported
in Table 26, reveals that substantial increases in labor time allocated to farming are
possible and, if accomplished, would ultimately result in a tremendous increase in
labor productivity. Conservatively, it is possible for all households in the area to achieve,
with the current resource base, an average level of labor productivity of about 3.88
kilograms of cereal-equivalents per man-hour. This represents an increase of about 26
percent over the labor productivity of 3.08 kilograms of cereal-equivalents per man-hour
attained in the study area as a whole.
This is not sufficiently brought out in most studies simply because researchers tend
to estimate resource productivities computed at the mean values of the inputs for the
study sample as a whole. This procedure unwittingly masks the fact that a disaggregation
of the study sample indicates that a large part of the study households could be productive
at a level that is compatible with their current resource base if appropriate actions
were taken. This leads to the conclusion that by increasing the time spent performing
land preparation and other farm operations, productivity would improve. As Chapter
5 has shown, increasing the labor allocated to farming requires that both the domestic
terms of trade of farm products and the content of capital input in the farm sector be
improved.
Finally, it needs to be stressed that in increasing capital content not only should
the availability of tools and seeds be increased through appropriate distribution, market-
ing, and credit policies, but also the nature of these capital inputs should be changed.
High-yielding seeds in conjunction with fertilizers and pesticides should be provided-
particularly for rice and maize production. Introducing these technological changes
into the production process will improve the productivity of labor and other primary
inputs (Dodge 1977; Mellor and Ranade 1986; Hossain 1988).








Table 26-Allocation and productivity of labor, 1982/83


Cultivated
Area


Available Average Labor Input to Farming Volume of Output
Number of Labor Force Area per Per Per Man- Per Per Man- Intensity of Labor
Households perHousehold Household Household Equivalent Household Equivalent LaborUse Productivity


(hectare) (percent) (man- (hectares) (man-hours) (kilograms of cereal- (man-hours/ (kilograms of
equivalents) equivalents) hectare) cereal-
equivalents/
man-hour)
0.25orless 12.88 2.83 0.19 442.72 156.44 726.06 256.56 2,330.11 1.64
0.26-0.50 24.24 3.01 0.38 617.87 205.27 1,278.99 424.91 1,625.97 2.07
0.51-0.75 29.55 2.97 0.62 805.23 271.12 2,455.95 826.92 1,298.76 3.05
0.76-1.00 14.39 3.01 0.86 996.50 331.06 3,537.58 1,175.28 1,158.72 3.55
More than 1.00 18.94 3.35 1.32 1,247.56 372.41 4,840.53 1,444.93 945.12 3.88
Studyarea 100.00 3.04 0.67 824.42 271.19 2,539.21 835.27 1,230.48 3.08

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut Facultaire des Sciences Agronomiques,
Yangambi, Zaire, and International Food Policy Research Institute, Washington, D.C., 1986 (computer printout).









8


POLICY IMPLICATIONS AND CONCLUSIONS


Resource Use and Allocation
The survey indicates that, despite ready availability of land, the cultivated area per
household and the cultivated land-man ratio remain small, largely because the amount
of labor time allocated to farming is limited and capital inputs are scarce. The elasticity
of cultivated area with respect to labor allocated to land preparation was estimated at
about 0.75 and that with respect to input of capital tools at 0.25, which suggests that
the expansion of cultivated area is primarily determined by the labor time spent on
land preparation and secondarily by the amount of capital tools.
According to this study, only 271 man-hours per man-equivalent were spent on
farming in this zone during 1982/83-a much smaller time than has been recorded
elsewhere in Sub-Saharan Africa. Cleave (1974) reported an average of 1,000 man-hours
and Haswell (1953) and Norman (1972) mention 500 to 600 man-hours per man-
equivalent per year. Farming's share of productive labor time was only about 35 percent;
the remaining 65 percent was spent on activities such as hunting, fishing, and gathering.
The finding that a 1.00 percent increase in the preplanting farmgate price of farm
products relative to nonfarm goods was associated with a 0.38 percent increase in the
amount of labor allocated to farming indicates that farmers are responsive to price
increases, but they perceive that other activities are more productive than farming.
Another critical factor affecting the amount of labor devoted to farming in the study
area is the availability of capital inputs at the household level. A 1.00 percent increase
in the amount of capital input per man-equivalent allocated to farming resulted in a
0.28 percent increase in the amount of labor input devoted to farming.
Although time spent on agricultural labor is low for small farmers in the Zairian
Basin, the amount of labor required to farm a hectare of land is high. For all crops
combined (rice, maize, cassava, and plantain), an average labor input of about 1,230
man-hours per hectare was recorded during the 1982/83 crop year. Labor intensity
ranged from 525 man-hours per hectare for plantain to 1,547 man-hours for rice. This
per hectare labor use is high compared with other parts of Sub-Saharan Africa and
Asia. For example, 822 man-hours of labor per hectare of sorghum were recorded in
Eastern Burkina Faso and 568 man-hours per hectare of maize in the Mumbwa area
of Zambia. The production of local varieties of sorghum in Maharashtra State, India,
required 477 man-hours per hectare, and that of local varieties of aman rice in
Bangladesh 1,067 man-hours.
The high labor input per hectare in the Zairian Basin is mostly explained by the
large amount of energy consumed in land preparation with hand tools in this rain-forest
area. The tropical climate saps farmers' energy and contributes to the drudgery of
farming. To these environmental constraints, the relatively small amount of labor
allocated to land preparation constitutes an additional hindrance. The intensity of labor
use per hectare was found to be negatively related to the size of cultivated area. A
1.00 percent decrease in cultivated area is associated with a 0.42 percent increase in
the per hectare labor use. Because so little land is prepared, the intensity of labor use
per hectare tends to be higher. Increasing labor allocated to land preparation would








help reduce the intensity of labor use per hectare. A 1.00 percent increase in labor
allocated to land preparation is associated with a 0.75 percent increase in cultivated
area, which, in turn, translates into a 0.32 percent decline in the amount of labor
input per hectare.


Resource Productivities
Considering the productivities of resources-that is, the average products of inputs
and their values-the study shows that cassava and plantain are the most productive
and profitable crops in the area. The expansion of cassava and plantain output has to
be tied to the development of livestock and related industries. Without this vertical
integration, the long-term prospects are not encouraging, because demand for these
crops (for human consumption) tends to decline as incomes rise. Rice and maize, the
two cereals produced in the area, have not been very productive, although agronomic
studies conducted in the area suggest that the production of improved varieties could
yield at least 2,500 and 3,000 kilograms per hectare, respectively, compared with
about 900 and 1,100 kilograms per hectare at present.
Low productivity of labor in the study area, particularly in the production of cereals,
is indicative of the small-farm sector in Zaire. And this is one of the key factors
underlying the decrease in per capital food output that has characterized the country
for the last two decades. To reverse the situation, substantial efforts to enhance the
productivity of labor input have to be initiated.

Conclusions
To increase the productivity of agricultural labor in the Zairian Basin, the study
suggests that efforts to stimulate farmers to allocate more labor to farming should be
initiated by the Zairian Government. Since allocation of labor to farming is mostly
determined by the domestic terms of trade between farm and nonfarm products and
by the content of capital input available to farmers, two sets of related policy actions
need to be considered.
First, the public sector should assume a larger role in efforts to integrate this remote
area into the mainstream of the national economy through investment in infrastructural
development such as roads and transportation systems. This integration will enhance
the terms of trade of farming in the area.
Second, steps to promote the capital market and to increase farmers' access to this
market through a set of policies including credit and pricing policies must be taken.
Specifically, measures to establish a distribution and marketing system for tools, equip-
ment, and seeds need to be examined in the short-to-medium term.
Furthermore, with respect to the introduction of improved rice seeds, investments
in biological and chemical technologies offer opportunities to improve the productivity
of labor and other primary inputs, particularly in the production of cereal crops.
Each of these policy actions is indispensable for getting agriculture moving in the
area. The cost of these policy actions is important, but whether these expenditures
have to be made is not a debatable issue. The Zairian Basin represents almost 40
percent of the total land mass. Investments in this region are a must if a stimulating
economic environment is to be created despite the region's sparse population. The
development of the farm sector alone (food and nonfood crops) will not cover the costs,
but because of these investments, other economic activities will develop.








Perhaps the most important open question is that of sequencing and timing of
investments in view of limited government and private resources. A detailed analysis
of this issue is a necessary and useful complement to this study. Based on the findings
here, however, the first actions should include maintenance of the existing road net-
work, the development of an input delivery system for tools, equipment, and seeds,
and the provision of credit. Finally, it should be stressed that the rain forest zone of
Sub-Saharan Africa is more suited to tree crops than to annual food crops. Logically,
therefore, the appropriate smallholder agriculture to promote in the Zairian Basin in
the long run is a tree crop production system, with farm households relying largely on
purchased food. But, in the short and medium terms, smallholders in the area will
continue to be primarily food-producing.
It is hoped that the findings of this study will provide a guideline for policymakers
to follow in designing policies to enhance the production of food crops in the short
and medium terms, as well as the production of tree crops by smallholders in the long
run. Although this study is concerned with a rather small part of rural Zaire, both the
findings and the policy implications should have wider relevance for Zaire as a whole.








APPENDIX 1:

SOME METHODOLOGICAL NOTES


Statistical Method Used in Allocating Cultivated
Area Among Mixed Crops

The method formulated here was proposed by the Institut National des Statistiques
et d'Etudes Economiques de France in the book entitled La Statistique Agricole dans
les Pays en vole de DAveloppemen4 as presented in Kaseko (1976).
The symbols are defined as follows:

A and B = two crops in mixed stands,
dAand dB = densities of these crops in
a pure cropping system,
d, and dB = densities of the same crops
when they are mixed,
TA and TB = areas effectively occupied
by these crops, and
T = total area.
Let
aA = di/dA and aB = dg/dB.

It is accepted that the areas occupied by A and B are proportionate to aA and aB
respectively. It is also accepted that

TA/aA = TB/aB, and TA + TB = T.
Thus,
TA = T(aA/aA + aB), and TB = T(aB/aA + aB).

This can be generalized to include n crops in mixed stands as follows:

Ti = ciT,
where
n
ci = Ai/ a,.


The coefficient, cp, is referred to as the part of the cultivated land effectively occupied
by crop i.
For the purpose of the study, pure crop densities were computed using the following crop
spacings proposed by Institut National d'Etude et de la Recherche Agronomique (1971):








0.30 meter x 0.20 meter for rice,
0.75 meter x 0.50 meter for maize,
1.00 meter x 1.00 meter for cassava, and
4.00 meters x 4.00 meters for plantain.
Mixed crop densities were estimated directly from yield plots.

Yield Plot Method
Yield plots of 9 square meters each were selected for rice, a high-density crop. For
maize and cassava, plots of 16 square meters and 25 square meters each were consid-
ered. Because the number of plantains in the fields was limited, a complete count was
performed. The operation was spread over a long period of time during the survey for
cassava and plantain, slow-maturing crops.
The yield plot method tends to overestimate actual crop yields, particularly when
the fields are heterogeneous and contain mixed crops (Tollens 1975; Poate and Casley
1985). To reduce the level of this error, five plots were delimited in each household
field. The number of plants in these five plots were counted. Due to variation in soil
fertility, crop density, and a number of other factors, the pattern of crop yield variation
within a field approximates a clustered rather than a random distribution. Therefore,
at harvest time, each field was divided into four clusters. The number of plants counted
in the five yield plots were randomly harvested all over each cluster; and the crop
yield for the field was then derived.
This method provides an estimate of the biological rather than the economic yield
because it does not account for crop losses. Nevertheless, the estimates of biological
crop yields are still relevant for policy analysis. The estimation of crop losses that occur
between harvest and the disposition of the crop constitutes an important research topic
on its own merits.

Collinson's Conversion Factors for Man-Equivalents
In his book entitled Farm Management in Peasant Agriculture, A Handbook for
Rural Development Planning in Africa, Collinson (1972) proposes a set of conversion
factors that take age and sex into account in transforming farm household size into
man-equivalents. Spencer (1972) in his rice study conducted in Sierra Leone proposes
a different set of conversion factors. Norman (1973b) in his study of peasant agriculture
among the Hausa in northern Nigeria suggests yet another set of conversion factors.
After a critical examination of these three sets of conversion factors in the context of
small farming in the Zairian Basin, Collinson's conversion factors were adopted (see
Table 27). The breakdown of man-equivalents per household for the entire study area
is given in Table 28.









Table 27-Collinson's conversion factors for man-equivalents

Age Group Male Female
(years)
Less than 10 0.00 0.00
10-14 0.25 0.25
15-19 0.67 0.50
20-55 1.00 0.67
More than 55 0.67 0.50

Source: M. P. Collinson, Farm Management In Peasant Agriculture, a Handbook for Rural Development Planning
in Africa (New York: Praeger, 1972).


Table 28-Number of man-equivalents per household in the study area

Number of
Rural Community Households Males Females Total
(man-equivalents/household)
Turumbu 53 2.19 1.69 3.88
(1.12) (0.73) (1.53)
Yangambi-Kisangani roada 17 2.47 1.77 4.24
(1.11) (0.88) (1.69)
Weko and Yambaw 17 2.12 1.87 3.99
(1.23) (0.73) (1.49)
Yangambi-lsangi roadb 19 1.99 1.46 3.45
(1.02) (0.56) (1.38)
Mongandjo 53 1.29 1.23 2.52
(0.59) (0.59) (0.92)
Bolikango 25 1.35 1.33 2.68
(0.68) (0.69) (1.05)
Babendja 28 1.24 1.13 2.37
(0.50) (0.48) (0.77)
Bomenge
Bomaneh 26 1.28 1.10 2.38
(0.41) (0.56) (1.73)
Studyarea 132 1.65 1.39 3.04
(0.41) (0.69) (1.36)

Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
Note: Figures in parentheses are standard deviations. Collinson's conversion factors reported in Table 27 are
used to compute man-equivalents.
a These villages were located up to 22 kilometers from Yangambi.
b These villages were located up to 20 kilometers from Yangambi.











APPENDIX 2: SUPPLEMENTARY TABLES



Table 29-Agricultural calendar for study villages, 1982/83

Location/Farm Operation J F M A M J J A S O N D


Yangambi-Kisangani road'
Slashing
Burning and clearing
Planting
Weeding
Harvesting, processing, and marketing
Weko and Yambaw
Slashing
Burning and clearing
Planting
Weeding
Harvesting, processing, and marketing
Yangambi-Isangi roadb
Slashing
Burning and clearing
Planting
Weeding
Harvesting, processing, and marketing
Bolikango
Slashing
Burning and clearing
Planting
Weeding
Harvesting, processing, and marketing
Babendja
Slashing
Burning and clearing
Planting
Weeding
Harvesting, processing, and marketing
Bomaneh
Slashing
Burning and clearing
Planting
Weeding
Harvesting, processing and marketing


0 0 0
0


* 0












* 0


* 0 0


0


* 0






0
0


0





0
0





0


Notes: This calendar includes only farm operations carried out in the ongoing crop year. Operations related to
last crop season performed mainly by female members are weeding of cassava and plantain, and marketing
of rice and maize during the January-February period. Harvesting and marketing of cassava and plantain
continue throughout the year.
SThese villages were located up to 22 kilometers from Yangambi.
b These villages were located up to 20 kilometers from Yangambi.


* *




* *
* *




*
* *


*
*
* *
* *




* *
* *




* *
* *
* 0




* 0
* 0







Table 30-Monthly intensity of labor allocated to farm and nonfarm activities per member household, 1982/83

Turumbu Villages
AlongYangambi- Wekoand AlongYnggambi- Mongandjo Bomenge Average for
KisanganiRoad Yambaw Isang Road Bolikango Babendja Bomaneh StudyArea
Non- Non- Non- Non- Non- Non- Non-
Month Farming farming Farming farming Farming farming Farming farming Farming farming Farming farming Farming farming


Males
January 20
February 23
March 25
April 24
May 18
June 17
July 21
August 14
September 18
October 20
November 17
December 14
Females
January 18
February 0
March 1
April 1
May 4
June 9
July 11
August 14
September 11
October 24
November 28
December 25


(number of hours/month)

28 69 44 61 20
51 50 55 36 37
37 40 59 29 35
66 28 55 45 23
59 20 15 81 24
46 27 21 114 12
45 36 6 180 6
22 53 7 200 3
19 58 7 186 14
9 67 14 109 12
9 77 12 152 3
3 84 18 155 1

3 45 0 13 9
0 49 0 28 1
1 67 0 28 3
1 19 8 34 10
1 5 6 36 56
31 6 49 55 24
76 25 30 73 6
74 36 22 197 7
38 22 28 140 15
25 23 55 26 12
58 29 39 90 6
53 25 55 61 4


Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut Facultaire des Sciences Agronomiques,
Yangambi, Zaire, and International Food Policy Research Institute, Washington, D.C., 1986 (computer printout).
Note: These are hours allocated to farming and nonfarming by average household member of either gender regardless of age. They should not be regarded as
man-hours. Nonfarm activities refer to hunting, fishing, and gathering.









Table 31-Preplanting farmgate prices in the study area, 1982/83

Rural Community/ Rice Cassava Plantain All Four Game
Farm Number Villages Paddy Maize (Raw) Bunch Crops* Meat

(Z/kilogram)
Turumbu
2-16 YalunguandYaselia 3.65 1.10 0.59 0.78 0.87 13.04
17-23 Bosukulu 3.98 1.26 0.64 0.79 0.95 15.47
24-29 YaondolollandYandimbiaII 5.00 1.14 0.65 0.71 1.03 13.81
30-35 Yalolia II 5.00 1.18 0.73 0.64 1.10 15.86
36-48 Weko 3.16 1.01 0.29 0.53 0.57 18.23
49-61 Yambaw 1.78 1.04 0.25 0.40 0.43 14.19
62-76 YakwondiandObiloto 3.73 1.20 0.75 1.47 0.73 9.48
77-82 Lilanda 3.99 1.40 1.46 1.75 1.66 8.31
83-85 Yambele and Yaisowa 2.91 1.24 1.26 1.64 1.39 7.52
Mongandjo
86-105 Bolikango I and I 3.88 1.43 1.01 1.08 1.27 5.28
106-115 Bomboma 3.45 1.43 1.27 1.43 1.45 14.00b
116-137 BolimaandBokangela 2.75 1.38 1.33 2.25 1.45 13.06
138-145 BokobangoloandBodjonga 2.50 1.00 1.00 1.35 1.11 9.81
Bomenge
146-175 Bomaneh 5.53 2.03 1.45 1.87 1.82 5.89b
Studyarea
2-195 3.67 1.27 0.91 1.19 1.12 11.71

Source: Based on data from Tshikala B. Tshlbaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
* The share of each crop in total labor input allocated to farming was used as a weight for the weighted average
farmgate price for the period of September 1982 to March 1983. During the period between September and
March, most of the land preparation is performed.
b The price of fish is given instead of meat.










Table 32-Regression results for food crop production functions using capital
as a deflator for rice and labor for other crops

Explanatory Crop/DependentVariable
Variables and Rice Maize Cassava Plantain All Crops
Important Statistics LnO/K LnQ/L LnQ/L LnQ/L Ln /L


Intercept


LnL/K

LnK/L

Dr

Df

S

Dummies for location
D,


R2
F
n
Elasticity of output
with respect to
L
K


-1.06
(-2.14)*
0.13
(0.78)


2.14
(5.06)**
0.21
(0.99)



0.08
(0.18)
1.09
(3.39)**
0.58
(1.64)
0.01
(0.05)


0.42
9.11
80


0.13
0.87


0.60
(1.96)


0.37
(2.64)*


-0.22
(-1.03)



0.73
(2.13)*
-0.16
(-0.41)
-0.07
(-0.17)
-0.52
(-1.55)
-0.09
(-0.25)
0.26
5.02
82


0.63
0.37


2.72 2.92
(17.63)** (12.13)**


0.02
(0.37)


0.04
(0.34)



-0.58
(-2.96)**
-1.17
(-5.60)**
-0.24
(-1.22)
0.65
(3.60)**
0.61
(3.91)**
0.56
23.59
128


0.98
0.02


0.15
(1.61)


0.12
(0.61)



0.41
(1.58)
0.27
(0.66)
-0.10
(-0.38)





0.04
1.39
52


0.85
0.15


Source: Based on data from Tshikala B. Tshibaka, "Food Production in the Zairian Basin, Zaire, 1982/83," Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire, and International Food Policy Research Institute,
Washington, D.C., 1986 (computer printout).
Notes: Dr is a dummy to reflect the date of introduction of rice in different communities, Df is a dummy to
account for the length of the fallow period, and S is a dummy for the share of cassava in cultivated area
to reflect the effect of the crop mix adopted.


0.94
(4.31)**


0.22
(3.09)**


0.05
(0.44)
0.83
(2.57)**

0.11
(0.54)
-0.78
(-3.34)**
-0.25
(-1.31)
0.53
(3.06)**
0.34
(2.22)*
0.53
19.34
132


0.78
0.22








BIBLIOGRAPHY



Bernard, G. 1945. Climate ecologique de la cuvette Congolaise: Influence de complex
climatique sur la vegetation. Brussels, Belgium: Institut National pour l'Etude
Agronomique du Congo.
Booth, A., and R. M. Sundhrum. 1985. Labour absorption in agriculture. Delhi: Oxford
University Press.
Cleave, J. H. 1974. African farmers: Labor use in the development of smallholder
agriculture. New York: Praeger.
Collinson, M. P. 1972. Farm management in peasant agriculture: A handbook for rural
development planning in Africa. New York: Praeger.
Crabbe, M. 1970. Ecolimat de Yangambi. Yangambi, Zaire: Institut National d'Etude
et de la Recherche Agronomique.
Delgado, Christopher L. 1982. Livestock versus foodgrain production in Southeast
Upper Volta: A resource allocation analysis. In Livestock production and market-
ing in the Entente states of West Africa. Monograph No.1. Ann Arbor, Mich.,
U.S.A.: Michigan State University, Department of Agriculture.
den Tuinder, B. A. 1978. Ivory Coast: The challenge ofsuccess. Baltimore, Md., U.S.A.:
Johns Hopkins University Press.
de Wilde, J. C. 1967. Experiences with agricultural development in tropical Africa, 2
vols. Baltimore, Md., U.S.A.: Johns Hopkins University Press.
Dodge, D. J. 1977. Agricultural policy and performance in Zambia. Berkeley, Cal.,
U.S.A.: University of California, Institute of International Studies.
Durocher, J. 1983. Systeme agricole du haute-altitude Zaire. M.Sc. thesis, l'Universitd
Laval, Quebec, Canada.
Eicher, C. K., and D. C. Baker. 1982. Research on agricultural development in Sub-
Saharan Africa: A critical survey. MSU International Development Paper. East
Lansing, Mich., U.S.A.: Michigan State University.
Elliott, C. M., J. E. Bessell, R. A. J. Roberts, and N. Vanzetti. 1970. Some determinants
of agricultural labour productivity in Zambia. Agricultural Labour Productivity
Investigation Report No. 3. University of Nottingham, United Kingdom, and
University of Lusaka, Lusaka, Zambia.
Greenland, D. J., and B. N. Okigbo. 1983. Crop production under shifting cultivation
and maintenance of soil fertility. In Potential productivity offield crops under
different environments. Los Bafios, Philippines: International Rice Research
Institute.
Haswell, M. R. 1953. Economics of agriculture in a savannah village, report on three
years study in Genieri Village and its lands: The Gambia. Colonial Research
Study No. 8. London: Her Majesty's Stationery Office.







Hayami, Yujiro, and Vernon W. Ruttan. 1985. Agricultural development. Baltimore,
Md., U.S.A.: Johns Hopkins University Press.
Hazell, Peter B. R., and Ailsa Rbell. 1983. Ruralgrowth linkages: household expenditure
patterns in Malaysia and Nigeria. Research Report 41. Washington, D.C.: Inter-
national Food Policy Research Institute.
Hossain, Mahabub. 1988. Nature and impact of the green revolution in Bangladesh.
Research Report 67. Washington, D.C.: International Food Policy Research Institute.
INERA (Institut National d'Etude et de la Recherche Agronomique). 1971. Rapport
annuel. Yangambi, Zaire: INERA.
Jurion, F., and J. Henry. 1967. De l'agriculture itindrante ad 'agriculture intensive.
Brussels: Institut National d'Etudes Agronomiques du Congo.
Kapambwe, N. M. 1974. La reliance des cooperatives agricoles au Zaire: Etude par-
ticulibre de la cooperative des Turumbu. M6moire de fin d'6tudes, Universite
National du Zaire, Campus de Kisangani, Faculte des Sciences Agronomiques,
Yangambi, Zaire.
Kaseko, U. 1976. Application de la function Cobb-Douglas dans une agriculture
paysanne: Cas des paysans producteurs du riz A Yalibwa. Memoire de fin d'etudes,
Institute Facultaire des Sciences Agronomiques, Yangambi, Zaire.
Kilumba, N. 1975. Determination de revenues, recettes et d6penses chez les planteurs
du riz A Yalibwa. M6moire de fin d'etudes, Universitd Nationale du Zaire, Campus
de Kisangani, Facultd des Sciences Agronomiques, Yangambi, Zaire.
Koutsoyiannis, A. 1984. Theory of econometrics. Hong Kong: Macmillan Press.
Lassiter, G. C. 1982. The impact of animal traction on farming system in Eastern Upper
Volta. Ph.D. diss., Cornell University, Ithaca, N.Y., U.S.A.
Lititiyo, B. 1977. Etude socio-6conomique sur l'action de la Mission Agricole Chinoise
chez les riziculteurs du paysannat Turumbu. Memoire de fin d'etudes, Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire.
Massell, B. F., and R. W. M. Johnson. 1968. Economics of smallholder farming in
Rhodesia. Food Research Institute Studies 8 (supplement).
Matlon, P. et al. 1979. Poor rural households, technical change and income distribution
in developing countries: Two case studies from West Africa. African Rural Econ-
omy Working Paper No. 29. East Lansing, Mich., U.S.A.: Michigan State Univer-
sity, Department of Agricultural Economics.
Mellor, John W. 1976. The new economics of growth: A strategy for India and the
developing world. Twentieth Century Fund Study. Ithaca, N.Y., U.S.A.: Cornell
University Press.
Mellor, John W., and Chandra Ranade. 1986. Technological change in a low-labor-
productivity, land-surplus developing economy: The African development prob-
lem. International Food Policy Research Institute, Washington, D.C. Mimeo.
Mellor, John W., Christopher L. Delgado, and Malcolm J. Blackie, eds. 1987. Accelerat-
ingfoodproduction in Sub-Saharan Africa. Baltimore, Md., U.S.A.: Johns Hopkins
University Press.







Norman, Denis W. 1972. An economic study of three villages in Zaria Province: An
input-output study, Vol. 1, Text. Miscellaneous Paper No. 37. Samaru, Zaria,
Nigeria: Institute of Agricultural Research, Ahmadu Bello University.

1973a. Methodology and problems of farm management investigations: Ex-
periences from Northern Nigeria. African Rural Employment Paper No. 8. East
Lansing, Mich., U.S.A.: Michigan State University.

1973b. Economic analysis of agricultural production and labour utilization
among the Hausa in the north of Nigeria. African Rural Employment Paper No.
4. East Lansing, Mich., U.S.A.: Michigan State University.

Norman, Denis W., D. H. Pryor, and C. J. W. Gibbs. 1979. Technological change and
the small farmer in Hausaland, Northern Nigeria. African Rural Economy Paper
No. 21. East Lansing, Mich., U.S.A.: Michigan State University.
Ntamulyango, B. 1975. Analyse economique des associations des cultures chez les
paysans Turumbu de Yalibwa (Yangambi). Mdmoire de fin d'6tudes, Institut
Facultaire des Sciences Agronomiques, Yangambi, Zaire.
Paulino, Leonardo A. 1986. Food in the Third World: Past trends and projections to
2000. Research Report 52. Washington, D.C.: International Food Policy Research
Institute.
1987. The evolving food situation. In Acceleratingfoodproduction in Sub-Saha-
ran Africa, ed. J. W. Mellor, Christopher Delgado, and Malcolm Blackie. Balti-
more, Md., U.S.A.: Johns Hopkins University Press.
Perreault, P. T. 1978. Banana-manioc farming systems of the tropical forest: A case
study in Zaire. Ph.D. diss., Food Research Institute, Stanford University, Stanford,
Cal., U.S.A.
Poate, C. D., and D. J. Casley. 1985. Estimating crop production in development
projects. Washington, D.C.: World Bank.
Purdue University. 1980. Workshop on Sahelian agriculture. Proceedings of a meeting
held at Purdue University, Lafayette, Ind., U.S.A., May 1980.
Rossi, P. H., V. D. Wright, and A. B. Anderson, eds. 1983. Handbook ofsurvey research.
New York: Academic Press.
Sanchez, P. A. 1976. Properties and management of soils in the tropics. New York:
John Wiley and Sons.
Sen, A. R. 1964. Size of holdings and productivity. Economic and Political Weekly
(February): 323-326.
Shapiro, K. H. 1973. Micro-levelfarm management and production economics research
among traditional African farmers: Lessons from Sierra Leone. African Rural
Employment Paper No. 3. East Lansing, Mich., U.S.A.: Michigan State University.

Spencer, Dunstan. 1972. Micro-level farm management and production economics:
Research among traditional African farmers: Lessons from Sierra Leone. East
Lansing, Mich., U.S.A.: Michigan State University.







Tollens, E. F. 1975. Problmes de rassemblement des donnees micro-dconomiques
dans les exploitations agricoles du Nord-Zaire. African Rural Economy Working
Paper No. 7. East Lansing, Mich., U.S.A.: Michigan State University.
Tshibaka, Mukendi M. 1975. Etude de l'utilisation du travail chez le planteur du riz
dans la locality de Yalibwa (Turumbu). Mdmoire de fin d'dtudes, Institute Facul-
taire des Sciences Agronomiques, Yangambi, Zaire. Mimeo.
Tshibaka, Tshikala B. 1986. The effects of trade and exchange rate policies on agricul-
ture in Zaire. Research Report 56. Washington, D.C.: International Food Policy
Research Institute.
Tshibaka, Tshikala B., and Kamanda Lumpungu. 1989. Trends and prospectsfor cassava
in Zaire. Working Paper 4 on Cassava. Washington, D.C.: International Food
Policy Research Institute.
Upton, M. 1973. Farm management in Africa: The principles of production and plan-
ning. London: Oxford University Press.
Wanmali, Sudhir. 1981. Periodic markets and rural development in India. New Delhi:
B. R. Publishing.
Winch, F. 1976. Costs and returns of alternative rice production systems in Northern
Ghana: Implications for output, employment and income distribution. Ph.D.
diss. Michigan State University, East Lansing, Mich., U.S.A.
World Bank. 1980. Country study: Zaire. Washington D.C.: World Bank.
Zaire. 1980. Calendrier agricole approuv6 pour la sous/r6gion de la Tshopo. Kisangani,
Zaire.
Zaire. 1983. Donnees ddmographiques des collectivitds Turumbu, Mogandjo et
Bomenge: Zone d'Isangi et Zone de Basoko, S/Rdgion de la Tshopo. Haut-Zaire,
Zaire.
Zuckerman, P. S. 1979. A micro-level farm management study in Western Nigeria:
Field work design and application. Discussion Paper No. 7/79. Ibadan, Nigeria:
International Institute for Tropical Agriculture.











RECENT IFPRI RESEARCH REPORTS (continued)


55 THE EFFECTS OF TRADE AND EXCHANGE RATE POLICIES ON AGRICULTURE IN NIGERIA,
October 1986, by T. Ademola Oyejide
54 WEATHER AND GRAIN YIELDS IN THE SOVIET UNION, September 1986, by Padma Desai
53 REGIONAL COOPERATION TO IMPROVE FOOD SECURITY IN SOUTHERN AND EASTERN
AFRICAN COUNTRIES, July 1986, by Ulrich Koester
52 FOOD IN THE THIRD WORLD: PAST TRENDS AND PROJECTIONS TO 2000, June 1986, by
Leonardo A. Paulino
51 DETERMINANTS OFAGRICULTURAL POLICIES IN THE UNITED STATES AND THE EUROPEAN
COMMUNITY, November 1985, by Michel Petit
50 GOVERNMENTEXPENDITURES ONAGRICULTUREANDAGRICULTURAL GROWTH INLATIN
AMERICA, October 1985, by Victor J. Elias
49 LIVESTOCK PRODUCTS IN THE THIRD WORLD: PAST TRENDS AND PROJECTIONS TO 1990
AND 2000, April 1985, by J. S. Sarma and Patrick Yeung
48 RURAL HOUSEHOLD USE OF SERVICES: A STUDY OFMIRYALGUDA TALUKA, INDIA, March
1985, by Sudhir Wanmali
47 EVOLVING FOOD GAPS IN THE MIDDLE EAST/NORTH AFRICA: PROSPECTS AND POLICY
IMPLICATIONS, December 1984, by Nabil Khaldi
46 THE EFFECTS ON INCOME DISTRIBUTION AND NUTRITION OF ALTERNATIVE RICE PRICE
POLICIES IN THAILAND, November 1984, by Prasarn Trairatvorakul
45 THE EFFECTS OF THE EGYPTIAN FOOD RATIONAND SUBSIDY SYSTEM ON INCOME DISTRI-
BUTION AND CONSUMPTION, July 1984, by Harold Alderman and Joachim von Braun
44 CONSTRAINTS ONKENYA'S FOOD AND BEVERAGE EXPORTS, April 1984, by Michael Schluter
43 CLOSING THE CEREALS GAP WITH TRADEAND FOOD AID, January 1984, by Barbara Huddleston
42 THE EFFECTS OF FOOD PRICE AND SUBSIDY POLICIES ON EGYPTIAN AGRICULTURE,
November 1983, by Joachim von Braun and Hartwig de Haen
41 RURAL GROWTH LINKAGES: HOUSEHOLD EXPENDITURE PATTERNS IN MALAYSIA AND
NIGERIA, September 1983, by Peter B. R. Hazell and Ailsa Roell
40 FOOD SUBSIDIES IN EGYPT: THEIR IMPACT ON FOREIGN EXCHANGE AND TRADE, August
1983, by Grant M. Scobie
39 THE WORLD RICE MARKET: STRUCTURE, CONDUCT, AND PERFORMANCE, June 1983, by
Ammar Siamwalla and Stephen Haykin
38 POLICY MODELING OF A DUAL GRAIN MARKET: THE CASE OF WHEAT IN INDIA, May
1983, by Raj Krishna and Ajay Chhibber
37 SERVICE PROVISION AND RURAL DEVELOPMENT IN INDIA: A STUDY OF MIRYALGUDA
TALUKA, February 1983, by Sudhir Wanmali
36 AGRICULTURE AND ECONOMIC GROWTH INAN OPEN ECONOMY: THE CASE OFARGEN-
TINA, December 1982, by Domingo Cavallo and Yair Mundlak
35 POLICY OPTIONS FOR THE GRAIN ECONOMY OF THE EUROPEAN COMMUNITY: IMPLICA-
TIONS FOR DEVELOPING COUNTRIES, November 1982, by Ulrich Koester



S Tshikala B. Tshibaka has been a research fellow at IFPRI since 1984.




RECENT IFPRI RESEARCH REPORTS


73 NONTRADITIONAL EXPORT CROPS IN GUATEMALA: EFFECTS ON PRODUCTION, INCOME,
AND NUTRITION, May 1989, by Joachim von Braun, David Hotchkiss, and Maarten Immink
72 RICE PRICE FLUCTUATIONAND AN APPROACH TO PRICE STABILIZATIONIN BANGLADESH,
February 1989, by Raisuddin Ahmed and Andrew Bernard
71 STORAGE, TRADE, AND PRICE POLICY UNDER PRODUCTION INSTABILITY: MAIZE IN
KENYA, December 1988, by Thomas C. Pinckney
70 AGRICULTURE IN THE GATT: AN ANALYSIS OFALTERNATIVE APPROACHES TO REFORM,
November 1988, by Joachim Zietz and Alberto Valdds
69 CONSEQUENCES OFDEFORESTATIONFOR WOMEN'S TIME ALLOCATION, AGRICULTURAL
PRODUCTION, AND NUTRITION IN HILL AREAS OF NEPAL, October 1988, by Shubh K.
Kumar and David Hotchkiss
68 COFFEE BOOM, GOVERNMENTEXPENDITURE, AND AGRICULTURAL PRICES: THE COLOM-
BIAN EXPERIENCE, August 1988, by Jorge Garcia Garcia and Gabriel Montes Llamas
67 NATURE AND IMPACT OF THE GREEN REVOLUTION IN BANGLADESH, July 1988, byMahabub
Hossain
66 THE BRAZILIAN WHEAT POLICY: ITS COSTS, BENEFITS, AND EFFECTS ONFOOD CONSUMP-
TION, May 1988, by Geraldo M. Calegar and G. Edward Schuh
65 CREDIT FOR ALLEVIATION OF RURAL POVERTY: THE GRAMEEN BANK IN BANGLADESH,
February 1988, by Mahabub Hossain
64 COOPERATIVE DAIRY DEVELOPMENT INKARNATAKA, INDIA: AN ASSESSMENT, December
1987, by Harold Alderman
63 INCOME AND NUTRITIONAL EFFECTS OF THE COMMERCIALIZATION OF AGRICULTURE
IN SOUTHWESTERN KENYA November 1987, by Eileen T. Kennedy and Bruce Cogill
62 AGRICULTURAL RESEARCH IN NEPAL: RESOURCE ALLOCATION, STRUCTURE, AND INCEN-
TIVES, September 1987, by Ram P. Yadav
61 THE PILOT FOOD PRICE SUBSIDY SCHEME IN THE PHILIPPINES: ITS IMPACT ON INCOME,
FOOD CONSUMPTION, AND NUTRITIONAL STATUS, August 1987, by Marito Garcia and Per
Pinstrup-Andersen
60 POPULATION POLICY AND INDIVIDUAL CHOICE: A THEORETICAL INVESTIGATION, June
1987, by Marc Nerlove, Assaf Razin, and Efraim Sadka
59 PRODUCTION INCENTIVES IN PHILIPPINE AGRICULTURE: EFFECTS OF TRADE AND EX-
CHANGE RATE POLICIES, May 1987, by Romeo M. Bautista
58 THE FOOD STAMP SCHEME IN SRI LANKA: COSTS, BENEFITS, AND OPTIONS FOR MODIFI-
CATION, March 1987, by Neville Edirisinghe
57 CEREAL FEED USE IN THE THIRD WORLD: PAST TRENDS AND PROJECTIONS TO 2000,
December 1986, by J. S. Sarma
56 THE EFFECTS OF TRADE AND EXCHANGE RATE POLICIES ON AGRICULTURE IN ZAIRE,
November 1986, by Tshikala B. Tshibaka
fontinned on inside back cover)


Wah0gtn D.e* 2003 US




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

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