34246
Africa Region
Working Paper Series No. 85
Poverty Reducing Potential of
Smallholder Agriculture in Zambia:
Opportunities and Constraints
Prepared by
Paul B. Siegel
Jeffrey Alwang
for
Social Development Department (SDV)
The World Bank
Washington, DC, USA
July 2005
Poverty Reducing Potential of
Smallholder Agriculture in Zambia:
Opportunities and Constraints
Africa Region Working Paper Series No. 85
July 2005
Abstract
T he objective of this paper is to investigate poverty-reducing potential of smallholder agriculture in
Zambia, considering suggested public actions expected to stimulate broad-based growth in the rural
economy. This paper was prepared as part of the Poverty and Social Impact Analysis (PSIA) work carried
out to inform the Zambia Country Economic Memorandum (CEM). Following PSIA guidelines, it combines
quantitative and qualitative analyses from a variety of sources
Several analytical techniques are used to examine opportunities and constraints facing typical Zambian
smallholders. These include enterprise budgets, a smallholder household model, and reviews of recent
studies. Using a one-period linear programming model of agricultural production activities by smallholder
households, the authors examine potential impacts of land tenure reforms, changes in fertilizer policy,
infrastructure investments, and HIV/AIDS on land and labor use and income-generation.
Based on the results of the smallholder model and other analysis, the authors are not very optimistic about
the potential for agriculture-led poverty reduction, especially in the short term. A major transformation is
required for smallholders' agricultural potential to become reality. Large-scale investments are needed in
research and extension, market and transport infrastructure, capacity-building for individual farmers and
groups of farmers, and more. These investments require time to reach fruition. Policies to stimulate non-
agricultural economic activities in rural areas should also be explored. Although high seasonal labor
demands create bottlenecks at key times, there is an opportunity to spread labor more evenly through the
year as a means to increase household incomes. In all cases, a more comprehensive and holistic approach to
rural development is needed, not just an agricultural or commodity-specific strategy.
The Africa Region Working Paper Series expedites dissemination of applied research and policy studies with
potential for improving economic performance and social conditions in Sub-Saharan Africa. The series publishes
papers at preliminary stages to stimulate timely discussions within the Region and among client countries, donors,
and the policy research community. The editorial board for the series consists of representatives from professional
families appointed by the Region's Sector Directors. For additional information, please contact Momar Gueye,
(82220), Email: mgueye@worldbank.org or visit the Web Site: http://www.worldbank.org/afr/wps/index.htm.
The findings, interpretations, and conclusions in this paper are those of the authors. They do not necessarily
represent the views of the World Bank, its Executive Directors, or the countries that they represent and
should not be attributed to them.
Authors' Affiliation and Sponsorship
Paul B. Siegel psiegel@worldbank.org, pbs11pbs@yahoo.com
Jeffrey Alwang alwangj@vt.edu
Acknowledgement
Guidance for this research was provided by Steen Lau Jorgensen (Sector Director, SDV) and Zlatina Loudjeva
(Junior Professional Associate, SDV). Research assistance was provided by Mr. Sibusiso Moyo, a graduate student
at Virginia Tech. Dr. Mwape, of UNZA in Zambia, prepared a background paper on cropping patterns and crop
budgets that was a valuable resource. Helpful comments were received from Abebe Adugna (AFTP1), Stefano
Paternostro (PRMPR) and Anis Dani (SDV). Administrative support was provided by Milagros Benedicto (SDV).
Editorial assistance was provided by John-Paul Ferguson. Any errors are solely the responsibility of the authors, as
are all the opinions.
This paper was prepared as part of the Poverty and Social Impact Analysis (PSIA) work carried out to inform the
Zambia Country Economic Memorandum (CEM). Funding was provided by the Norwegian government. The
original draft of this report was delivered in June 2003. Revisions of the original draft report were coordinated with
Abebe Adugna (AFTP1). Momar Gueye provided logistical support.
More information on the PSIA approach can be found at: : www.worldbank.org/psia
Poverty Reducing Potential of Smallholder Agriculture in Zambia:
Opportunities and Constraints
Prepared by:
Paul B. Siegel
Consultant
The World Bank
Washington, D.C., USA
psiegel@worldbank.org, pbs11pbs@yahoo.com
and
Jeffrey Alwang
Professor
Virginia Tech
Blacksburg, VA, USA
alwangj@vt.edu
For
Social Development Department (SDV)
The World Bank
Washington, DC, USA
Abbreviations and Acronyms
ACP Agriculture Commercialization Programme
ASIP Agricultural Sector Investment Program
CEM Country Economic Memorandum
CF Conservation Farming
CIF Cost, Insurance and Freight
CLUSA Cooperative League of United States of America
CSO Central Statistics Organization
FHH Female-Headed Household
FAO Food and Agricultural Organization
GRZ Government of the Republic of Zambia
Ha Hectare
HH Household
HYV High Yielding Variety
IDS Institute for Development Studies
Kg Kilogram
Km Kilometer
LP Linear Programming
MAC Ministry of Agriculture and Cooperatives
MAFF Ministry of Agriculture, Food, and Fisheries.
MHH Male-Headed Household
Mt Metric Tons
NGO Non-governmental Organization
PHS Population and Health Study
PRSP Poverty Reduction Strategy Paper
PSIA Poverty and Social Impact Assessment
ROADSIP Road Sector Investment Project
ZK Zambian Kwacha
TABLE OF CONTENTS
EXECUTIVE SUMMARY .............................................................................................................I
1. INTRODUCTION .................................................................................................................. 1
Background and Objectives.................................................................................................... 1
Structure of the Agricultural Sector........................................................................................ 2
Constraints to Smallholder Production................................................................................... 4
Changes in the Agricultural Sector since the 1990s ............................................................... 6
2. THE SMALLHOLDER MODEL......................................................................................... 15
Model Description ................................................................................................................ 15
Main Findings from the 1994 Baseline Model and Scenarios.............................................. 16
Updating the Model in View of the Changes in the 1990s................................................... 17
3. UPDATED BASELINE SMALLHOLDER MODEL, POLICY REFORM SCENARIOS,
AND RESULTS................................................................................................................... 19
Updated Baseline Model....................................................................................................... 19
Stylized Policy Reform Scenarios ........................................................................................ 21
Model Results ....................................................................................................................... 22
Beyond the Model: Moving into Higher Value Crop Activities and Conservation Farming25
4. SUMMARY AND CONCLUSIONS ................................................................................... 28
Looking Ahead: Tapping Smallholder Potential .................................................................. 28
Suggestions for Further Research......................................................................................... 30
REFERENCES ............................................................................................................................. 31
Tables
Table 1: Typology of Agricultural Producers in Zambia.............................................................. 36
Table 2: Estimated Shares (percent) of Zambian National Production by Typology of Producers
............................................................................................................................................... 36
Table 3: Major Agro-Ecological Zones of Zambia...................................................................... 37
Table 4: Distribution of Rural Households, Land per Household and Agricultural Activities, by
Province ......................................................................................................................... 37
Table 5: Major Smallholder Crops in Zambia 2000 � 2002 and Distribution by Province......... 38
Table 6: Estimated Composition of Zambia's Agricultural GDP (percent) ................................ 38
Table 7: Major Crops in Southern, Eastern and Northern Provinces (in hectares)....................... 38
Table 8: Baseline model results................................................................................................... 39
Table 9: Model results with land constraint, output and input price adjustments ....................... 40
Table 10: Model results under remoteness scenario, Eastern Province....................................... 41
Table 11: Labor decreased by 20 percent (columns 2 and 4). ..................................................... 42
Table 12: Net incomes by crop, financial model......................................................................... 43
Table 13: Net incomes by crop, financial model......................................................................... 43
Table 14: Comparison of Yields, Inputs, Labor, and Returns of Different Smallholder Crops
1994/5 and 2001/2....................................................................................................... 43
Table 15: Estimated Costs of Entry into Commercial Farming for Selected Enterprises ........... 44
Figure
Figure 1: Maize land planted and total production, 1993-2002................................................... 45
Figure 2: Land planted to various crops, 1993-2002................................................................... 46
Figure 3: Production of various crops, 1993-2002 ...................................................................... 47
Figure 4. Impacts of landholding on net returns.......................................................................... 47
Figure 5: Impacts of increased labor availability on net returns and acreage under cultivation,
hand-hoe and oxen technology households. .................................................................. 48
Figure 6: Impacts of cash constraints on model solution, Eastern Province. .............................. 48
Figure 7: Impacts of fertilizer price changes on model solution, Eastern Province. ................... 49
Boxes
Box 1: Agricultural Performance Since the mid-1990s................................................................. 6
Box 2: Fertilizer Prices and Maize Production.............................................................................. 8
Box 3: Inter-relationships Among Constraints in Fertilizer Supply.............................................. 9
Box 4: Changes in Agricultural Production Patterns................................................................... 10
Box 5: Data Constraints in Zambia Limit Analytical Techniques............................................... 16
Box 6: Using the Smallholder Model for Policy Dialogue with and Stakeholders ..................... 16
Box 7: Resettlement Schemes: Urban to Rural Migration........................................................... 28
Box 8: Needs for Sustainable Smallholder Farming Systems ..................................................... 29
Annexes
ANNEX A:.................................................................................................................................... 53
ANNEX B..................................................................................................................................... 56
EXECUTIVE SUMMARY
Zambia's recent Poverty Reduction Strategy Paper (PRSP) views development of the
agricultural sector as a critical pillar for rural growth and poverty reduction. The PRSP
proposes, inter alia, improving the land tenure system, liberalizing fertilizer markets, and
improving rural infrastructure. These actions are expected to stimulate broad-based growth in
the rural agricultural economy.
Given the importance of agriculture to the livelihoods of rural Zambians, most poverty
reduction strategies focus on improving income generation through agriculture. Since
smallholder agriculture predominates throughout the country, the impacts of policy reforms and
investments on smallholders will, to a large extent, determine effectiveness in reducing rural
poverty. An important consideration is whether agriculture can be a viable poverty exit strategy
for rural households in Zambia.
The objective of this paper is to assess the poverty-reducing potential of smallholder
agriculture in Zambia, in light of suggested public actions. Several analytical techniques are
used to examine the opportunities and constraints facing typical smallholders. These include
basic enterprise budgets, a smallholder household model, and reviews of recent studies. This
paper was prepared as part of the Poverty and Social Impact Analysis (PSIA) work carried out to
inform the Zambia Country Economic Memorandum (CEM). Following PSIA guidelines, it
combines quantitative and qualitative analyses from a variety of sources (see World Bank,
2003a).
Most traditional smallholder crops in Zambia, such as local maize, groundnuts and
cotton, have low input costs, but also relatively low returns. Given the lack of dependable input
and output markets and high prices of inputs relative to outputs, the low-input low-return
strategy is an economically rational strategy for smallholders, especially those located in remote
areas. In many cases, labor and credit constraints dominate land constraints. Higher-value
crops, such as paprika and tobacco, have higher cash-input and labor demands. Conservation
farming, a minimum tillage technique that is gaining popularity because it increases the
efficiency of inputs, can also require additional purchased inputs and labor. Furthermore, the
prices of many higher-value crops stagnated or fell since 1994, while input costs tended to rise.
Crop budgets for commercially oriented emergent farmers and large-scale farmers might
show more promise to increase incomes, but initial investments and operating costs for such
enterprises require substantial outlays. Complementary infrastructure and support services are
also needed. Hence, a major transformation of smallholder agriculture faces serious constraints.
Although adoption of improved technologies by smallholders might contribute to improvements
in household food security and supplement incomes, given their limited assets, they cannot be
expected to make major contributions to reducing rural poverty -- especially in the short term.
The negative impacts of HIV/AIDS on labor availability and increased dependency ratios have
exacerbated the constraints facing many smallholder households.
Using a single-period linear programming model of agricultural production activities by
smallholder households, we examine the potential impacts of land tenure reforms, changes in
fertilizer policy, infrastructure investments, and HIV/AIDS on land and labor use, and income-
generation. Major findings include:
� Food security concerns imply a significant (real and opportunity) cost, even though staple
food production to assure food security is the main goal of most smallholders, especially
in remote areas.
� Using hoes and oxen, smallholder households can only cultivate about two to four
hectares of land, respectively, before household labor constraints become binding.
Increased access to land would therefore not benefit most households unless hired labor
became available at a relatively low cost. Labor-saving technology may substitute for
labor and allow for improved labor allocation. Or technologies such as conservation
farming can help spread out labor demand more evenly though the year. However, the
"entry costs" of such technologies may be prohibitively expensive for the majority of
resource-poor smallholders.
� Higher official market prices for fertilizer and lower prices for maize have made hybrid
maize production less economically attractive for smallholders. Lower maize/fertilizer
ratios and the continued lack of dependable input supplies have slowed the adoption of
improved technologies and encouraged a "retreat" to semi-subsistence farming,
especially in remote areas.
� For most smallholders, there are relatively low returns to farming--less than $1 per
person per day. This low return to labor makes it difficult to generate surpluses that
could be saved or invested in improved technologies, and thus contributes to a cycle of
poverty.
� Market liberalization and commercialization increases the need for cash, other liquid
assets, or credit. The scant amounts of cash available and the high returns to cash suggest
that returns to and demand for credit will likely be robust.
� Improved infrastructure should, in theory, lower transport and transaction costs, and
increase the commercial orientation of smallholders. However, lower costs might not
compensate for the drop in output prices and rise in input costs that have hurt
smallholders. Thus, the incentives to diversify into higher-value crops using improved
technologies are limited, and further constrained by the lack of credit.
� Remoteness leads to less land under cultivation, lower returns per household member and
lower returns to land. This indicates the welfare costs associated with remoteness.
Improved transport should benefit remote households and encourage more intensive and
extensive utilization of land, and higher income-earning potential.
ii
� Improved infrastructure may also increase confidence in markets. Such confidence could,
over time, reduce the tendency of smallholders to allocate scare resources to the
production of staple foods for own consumption. The combination of less remoteness
and increased confidence in the market can lead to significant improvements in
household well-being. The food security constraint severely lowers household returns,
and its "removal" should allow smallholders to produce higher-value crops instead of
traditional low-value staple crops. Over time, these households might adopt improved
technologies and possibly increase their demand for land.
� HIV/AIDS and other serious illnesses will likely continue to negatively impact household
welfare. Labor reductions lead to less land under cultivation and more staple foods as a
share of total crops. Purchased inputs increase as a share of the total value of production,
indicating some substitution of purchased inputs for scarce labor.
Land reform that increases access to land can not be expected to help most smallholders.
The "land as collateral" argument, whereby access to land helps guarantee loans, needs to be
examined, and alternative means of financing (e.g., micro-finance, contract farming) should be
considered. The absence of affordable transport and more dependable markets are interrelated,
and make fertilizer reform per se a questionable exercise. How would the reform of fertilizer
markets affect the price and availability of fertilizer? Access to fertilizer appears to be more
important than its price. Improving transport and access to markets is necessary for improving
rural welfare, but complementary investments in labor and financial markets are also required.
New opportunities from higher value crops and conservation farming hold out some promise for
smallholders, but high costs of entry and are a constraint.
Based on the results of the smallholder household model and other analysis, the authors are
not very optimistic about the potential for smallholder agriculture-led poverty reduction,
especially in the short term. A major transformation is required for smallholders' agricultural
potential to become agricultural reality. Large-scale investments are needed in research and
extension, market and transport infrastructure, capacity-building for individual farmers and
groups of farmers, and more. These investments require time to reach fruition. Policies to
stimulate non-agricultural economic activities in rural areas should also be explored. Although
high seasonal labor demands create bottlenecks at key times, there is an opportunity to spread
labor more evenly through the year as a means to increase household incomes. In all cases, a
more comprehensive and holistic approach to rural development is needed, not just an
agricultural or commodity-specific strategy.
iii
1. INTRODUCTION
Background and Objectives
Zambia's recent Poverty Reduction Strategy Paper (PRSP) views the agricultural sector
as a critical pillar for rural growth and poverty reduction. "Agriculture is given the highest
priority in the PRSP for diversifying production and exports, creating employment, increasing
incomes, and improving food security (World Bank, 2002a, p.7)." To stimulate broad-based
agricultural growth the PRSP proposes, inter alia, improving the land tenure system, liberalizing
fertilizer markets, and expanding rural infrastructure (notably roads) through a combination of
policy reforms and investments.
There is optimism about the potential for Zambia's agricultural sector to be an engine of
growth (see, for example, World Bank, 2003b). This optimism is based on the country's
abundant and underutilized land and water resources, and rural labor force. Although Zambia's
agricultural sector has significant untapped potential, it is characterized by structural problems
and risks that constrain realization of that potential. The smallholder sector, which accounts for
over 90 percent of rural households, faces unique constraints.
Most rural residents are smallholders who derive most of their income from agriculture.
There is a general lack of non-farm, non-agricultural activities in rural Zambia (Milimo, Shilito,
Brock, 2003; Skjonsberg, 2003). Given the importance of agriculture to the livelihoods of rural
Zambians, most poverty reduction strategies focus on improving income generation through
agriculture. Since smallholder agriculture predominates throughout the country, the impacts of
public actions on smallholders will, to a large extent, determine effectiveness in reducing rural
poverty.
The objective of this paper is to investigate the poverty-reducing potential of smallholder
agriculture in Zambia, in light of suggested policy changes and investments. This paper was
prepared as part of the Poverty and Social Impact Analysis (PSIA) work carried out to inform the
Zambia Country Economic Memorandum (CEM). Following PSIA guidelines, this study
combines both quantitative and qualitative analyses from a variety of sources (see World Bank,
2003a). We first review major changes that have taken place in the agricultural sector in Zambia
since the early 1990s, with particular focus on changes in the smallholder sector. Results from
an analysis undertaken for the 1994 Zambia Poverty Assessment using a smallholder household
model are also reviewed. The model has been updated to examine the current opportunities and
constraints facing typical smallholder households. We examine impacts of proposed reforms and
investments for land, fertilizer and infrastructure on typical smallholder households. In addition,
we examine impacts of HIV/AIDS on smallholder households. An important consideration is
whether smallholder agriculture can generate sufficient surpluses to generate a virtuous cycle and
lift rural households out of poverty
The model captures many of the major features of the smallholder decision making
environment, but other important factors are not easily modeled. Thus, we also examine findings
from other quantitative and qualitative studies and discuss their implications with respect to the
model parameters, scenarios, and results. The paper concludes with policy implications and
suggestions for future research.
Structure of the Agricultural Sector
Structural problems are manifested in the dualistic structure of the Zambian agricultural
sector.1 Dualism has been influenced by historical factors � which are beyond the scope of this
paper (see Jensen, 1977). In addition, policy incentives and investments in public infrastructure
have contributed to the differentiation in crop and livestock enterprises among different types of
farmers.
The overwhelming majority of Zambia's agricultural producers are smallholders, who
use simple technologies (hand hoes and oxen) and cultivation practices (minimal purchased
inputs such as hybrid seed or fertilizer). Productivity on smallholder farms tends to be relatively
low. They mainly produce rain-fed maize, groundnuts, roots and tubers, primarily for own
consumption on five or fewer hectares (see table 1). Most lack access to functioning input and
output markets and support services.2 It is estimated that only 40 percent of smallholders sold
crops during the 1999-2000 season (World Bank, 2003b).
At the other extreme are large scale commercial farms using modern inputs with access to
global input and output marketing chains. In some cases, large commercial farms are vertically
integrated with agro-processing (Francis, et al., 1997; World Bank, 2003b)3 More commercially
oriented medium-sized ("emergent") farmers use animal traction, hybrid seed and fertilizer to
grow rain-fed crops. Some attempts are being made to introduce micro-irrigation for emergent
farmers and some smallholders.
Zambian producers can also be differentiated by the types of crops they produce.
Smallholders tend to produce low-value-to-weight food staples, including about 60 percent of the
country's maize, 90 percent of sorghum, 85 percent of groundnuts and virtually all the cassava
and other starchy staples (see table 2). Smallholders produce some higher-value cash crops, such
as cotton, tobacco and paprika, and have small livestock, primarily poultry and pigs for home
consumption. The differentiation of crops and livestock by producer type is largely a function of
the higher capital requirements for higher-value enterprises. Emergent and large-scale producers
generally have better access to markets and infrastructure. Land tenure arrangements also
1The structure of Zambia's agricultural sector is dualistic with respect to technology, cultivation practices and
market orientation, crops produced, locational factors such as agro-ecological conditions and proximity to transport
and markets, and the distribution of land and other household assets (e.g., human capital and financial assets).
2The GRZ's "Agriculture Commercialization Programme" (ACP), which was designed to implement the
agricultural component of the PRSP specifically notes problems with input and credit markets, which "were not
adequately addressed in the PRSP (GRZ, 2001, p.7)."
3Emergent and larger-scale farmers use hired labor and contribute to national agricultural production and exports,
but their relative small numbers and reliance on labor-saving technologies limit their contributions to poverty
reduction. However, as mentioned in the final section on suggestions for additional research, more empirical
information is needed on the labor generation and linkages of emergent and larger scale commercial farmers.
2
differentiate producers and affect access to credit. Commercial farmers have title to lease state
lands, while smallholders produce on lands under customary tenure. Customary land is
distributed by chiefs, which can lead to more or less land access and security, depending on
factors such as types of marriage/social organizational systems and ties, legth of residence, and
the integrity of chiefs.
Crop and livestock patterns are influenced by agro-ecological zones. Zambia can be
divided into three major agro-ecological zones (see table 3). Region II, including most parts of
Central, Southern, Eastern, and Lusaka Provinces has the most favorable agro-ecological
conditions in terms of rainfall, soil quality, and absence of tse-tse fly that allow for a diverse mix
of crop and livestock enterprises.4 Besides agro-ecological factors, another important locational
factor is proximity to the "line-of-rail" and major urban centers and markets. The railroad line
runs from copper mine areas in the Copperbelt to Southern Province. As mentioned previously,
most commercial farms are located close to the line-of-rail, whereas many smallholders are
located in more remote areas with less favorable agro-ecological conditions. Region I, has drier
conditions than Region II and has experienced numerous droughts since the early 1990s.
Eastern and Northern Provinces contain the most rural households (see table 4), but
farming systems greatly differ in the two areas. Eastern Province, known as the "maize basket,"
is characterized by relatively more productive smallholders and widespread livestock production,
and has higher population and less land abundance than other areas of the country. Northern
Province (Region III) has higher rainfall and more humid conditions and problems with soil
acidity and declining soil fertility5 Land is relatively abundant in Northern Province and shifting
cultivation (e.g., slash and burn) has been widespread until recently, and is still the norm in some
areas. Unlike other areas of Zambia, cassava (and not maize) is the major food staple. Conditions
in Southern Province are fairly similar to Eastern Province (tradition of maize and livestock
production), except for less rainfall. Southern Province contains the third most rural households
and is of special interest because there are relatively abundant land and water resources.
Southern Province is an area where cotton production has expanded in recent years, and there has
been some growth in the number of emergent and commercial farmers. Western Province is the
driest area of Zambia. Northwestern Province, in contrast, has high rainfall and conditions are
more similar to Northern Province. Central and Lusaka Provinces have agro-ecological
conditions similar to Eastern and Southern Provinces, and are located along the line-of-rail.
4Average yields vary substantially among provinces. In general, yields are higher in agro-ecological zone II for
most crops. Higher yields are due to more favorable agro-ecological conditions and the fact that much of this zone is
located along the "line-of-rail." This area has better access to infrastructure, higher use of inputs, and a higher share
of medium- and large-sized farms using improved technologies.
5Northern Province accounts for 20 percent of Zambia's land area and about 14 percent of the population. The
poverty rate is about 80 percent, and non-agricultural employment is virtually non-existent. There is relatively high
rainfall, but the highly acidic soils need lime in addition to fertilizer. Since the early 1990s the province's maize
output has fallen to less than one-half of pre-reform levels. In the absence of fertilizer, some smallholders have
reverted to "slash-and-burn" and growing cassava. Coffee and sugar plantations are relatively minor, yet important.
Sheshami and others (2002) identified the major constraints on agricultural production in Northern Province as poor
infrastructure and market integration, and the lack of irrigation, safe water and sanitation and credit.
3
On average, rural households have access to approximately three hectares of land (see
table 4). This amount of land should allow households, in most areas, to produce enough food
staples and other foods to cover consumption needs. Access to land and the quality of land,
however, varies by Province. The lowest mean access to land is in Northwestern Province and
the highest is in Central. Eastern and Southern Province have areas (notably near line-of-rail or
towns) that are relatively densely populated and average land availability per household is often
insufficient to cover food staple consumption needs.
Constraints to Smallholder Production
In the past, price and institutional factors made maize production an economically viable
activity for farmers throughout the country. Subsidized maize and fertilizer prices and pan-
territorial pricing, maize-biased public agricultural research, extension and credit systems all
contributed to maize being produced in areas not particularly suited for it. Past policies
encouraged dependence on maize as a staple food and on government institutions for marketing,
extension and credit. Over-cultivation led to decreasing soil fertility, natural resource
degradation and inefficient use of human resources. Deininger and Olinto (2000, p.3) note:
"Producer subsidies for fertilizer led to the extension of maize cultivation into unsuitable areas
which increased vulnerability to drought, distorted factor prices, and biased the direction of
research away from high value export crops to staples with low profitability."
Despite its low value, economic logic influences smallholder decisions to produce maize,
primarily for home consumption. "The low value to weight ratio of maize adds to the
profitability of producing for own or local consumption, while at the same time restricting
opportunities for sale to urban consumers with good market access (Copesake, 1997, p.24)." As
long as markets are poorly integrated and transaction costs high for low value food staples (like
maize), in remote areas the costs of such staples is relatively high.6 In fact, there is evidence that
most smallholders practice a "safety-first" strategy to allocate household assets first to producing
enough food staples for own consumption and only adopt a profit maximizing strategy once food
requirements are fulfilled (See Alwang, Siegel, and Jorgensen (1999)).
Deininger and Olinto (2000) highlight the constraints to increased productivity and
adoption of higher-value crops faced by Zambian smallholders. They note:
i. Purchased inputs and adoption of improved technology can be profitable, but non-
price factors--such as the lack of markets for timely buying and selling, and the lack
of support services like credit and extension--hinder increased productivity.
ii. Under conditions of land abundance, access to complementary productive assets is a
key constraint on productivity, the amount of land cultivated, the use of credit and
purchased inputs, and adaptation to climatic risks.
6 Loy and Wichern (2000) examined regional and international integration of maize markets in Zambia to test
whether the liberalization policy has led to integrated food markets. Results indicate that regional maize markets are
integrated. However, the level of regional and international market integration is low and has not increased over
time. Transaction costs in these markets are still high.
4
iii. "Constraints facing rural producers in Zambia are still related more to market access
and the ability to obtain necessary inputs in a highly volatile economic environment
rather than application of more productive technology. At the same time, this calls
for a thorough assessment of the technology available and the messages that are being
disseminated through public technical assistance (p.15)."
Deininger and Olinto (2000) conclude: "this reinforces the importance of providing
public goods, in addition to price policies, to bring about a sustained supply response much
needs to be done to address non-price related constraints to agricultural production, generate and
disseminate technology, and thus help rural producers make better use of the resources at their
disposal." For example, confidence in markets is a critical non-price factor. In the past, market
imperfections such as the undependable and untimely availability of inputs and basic
consumption goods have undermined market confidence, leading to conservative decision-
making. Also, the lack of credit and insurance markets for smallholders slow the adoption of
new technologies and/or new enterprise mixes. In addition, new technologies and crop
enterprises often have relatively high initial investment costs, creating barriers to entry. These
problems are exacerbated by the low value of surpluses and savings generated by smallholder
agriculture, which preclude new investments in improved technology.
Risks Faced by Smallholders. Structural problems are exacerbated by numerous risks
facing agricultural households. Risks include weather- and price-related fluctuations in harvests
and crop prices, death or distress sales of livestock, and human health risks that affect labor
availability. Smallholders tend to have greater exposure to risks and less ability to manage the
risks.
Risk factors also affect smallholder decisions about crops, technology adoption, input use
and productivity (Ruiske, et al.,1997). For example, since 1990, about three out of every 5 years
have essentially been drought years in Zambia (Bwalya, 1999). Price and yield risks may
undermine an emerging liberalized fertilizer market and "efforts to increase fertilizer use on
small farms are plagued with fundamental problems such as climatic risk, a dearth of technology
packages that are farmer tested to be profitable and risk-decreasing, especially in less-favorable
agro-climatic zones; lack of technical and management skills and information on application
rates and agronomic methods to increase efficiency and profitability of fertilizer use; high
transport costs; underdeveloped credit markets; and risky output markets. Sustainable strategies
for increasing fertilizer use must address these fundamental problems (Ruiske, et al., 1997, p.3)."
Uncertainty with respect to input and output markets also influences decisions about crops,
technology adoption, and input use (Gordon, 2000). Because of widespread risk and uncertainty,
many Zambian smallholders have retreated to semi-subsistence farming systems in recent years.
Smallholders also face risks related to human illnesses, and diseases and pests that affect crops
and livestock. The proliferation of HIV/AIDS7 has highlighted health risks and impacts on
household labor and dependency ratios, but rural Zambians have long suffered from many
7Research indicates that one in every four Zambian adults is HIV positive. The high incidence in rural areas has
become more than a public health concern placing demands on limited household resources, exacerbating labor
constraints, and lowering productivity (World Bank, 2003b; Pillai, Sunil, and Gupta, 2003).
5
illnesses, which affect labor availability, household composition and household production
decisions.
Changes in the Agricultural Sector since the 1990s
Performance of the agricultural sector since reforms began in the early 1990s has been
uneven. Reforms aimed at market and trade liberalization were expected to benefit smallholders,
but evidence shows that the reforms harmed many such households, particularly those in remote
rural areas (Evans, 2001). In particular, the collapse of markets for credit and inputs that
followed liberalization seriously hurt smallholders. Reductions in government expenditures on
transport and communication infrastructure compounded the problem. The collapse of
expenditures on transport and communications appears to have exacerbated difficulties faced by
rural households. This collapse discouraged the private sector involvement, and, as a result,
many poor farmers were unable to exploit their agricultural potential. (McCulloch, Baulch, and
Cherel-Robson, 2000, p. 31-32).
Many of the core weaknesses of Zambian agribusiness lie outside its own confines, in the
forms of weak overall infrastructure within the country, an unstable and unpredictable exchange
rate, extremely high costs of finance, and weak capacity of government to address trade-related
anomalies and problems (World Bank, 2003b, p.xiii). For example, throughout the past 15 years,
Zambia has experienced macroeconomic instability (World Bank, 2002b). Also, its public
institutions for agricultural development (such as the Ministry of Agriculture and Cooperatives
(MAC)) are weak and fragmented (World Bank, 2002b).
Many of the changes in the agricultural sector during the 1990s were the results of policy
reforms, while others stem from factors such as changing conditions in international commodity
markets and recurring droughts. These changes have had different impacts on technology
adoption, enterprise mixes, profitability, and market orientation on different types of farmers
(World Bank, 2003b). Below we highlight some of the changes that are relevant to smallholders
- and our subsequent analyses.
Policy shifts
Major policy shifts in the 1990s include the dismantling of state institutions for marketing
and distribution of agricultural outputs and inputs, the abolition of producer subsidies, the
liberalization of trade in food items, and the introduction of market-determined input and output
prices (see GRZ, 2001; World Bank, 2003b, p. 64-66 for details). It was hoped that these policy
shifts could benefit producers of all sizes, but this has not been the case (janyne, et. Al., 1991,
Box 1: Agricultural Performance Since the mid-1990s.
World Bank (2002b, p.5) highlights the following changes in performance of the agricultural sector since the mid-
1990s: "The food production index fell from 130 in 1995 (1981=100) to 128 in 2001. Agricultural gross domestic
product barely rose from 395 billion kwacha in 1995 to 419 billion kwacha in 2001. During this period, agricultural
production fluctuated significantly around a stagnant mean. Agricultural productivity continued its long-term
decline. Moreover, agricultural incomes fell from 70,000 kwacha to 62,000 kwacha in 2001. {However}...
Agricultural exports increased dramatically, particularly of nontraditional exports such as horticulture and
floriculture. The total value of agricultural exports rose from USD$55 million in 1995 to USD$ 160 in 1998,
although it declined to USD$125 in 2001."
6
SGS Zambia, 1999; Milimo, Shilito, and Brock, 2000). See box 1. Many cite the absence of
private service providers and structural constraints to entering agricultural input and output
markets as a major source of ongoing market imperfection and uncertainty (Smale and Jayne,
2004).
Land Policy Reform Initiatives8
Land reform has been on Zambia's political agenda since independence in 1965. More
recently the PRSP highlighted the need to amend the 1995 Land Act.9 The Draft Land Policy is
an attempt to revise the act, but this initiative has stalled. Pre-independence the most fertile
lands were designated "Crown Land" and reserved for the white settlers, while the less fertile, or
"Reserve and Trust Lands," were made available to blacks. Upon independence, the government
repossessed and re-leased most of the land. Currently all land in Zambia is vested in the
President, who holds it in perpetuity for and on behalf of the people of Zambia. Ninety-four
percent of the land is under customary tenure, distributed by the chiefs. Customary tenure
comprises more than 70 ethnic traditions and four types of marriage/social organizational
systems, which affect access to land. The remaining 6 percent, which are held under state title,
includes cities and towns, game reserves, mining areas and land set aside for commercial
agriculture. Commercial farming areas constitute about half the state-titled land.
Considered a land-abundant country, the under-utilization and inefficient use of Zambia's
land base is perceived to be a problem. The PRSP explains the under-utilization of land is due,
in part, to complicated procedures for obtaining title deeds and the lack of incentives to utilize
(or sell) idle land (such as collectable land taxes). One factor influencing the re-evaluation of
land tenure practices is the desire by GRZ to provide more land for commercial and emergent
farmers. Demand for land by disenfranchised Zimbabwean commercial farmers is one factor
driving this search for ways to better utilize Zambia's idle agricultural natural resources. One
suggested solution is the selling of 1,000-hectare "farm blocks" to qualified local
businessmen/farmers to develop commercial agriculture. This would not require immediate
reform of the customary land system, but could rather quickly get un-utilized lands under
production and add to government revenues. The revenues from land sales could be used to
support smallholder agriculture. The reform of traditional agricultural land tenure practices
should strive to provide greater equity and efficiency in land use (especially equal access for
women, new residents, and other groups whose rights to land are not protected in many cases).
8 Information on land issues is largely drawn sections of the Zambia PSIA draft report.
9 The 1995 Land Act tried to make customary land arrangements more efficient and equitable. The act introduced a
Land Tribunal to address failures of customary land dispute-resolution systems and provide assistance for those who
could not afford the courts. The tribunal was meant to balance the power of the chiefs; yet it has been moribund, and
its impact has been marginal. The 1995 act also introduced a 30 percent community set-aside for women and other
excluded groups and individuals.
7
Land quality issues are gaining prominence in Zambia. Despite claims of
underutilization of land, in some rural areas with land scarcity there has been a trend toward
continuous cultivation and reduced fallow. This has led to increases in soil acidity and decreases
in fertility (Haggblade and Tembo, 2003). To maintain productivity, many need to apply lime
and increasing amounts of fertilizer over time. Conservation farming, which we discuss later, is
an effort to preserve and enhance soil quality and improve productivity and profitability � partly
through more efficient fertilizer use.
Fertilizer Policy and Prices
The price of fertilizer more than doubled during the 1990s and fertilizer use fell by about
one-half (Howard and Mungoma, 1996; Jayne, et al., 1998; Kherallah, et al., 2000, GRZ, 2002).
See box 2.
Box 2: Fertilizer Prices and Maize Production
According to participatory studies in rural Zambia, increased fertilizer prices were the most commonly mentioned
constraint on agricultural production in all agro-ecological zones in all years since 1992. "Maize is now perceived a
low value crop by farmers who can no longer afford the heavy dose of fertilizer which is necessary to make maize
cropping productive enough to be profitable (Milimo, Shilito, and Brock, 2000, p.15)." In addition, since the
beginning of market liberalization, farmers in all zones perceived that delivery of inputs had become less
dependable.
The data show, however, that fertilizer prices have not risen substantially relative to maize prices. In 1993, the
national maize/fertilizer price ratio was 0.41 (Alwang and Siegel, 1994); recent data from CSO show this ratio to be
0.51 in Eastern Province, 0.42 in Northern Province and 0.53 in Southern Province. While fertilizer prices have
risen, maize prices have risen more than proportionally, leading to increased profitability of fertilizer use on maize.
Despite this increased profitability, fertilizer use declined nationally between 1990 and 2000 (World Bank, 2003).
But, others note that the official prices used to calculate maize/fertilizer price ratios are misleading, since few can
buy fertilizer at official prices.
Timely access to fertilizer can be more important than its cost. Government-supplied
inputs, although less expensive, have a history of not being provided on time. Late application
of fertilizers can result in significant reductions in yields. Many studies highlight the lack of
timely inputs as a major constraint for smallholders (e.g., Gordon, 2000; Keyser, Helsop, and
Abel, 2001; Milimo, Shilito, and Brock, 2000; Skjonsberg, 2003; Haggblade and Tembo, 2003).
With fewer than 20 percent of Zambian smallholders using fertilizer in 1999-2000, GRZ
feared that discontinuing fertilizer subsidies would exacerbate food security problems (Mwape,
2004).10 "There are serious concerns over private traders' willingness to deliver inputs on credit
to resource-poor farmers. According to this view, government fertilizer and credit distribution
are indispensable for promoting smallholder agricultural productivity and growth (GRZ, 2002,
p.iii)."
Recent programs to improve smallholder access to fertilizer are MACO-FSP (50%
subsidy program) and MCDSS-FSP (100% grant). These programs were each supposed to reach
10An institutional analysis of fertilizer markets in Zambia was undertaken as part of the PSIA (see Mwape, 2004).
8
about 200,000 households in 2002/2003.11 About to two-thirds of the planned number of
households received assistance in 2002/2003, but problems persist, such as: late delivery, limited
program benefits per household, seemingly inefficient use of subsidized fertilizer, reselling of
subsidized fertilizer and/or use on other crops, and high costs to maintain the program (even if
the major share of costs are borne by donors and NGOs).
It has been argued that continuation of subsidized fertilizer programs inhibits private sector
development of fertilizer production, purchases and sales (GRZ, 2002; Mwape, 2004). In fact,
some contend that market privatization, with appropriate investments in infrastructure to lower
transport costs and make remote areas more accessible, could benefit smallholder farmers more
than current policies that subsidize fertilizer prices (GRZ, 2000). In addition to appropriate roles
of government and the private sector, questions about the transition to a private-sector-led input
supply system must be addressed. In addition, it is important to recall that efforts at lowering
transaction costs through road construction and improved communication and storage
infrastructure can take many years to materialize. With all these inter-related factors, it is
questinable what liberalization of fertilizer markets alone can achieve (see box 3).
Box 3: Inter-relationships Among Constraints in Fertilizer Supply
Constraints of fertilizer supply and access have been attributed to: limited purchasing power, poor
road and market infrastructure, inadequate private sector capacity especially in remote areas, late
delivery of inputs, reductions in Donor support for fertilizer subsidies, and reduction in domestic
production of fertilizer. Ironically, Zambian smallholders are reported to overuse fertilizer. Mwape
(2004) details many of the inefficiencies and high costs from lack of a coherent fertilizer policy. It
is clear that fertilizer supply and prices are affected by numerous factors outside the realm of the
Ministry of Agriculture and Cooperatives. It is thus important to have a multi-sectoral and inter-
Ministerial approach. That includes investments in infrastructure combined with improved
extension messages, and incentives for more efficient use of fertilizer.
Rural Infrastructure and Transport Costs
Zambia is a large landlocked country with relatively unreliable and expensive access to
ports; be they in Tanzania or Mozambique (or South Africa). Smallholders are dispersed
throughout the country, many located far from the line-of-rail, which is a major transport artery.
The lack of direct port access, the poor state of rural infrastructure, and the physical dispersion of
smallholders all make the marketing of agricultural inputs and outputs expensive (Copestake,
1997; SGS Zambia, 1999; Keyser, Helsop and Abel, 2001; GRZ, 2002; Mwape, 2004).
Furthermore, many rural roads are in bad condition and often not passable during the rainy
season. Private traders therefore tend to concentrate their business to the line-of-rail and other
main roads, leaving farmers in remote areas without reliable service. These factors contribute to
high transaction costs and uncertainty about markets, and have had a negative impact on
11In 2003 MACO-FSP administered by FAO and assisted members of cooperatives who paid their required 50% of
input costs for a pack consisting of 4 50kg bags of urea, 4 50kg bags of D-compound, and 20 kg of maize seed.
Under MCDSS-FSP, administered by PAM, target households were poorer and affected by recent droughts. Their
pack consisted of inputs for 0.25 ha of a staple grain (maize or sorghum or millet), legume (groundnuts, beans or
cowpeas) and tubers (cassava or sweet potato) for home consumption.
9
Zambia's competitiveness in international commodity markets. For example, farmgate prices of
fertilizer can be about double the CIF price in many areas of the country (Mwape, 2004).
To some extent, past policies of pan-territorial pricing masked the problem of differential
access to transport infrastructure and services in the country. Since elimination of pan-territorial
pricing for fertilizer and maize, spatially differential prices of inputs and outputs have influenced
the distribution of different cropping patterns and cultivation practices.
During the 1990s GRZ made some efforts to improve major roads (primary and
secondary roads) and also has supported local and community efforts to improve and maintain
tertiary rural roads (World Bank, 2004).12 However, with such a large country and dispersed
country, there is still a long way to go to improve access to transport and markets for many rural
Zambians. Improving access to transport and markets and lowering transport costs are key
factors for improving market integration for staple food crops and higher value enterprises. In
addition, improved rural transport should help improve access to rural and urban labor markets,
and help improve conditions for development of non-agricultural activities.
Shifts in Production Patterns and Yields
Evidence suggests that the breakdown in maize marketing arrangements and the rising cost
of fertilizers led many smallholders - especially those in more remote areas - to shift to low-input
technologies and increased production of alternative staple crops for home consumption (Masters
et al., 1998; Milimo, Shilito, and Brock, 2000; World Bank, 2003b; Skjonsberg, 2003). This shift
has been quite pronounced in Northern Province, where agro-ecological conditions do not favor
maize production and transport costs are high (Francis, et al., 1997). See box 4.
Box 4: Changes in Agricultural Production Patterns
"In the early-to-mid-1990s, there occurred a noticeable shift in production patterns, with a reduction
in maize plantings and a corresponding increase in the production of an array of drought tolerant
cereals as well as crops that do not require much fertilizer (i.e., legumes). Some renewed interest was
taken in growing industrial crops (e.g., cotton) and there occurred an initial spurt of growth in non-
traditional agricultural exports. The elimination of maize meal and transport subsidies contributed to
the rapid emergence of many micro and small-scale milling and oil expressing
operations....Networks of private traders and framer-based marketing companies developed,
especially along the Livingstone-to-Copperbelt `line-of-rail' and in surplus producing areas
elsewhere. In contrast, in more remote locations, little private trading emerged and smallholder
farmers began to revert to more subsistence-oriented production patterns (World Bank, 2003, p.66)."
Maize continues to dominate agriculture, but its relative importance has lessened to some
extent (see table 5). Maize area planted declined somewhat since the early 1990s.13 Fluctuating
12Under the Road Sector Investment Program (ROADSIP) over 1,400 km of trunk, main, district, urban, and feeder
roads were rehabilitated, 6,400 km received periodic maintenance, and 1,500 of community roads were improved.
ROADSIP has also been an important source of employment (World Bank, 2004).
13Maize production increased at a 1.9% average annual rate from 1970 to 1989 and declined at a 2.4% average
annual rate from 1990 to 2000 (Smale and Jayne, 2004). Rural households use either local or hybrid maize seed.
10
climatic conditions, especially recurring droughts, contribute to variable yields (figure 1). The
area planted to maize declined by more than total production did, which indicates that yields
increased slightly over time.14
Patterns of production have shifted, but agricultural production is still concentrated in grains
and staple foods. Major shifts into cassava have taken place. Cassava requires no purchased
inputs and can produce good yields in a wide variety of soil-water conditions (including
drought). Its flexible planting and harvesting calendar make it one of the easiest crops for labor-
constrained households. There is some evidence from Zambia that many HIV/AIDS affected
households have shifted to cassava production (Nweke, Haggblade, and Zulu, 2004).
Land planted to groundnuts (which are widely grown throughout the country and used for
relish and oil extraction and have a relatively high value per weight), increased substantially
between 1993 and 2002 (figure 2). However, production increases have not mirrored the increase
in acreage (figure 3). Sorghum and millet plantings (mostly done by smallholders) increased
during the mid-1990s but leveled off and even declined since. Sorghum, which performs well in
drought-prone areas and stores well, is grown in many provinces, often inter-planted with local
maize varieties. Land planted to sunflowers, which are produced for sale to oil processing plants,
has fluctuated around a fairly constant level since 1997.
A major shift in crop production by smallholders has been the significant increase in land
planted to cotton (Boughton, et al, 2002). Much of the cotton production has taken place using
contract farming arrangements, whereby smallholders receive inputs on credit and extension
assistance along with guaranteed output markets. Over the past few decades, cotton production
by smallholders has gone through periods of expansion and contraction. Cotton production
increased from 20,000 mt in 1994 to about 100,00 mt by 1998 and evened off at about 80,000
mt. During 1998 to 2000, cotton exports were Zambia's major agricultural export. Cotton
planting is sensitive to prices; unusually high international prices in 1998 preceded a spike in
land planted and production in 1999. Cotton prices have continued to fluctuate in recent years
and there is evidence of a long term decline in real prices (Baffes, 2004).
As mentioned, there were significant increases in production of nontraditional exports such
as horticulture and floriculture. There were also significant increases in coffee production and
exports in recent years, mostly by commercial farmers. In most cases, commercial farmers and
some emergent farmers have been the major producers of these nontraditional exports. A notable
exception is paprika, which like cotton, has largely been produced by smallholders under
contract farming arrangements. In most cases declining or stagnant international prices have
constrained any significant increase in the value of agricultural exports or increased incomes for
producers.
14Note that in some cases we report that average maize yields have slightly increased and in others that it has
slightly decreased. Different publications have different starting and ending periods, and data is of questionable
quality, but nationally there have not been significant changes in average yields.
11
Shifts in Value of Production and Shares of GDP
Since the mid 1990s, the contribution of agriculture to national GDP increased from
about 15 to 20 percent (World Bank, 2002b). The contributions of different crops and livestock
to agricultural GDP have also changed (table 6). Maize still is the largest component of
agricultural GDP, but it declined from 26 percent in 1994-5 to 19 percent in 2000-1; while the
share of roots and tubers increased from 5 percent to 14 percent. Other notable changes are the
increases in the value of cotton (from 15 to 19 percent) and high-value cash crops such as
horticulture and floriculture products (from 8 to 14 percent), and the decrease in the share of
livestock (from 30 to 20 percent). The large-scale decline in livestock numbers (especially work
oxen and cattle) during the 1990s is a major reason for this decline in the value of livestock.
Decline in Livestock Numbers
Oxen use and mechanization increased some since the early 1990s in Central and Eastern
Provinces. But, oxen use has declined dramatically in Southern Province due to East Coast
Corridor disease (FAO, 2002). Official agricultural census statistics indicate that draught
animals numbers declined by 50 percent between 1994 and 2000. A similar decline was
recorded for other types of livestock. Reasons include repeated droughts, spread of the tick-
borne corridor disease and deterioration in the availability and affordability of veterinary care
(Copesake, 1997). Oxen use is virtually non-existent in Northern Province. Nationally, it is
estimated that only about 25% of smallholders use oxen (Haggeblade and Tembo, 2003).
In most areas virtually all farm labor is provided by household members. Labor markets
are consequently thin (Mwape, 2003). Oxen use can help raise average yields and save labor at
critical points in the cropping cycle. Such innovations can help smallholders plant more land
without resorting to hired labor, or be constrained by household labor availability. Increasing
land under production by smallholders is one of the keys for agricultural growth to be poverty
reducing.
Emergence of Private Sector Outgrower Schemes
Some evidence suggests that the private sector can fill the void left by the public sector
and that smallholders can adapt to new market conditions. Smallholder outgrower and contract
farming schemes proliferated in the 1990s, after the public marketing and agricultural finance
systems collapsed. This private-sector led initiative to fill the voids has certain appeal. In
addition to cotton, schemes have been set up for tobacco, paprika, sugarcane, castor, and for seed
production (Copestake, 1997; SGS Zambia Ltd., 1999; Keyser, Heslop, and Abel, 2001).
Many of these outgrower schemes are in danger of failing. A major problem is that
Zambia's legal system does not protect or enforce buyers' contracts; instead it protects the rights
of farmers not to honor their obligations. "Side selling" of outputs by smallholders is common15,
as are the resulting high rates of loan delinquency and non-payment. Buyers have grown weary
of such contractual arrangements and are either looking for other ways to encourage production
15Side-selling is when contracted farmers do not sell their outputs to the provider of inputs on credit, as stipulated.
12
and purchase crops or are simply pulling out of the business (SGS Zambia Ltd., 1999; Keyser,
Helsop, and Abel, 2001; Boughton, et al.,2002; Parker 2003).
Lonhro, the major private sector firm involved in contract farming for cotton, withdrew
from Zambia in 1999 due to problems with high credit defaults and the lack of legal sanctions for
defaulters (Boughton, et al, 2002). Dunavant Ltd. took over Lonhro's cotton business and
insituted a new system called "Dunavant Distributor System", whereby independent agents
contract with the company to receive inputs on credit and deliver them along with extension
services to farmers. Agents' earnings are a function of credit recovery, so they have an incentive
to provide quality inputs on time with extensionand to have high yields and repayment rates
(Boughton, et. al., 2002).
Changes in Kefa Village Over the Past 25 Years: Findings from a Participatory Study
For the World Bank's 1994 Zambia Poverty Assessment, a background paper by
Skjonsberg (1994) examined living conditions in a village in Eastern Province using a serious of
semi-structured interviews with residents.16 For the PSIA, Kefa Village was revisited and
changes documented (see Skjonsberg, 2003). The findings presented below are interesting, and
in most cases they corroborate the storyline presented above on changes taking place in rural
Zambia since the early 1990s. In particular, there are additional insights about changes taking
place and conditions in a typical village in Eastern Province.
Major findings from Kefa Village include:
(a) Demise of the GRZ extension system that provided farmers with technical
support, subsidized credit, seeds and fertilizer, and the lack of public or private
sector to replace these services.
(b) Land issues tend to be debated, but land reform per se is not an issue for the
community (only 25 percent of interviewed residents said they did not have
enough land).
(c) Land is not equally distributed (advantaged groups include clan members,
wealthier farmers and long-time residents).
(d) Soil fertility problems are increasing the need for fertilizer and improved seeds.
However, farmers do not have cash or credit to buy improved seeds and fertilizer.
(e) Seeds and credit are sometimes available from agents of the cotton and tobacco
buyers.
(f) Seeds and fertilizer are rarely available in a timely manner.
16The 1994 paper was an update of an extensive study undertaken in the 1970s and documented in a book: "Change
in an African Village: Kefa Speaks" (see Skjonsberg, 1979).
13
(g) On average, a household cultivates about 1 ha maize and 0.4 ha groundnuts,
primarily for home consumption. The major binding constraint tends to be
available household labor, not land.
(h) Livestock numbers, especially cattle, populations are declining due to drought and
illness.
(i) Small livestock (e.g., chickens and goats) are gaining in importance.
(j) Water scarcity is a major problem.
(k) Lack of fertilizers is claimed to be the biggest problem. When people say that
there is a "lack of fertilizer", this includes the lack of suppliers and the high cost
of fertilizer when available.
(l) GRZ programs for subsidized fertilizer do not reach Kefa farmers or reach them
in insufficient quantities.
(m) Lack of fertilizer and water scarcity are driving cropping choices. Cotton
production has increased because it needs less fertilizer and water, and because
and agents for cotton processors tend to provide advances of inputs and
guaranteed markets)
(n) There is also a lack of output markets. "One big problem that we all have is where
to sell our produce. There is no market so instead we just have to sell it one by
one."
(o) Increased problems with malaria, HIV/AIDS, and orphans.
(p) Increased dependence on assistance from donors and NGOs (church groups).
(q) Persistence of beliefs in witchcraft and fatalistic attitudes.
The findings from Kefa Village indicate that there are many factors that influence the
opportunities and constraints faced by smallholders., including important social and cultural
factors.
14
15
2. THE SMALLHOLDER MODEL
Model Description
As part of the World Bank's 1994 Zambia Poverty Assessment, a stylized linear
programming (LP) model was constructed to better understand the economic opportunities and
constraints of "representative" smallholder households. This model was used to measure
impacts of changes in policies and investments on smallholder cropping patterns and cultivation
practices, and subsequently on households' well-being outcomes (see Alwang and Siegel, 1994;
Alwang, Siegel, and Jorgensen, 1996). A strength and weakness of the household model is its
simplicity and ease with which parameters can be revised.
The model is a single-period linear programming (LP) model of smallholder households
in rural Zambia. It is comprised of an objective function, model activities, and constraints for
different resources, with labor disaggregated by month. 17 See the annex for a formal
presentation of the general LP model.
The model maximizes annual returns from agricultural production given fixed assets,
subject to the technologies and constraints. The model only considered labor incomes from on-
farm agricultural activities. Although could be expanded to include off-farm wage labor
activities.18
Production technologies (i.e., cultivation practices) are reflected in crop budgets.
Constraints include access to land, household labor availability, and household food security
objectives. In initial model runs, land was not considered to be a binding constraint, but for
different scenarios, the availability of land is constrained. The food security constraint is
modeled by applying a "safety-first" condition to the household's objective function, whereby
the household maximizes net income only after first producing enough food staples for their own
consumption. This food-security constraint is relaxed in different model runs to evaluate the
opportunity cost of household concerns about food security.
There is a lack of consistent data on Zambian agricultural production (see box 5). The
smallholder model combined census data (information on household composition, land holdings,
economic activities), socio-anthropological data (time-use studies of household members, gender
distribution of labor for different activities), with farm budgets for different cropping activities to
17The household model analyzes a single 12-month period, disaggregated by month.
18Smallholders derive a majority of their income from agriculture. There is a general lack of non-farm, non-
agricultural activities in rural Zambia (Milimo, Shilito, Brock, 2003; Skjonsberg, 2003). Only considering
agricultural incomes focus more clearly focuses on agricultural incomes and the potential for agriculture, by itself, to
reduce poverty of representative rural households. Expanding the model to consider off-farm agricultural or non-
agricultural incomes is possible, but requires additional data (see Alwang and Siegel, 1999 for a similar smallholder
model that does consider off-farm wage labor).
generate the economic variables of interest (crop production patterns, household income, shadow
prices of land and labor). See Siegel and Alwang (1994) for details.
Box 5: Data Constraints in Zambia Limit Analytical Techniques
"A major challenge in understanding the performance and competitiveness of Zambian agriculture and
agribusiness stems from the unreliability of production-related agricultural statistics and the paucity of available
data pertaining to agro-industry. There are wide variations among production statistics reported by different
sources and a high level of uncertainty regarding the accuracy of data on plantings, production, and yields for
most crops. Data on livestock production and marketing is even more tentative. There are serious problems with
both the, methods and underlying data used to calculate agricultural GDP in Zambia. Available data on
agricultural trade probably represent a reasonable approximation of actual trade flows, although much of the
informal cross-border trade in food may not be regularly measured. The depth and quality of available
agricultural statistics are not sufficient to subject this data to sophisticated analytical techniques (World Bank,
2003, p.63)."
The model contributes to policy dialogue by providing quantitative estimates of the benefits
and costs of stylized changes in policies and investments to smallholders. That is, once a baseline
model is constructed and results analyzed, it is possible to change parameters to reflect different
scenarios and to quantify the impacts. In the context of the PSIA and CEM, the model was used
to examine impacts of changes in land availability (and, implicitly, land reform), fertilizer prices
and availability, infrastructure investments and transport costs, and HIV/AIDS on cropping
patterns, labor allocation and input use, land cultivated and incomes for "representative" rural
households'.19 See box 6.
Box 6: Using the Smallholder Model for Policy Dialogue with and Stakeholders
One of the appealing features of the simple LP based smallholder model is the ease with which it can
be updated, revised for which different scenarios can be examined. The technical data and price data
for the LP model can all be easily viewed and updated/revised. As such, it is hoped that there will be
follow-up consultation and dissemination activities in Zambia with stakeholders whereby the
smallholder model can be adjusted and tested with stakeholders in "real time" to generate additional
discussions and debates about how to most accurately portray representative households.
Main Findings from the 1994 Baseline Model and Scenarios
The main findings from the 1994 smallholder baseline model and different scenarios are
presented below:
� Access to land was not a binding constraint for households, because labor constraints
were binding. A typical household using hand-hoe could cultivate about 2.5 hectares at
most. Labor constraints during land preparation, weeding and harvesting limited more
extensive use of land resources (we assumed that only household labor was used in farm
production activities).
19Several factors limit the model's insights. These include a) the prevalence of non-price constraints, b) the
importance of household assets and bundles of assets in addition to cash flows, c) risks associated with fluctuating
prices and climatic conditions, and d) consideration of fixed investment costs in addition to variable input costs. To
incorporate these factors, a multi-period dynamic model would be required; this would require additional data. See
suggestions for further research at end of the paper.
16
� Oxen (and implicitly other labor-saving) technology reduced labor constraints and led to
more output per hectare and brought more land into production. When oxen cultivation
was assumed to be available, the maximum land used by the typical household increased
to around 4 hectares. Labor constraints were still binding at this higher level of land use.
� The impacts of price liberalization depend critically on access to infrastructure. More
remote households were adversely affected by liberalization because of increased
transport costs, while those with better access to infrastructure and thus to markets saw
their well being improve.
� Input supplies for improved technologies were an important constraint for households.
Model results showed that most smallholder households were more adversely affected by
late and irregular access to hybrid maize seed and agricultural chemicals than they were
by increased prices of such inputs. When inputs become only irregularly available,
households substituted labor and land for scarce modern inputs. As a result, yields
declined and households could devote less labor to other agricultural and non-agricultural
jobs.
� The impact of HIV/AIDS and other serious illnesses depend on household composition.
In female-headed households (FHHs), a loss of 20 percent of household labor leads to a
proportional loss in income. For male-headed households (MHHs), who initially had
more labor at their disposal, impacts of such illness were less severe. It must be
emphasized that the modeling of HIV/AIDS did not include out-of-pocket expenses
associated with the disease, psychic costs, and time spent caring for the sick; but the
model still shows a strong impact of the pandemic.
� Gender-differentiated agricultural activities (in some Zambian cultures, women and men
perform separate tasks) represented a real cost to the household, because they constrained
available household labor at peak periods. The net value of agricultural production was
20 percent lower for gender differentiating households, compared to non-gender
differentiating households.
Updating the Model in View of the Changes in the 1990s
The model was updated to reflect current conditions--technologies, cropping patterns,
input and output prices, and household composition, constraints and asset endowments). In
addition to the updating, the model was regionalized, since market liberalization has widened
regional price differentials, cropping patterns and cultivation practices across regions. The
technical input-output parameters for the various crop production activities were adjusted to
reflect new technologies and cultivation practices. The updated model was calibrated for stylized
smallholder households in Eastern, Southern and Northern Provinces. The model also includes
additional crop activities to represent the major crops produced by smallholder farmers in the
respective areas. For Eastern and Southern Provinces we include an option for cotton production,
while in Northern Province we include a cassava production activity. See table 7 for major crops
produced in Eastern, Southern and Northern Provinces.
Over the years, hand-hoe and oxen technologies have not significantly changed,
therefore, the technical input/output coefficients are assumed to be similar to the past
17
(Mwape, 2003).20 In provinces with significant oxen populations, both hand-hoe and oxen
technical coefficients are considered as available options to household at all levels of
management. In the areas with small numbers of oxen, hand-hoe technical coefficients are the
most applicable at all management levels. Since gender differentiation of agricultural tasks is not
as prevalent as in the past, differentiated male and female labor allocation by task was not
considered as in 1994 (Mwape, 2003).
20While use of hybrid maize, fertilizers and other modern inputs changed since the reforms of the 1990s, few
innovations in maize or other technologies have been made available to small-scale producers. Thus, the technical
relationships between input uses and outputs have not changed much. These technical input/output relationships
form the basis of the household model. The main difference between the updated model and the 1994 model is the
use of higher yields for local maize. For Eastern Province, we use yields of 1200 kg/hectare under hand-hoe
conditions, compared to 800 kg in the 1994 model. We justify this change based on data from Zambia (Mwape,
2003; World Bank, 2003 statistics, and FAO Agricultural Statistics data)showing average maize yields to be about
1450 kg/hectare. With less than 30 percent of farmers in Zambia using hybrid maize, local yields would have to be
substantially higher than 800 kg/hectare and our best information is that yields are approximately 1200kg/hectare
using relatively low technology. Conservation farming (CF) was not modeled due to lack of data.
18
19
3. UPDATED BASELINE SMALLHOLDER MODEL, POLICY
REFORM SCENARIOS, AND RESULTS
Updated Baseline Model
The updated baseline models considered the following crops, technologies and
cultivation practices (see annex, tables A and B for details on assumptions on crops and
input/output relationships and prices for Eastern, Southern, Northern provinces):
3.1 maize (local/hybrid seed, early/late planting, good/bad management, hand-hoe/oxen and
fertilized/unfertilized),
(a) groundnuts (early/late planting and hand-hoe/oxen)
(b) sorghum (local/hybrid seed, hand-hoe/oxen and fertilized/unfertilized),
(c) millet (hand-hoe/oxen)
(d) sunflowers (hand-hoe/oxen)
(e) cassava (hand-hoe/oxen)
(f) sweet potato (hand-hoe/oxen)
(g) cotton (local/hybrid seed, hand-hoe/oxen and fertilized/unfertilized).
Early/late planting options have implications for monthly labor requirements and yields
(and reflect household labor/cash constraints and the problem of untimely input supply).
Problems from late planting due to untimely input supply are frequently cited in the literature on
Zambian smallholders (e.g., SGS Zambia, Ltd., 1999; Skjonsberg, 2003). We model the cost of
untimely inputs by running the model with and without an option for early planting. In addition,
good/bad management affects labor requirements, inputs and their costs, and yields. We model
this by making different assumptions about management levels. As expected, use of hand-hoe
compared to oxen technology has implications for labor requirements, timing of activities and
yields.
Updated Baseline Model Results
Model results generate the objective function value (net returns in ZK to the household),
the amount of land and labor devoted to different crops, and shadow prices for binding
constraints (see annex for details). We have added a few indicators that reflect household well
being, such as the share of household income devoted to staple foods (to reflect food insecurity),
and the value of purchased inputs as a share of household gross revenues (to reflect credit needs
and market exposure). Model results are presented in tables 8 to 11.
We begin with the baseline results (table 8), for the updated model representing
conditions in Eastern Province. The results reflect the finding that - under current conditions -
smallholder agriculture has a limited role to play in reducing rural poverty because of its low
income-generating potential. Given labor, cash and food-security constraints, prevailing
technologies and prices smallholders would choose to produce local maize, groundnuts and
millet. Net returns are about ZK2.7 million per household from 3.2 hectares of land using hand-
hoe technologies. On a per-household basis this represents about $1.7021 per day, about $0.30
per person per day.
Results from Southern and Northern Province are similar to those from Eastern Province.
The food-security constraint pushes smallholders in Eastern and Southern Provinces to primarily
produce maize and groundnuts for own-consumption. As expected, net returns are highest in
Eastern Province, followed by Northern. In Northern Province, relatively large amounts are
spent on purchased inputs, but, because staple food prices are relatively high, households also
receive larger net returns.22 Land under hand-hoe cultivation is highest in Eastern Province.
The baseline model for Eastern Province shows that, under existing prices and
technologies, hybrid maize yields would have to increase by 500 kg/ha (from 2200 to 2700 kg/ha
under hand-hoe cultivation) before hybrid maize would supplant local maize in the optimal
solution. Using hand-hoe technologies, the household cultivates approximately 3.2 hectares of
land, and cannot cultivate more due to limited household labor availability during November and
December, months of land preparation and beginning of season weeding.23 Without alleviating
these labor constraints, additional land availability will not benefit these households.
Oxen clearly help reduce the importance of these early-season labor constraints for
Eastern Province households.24 With oxen, land in production grows to 4.2 hectares and the
household is bound by labor availability in March and April, when the maize harvest is in full
swing. Oxen are still associated with local maize production (as opposed to hybrid maize), but if
the yield of local maize were to fall by 200 kg/hectare (to 1000 kg/ha) with the use of oxen,
hybrid maize would replace local maize production in the model solution. Oxen technology
leads 370 percent higher returns and because it increases hybrid yields by relatively more than it
increases local maize yields, it makes hybrid maize relatively more profitable.
Availability of oxen and the ability to alleviate early-season labor constraints allows the
household to produce sweet potatoes instead of millet. The former crop yields better returns, but
requires substantial amounts of labor for land preparation at about the same time land is being
prepared for maize production. Introduction of oxen also leads to higher use of purchased inputs
and cash exposure, as cash spent on inputs as a percentage of gross production value rises from 8
to 10 percent. In the case of oxen technology, the household spends more than ZK1.1 million
21Using $1= ZK4,300, which was a representative exchange rate for 2003.
22This is a problem when imputing household incomes based on food staple prices. Remote households are merely
eating higher value maize and beans than less remote households.
23Recall, the model assumes that smallholders only use household labor.
24Results for Southern and Northern Province are qualitatively identical; information is available from the authors.
20
(about $270) on inputs. This amount may seem small, but it is significant for a household with
agricultural activities generating about one US dollar per day, and represents a considerable
hurdle to adopting new technologies. Generating savings to purchase these inputs could be
difficult, given the low levels of "profit" from the enterprise. Credit access is thus an extremely
important component of the household's ability to take advantage of market opportunities and
increase output through use of new technologies.
Interestingly, cotton never entered the model solution in either the Eastern or Southern
Province scenarios. In fact, cotton prices would have to rise from MK 850/kg to MK 3900 (a
460% rise, holding all other prices constant) before it becomes attractive to smallholder
producers, according to our model estimates. Even with the food security constraint removed,
cotton does not enter the solution. This finding shows that, given prevailing technologies and
prices, cotton will not be a profitable alternative for Eastern and Southern Province smallholders.
Evidence that cotton production among smallholders is widespread is an indication of: (i) our
model assumptions are incorrect or (ii) other factors are affecting the decision to produce cotton.
An example of the latter might be access to credit or other inputs that induces smallholders to
enter into outgrower schemes.
Stylized Policy Reform Scenarios25
We want to examine the impacts on smallholder households of the following: land
reform, liberalization of fertilizer markets, and improved rural road infrastructure. In addition we
examine impacts of HIV/AIDS. The specific scenarios are described below.
Land reform: in the baseline model, land availability is not constrained. Instead, labor
constraints prevent the households from planting more than 3.2 hectares. Labor availability was
constraining during November and December, when planting and weeding occurs. Introduction
of labor-saving technologies (such as oxen) increases the amount of land the household plants to
4.2 hectares, when labor availability again constrains the household. We examine land-reform
impacts in three ways26. First, we limit the households to only two hectares in production and
examine the impacts on the farmers of progressive increases in land availability. Second, we
examine impacts of providing progressively more land to a poor land-scarce smallholder
household. Finally, we begin to eliminate the labor constraints and examine how land use
patterns change as more labor becomes available to the household. This final scenario shows
how demand for land use might change as labor-saving technologies are introduced into rural
areas.27
25The scenarios were not created with specific research-based information on the likely impacts of the reforms. The
authors know of no such research, but the scenarios could be modified to incorporate research findings on the
impacts of reforms. Alternatively, sensitivity analyses (carried out by adjusting various parameters of the model)
could easily be applied to examine a range of possible outcomes.
26Land reform might also increase access to credit because of improved collateral conditions. The model is not set
up to examine access to credit, although we could look at the implications of more binding cash constraints.
27The model reflects conditions in Zambia where land is abundant. Efforts to bring more land under cultivation
should include complementary moves to improve labor markets and to increase the productivity of existing
agricultural laborers.
21
Fertilizer: We examine the impacts of changes in fertilizer policy in two ways. First, we
examine how increased prices of fertilizer affect farm profits and production patterns. Fertilizer
price increases are likely to be more significant in Northern Province, because of its remoteness;
we raise prices by 10, 20 and 50 percent and examine the impacts on the model. Second, market
liberalization might make fertilizer more readily available. We constrain the model to not allow
timely application of fertilizers and examine the relative impacts of price increases (and timely
application) compared to late-arriving fertilizer.
Infrastructure: We assume that the major impact of remoteness is felt through higher
transaction costs and the farm-gate prices of inputs (higher) and outputs (lower). Therefore, we
assume that new infrastructure should lower the relative costs of purchased inputs and raise the
relative prices of tradeable goods (such as hybrid maize, sunflower and cotton). Improved
infrastructure should also weaken the impetus for smallholders to produce food staples for home
consumption, by increasing the cost associated with making cropping decisions based on a
"safety first" food-security constraint. Purchased food staples should be cheaper and returns on
marketed crops higher.
HIV/AIDS: We assume that households with members having HIV/AIDS will face
reductions in available labor and an increase in household food-security requirements. In other
words, we assume that dependency ratios increase (extra mouths to feed and less hands to work)
as a result of HIV/AIDS. We do not model the costs for medication or other treatments, or other
economic and social impacts of HIV/AIDS. The results can be viewed as a "lower bound" of the
costs.
Model Results
Land reform
The basic results for the scenarios in Eastern Province where land is assumed to be
limited (e.g. this assumes land reform that provides a minimal amount of land to poor
households) are shown in columns one and four of table 9 for hand-hoe and oxen technologies,
respectively. For the hand-hoe household, the solution is infeasible with two hectares, as the
household cannot produce enough maize (local) to meet its food-security constraint. With the
food security constraint imposed, the minimum feasible amount of land is about 2.2 hectares; the
model results with hand-hoe technology and 2.2 hectares are shown in column one of table 9.
For the oxen-using household, the results are shown with a 2 hectare allocation. The low amount
of land available to the household leads to almost all production being devoted to staple foods.
The same scenario, but with the food-security constraints removed, is shown in columns 2 and 5.
Concerns for food security clearly alter production patterns and prevent households from
specializing in higher value crops.
Under oxen cultivation, returns to household labor are more than twice that of the hand-
hoe household, and the land used is slightly less (two hectares versus 2.17). This result shows
how land reform, together with the introduction of labor-saving technology, can help even
relatively poor farmers. Notice however that the oxen-technology option assumes that oxen are
readily available for rental (a big assumption considering high recent oxen mortality rates). The
22
solution requires almost ZK150,000 in cash at the start of the planting season to rent oxen and
purchase other inputs (see below for more details).
Progressively increasing returns from increased land allocations clearly depend on the
technology used. Access to oxen is associated with a higher marginal return to increased land,
while the marginal value of land is lower (but still positive) for hand-hoe technologies (figure 4).
Among Eastern Province smallholders, oxen use is still quite limited, likely due to the high price
of buying and keeping oxen. Oxen rental markets require adequate cash on hand at the
beginning of the growing season; anecdotal evidence shows that seasonal credit for up-front
farming operations is limited. Returns to hand-hoe smallholders of maize also increase with
landholding size, to the maximum acreages noted above. As noted in the 1994 study, the food-
security constraint substantially lowers net returns at all production levels. More confidence in
the ability of food markets to provide minimum food needs will improve conditions (by a
substantial amount) among smallholders, regardless of the technology used.
Model results suggest that allocating more land to smallholders without addressing the
labor constraints will have no real impact on their ability to increase production. Allocations of
more than five hectares will lead to under-utilized land unless either labor markets can be created
(although the demand for labor is not likely to be strong, since current conditions in agricultural
labor markets suggest few transactions take place) or labor saving technologies can be
introduced. We examine the case where household labor is increased by 20 percent for hand-hoe
(see table 9 column 3) and oxen (column 6) technology. Additional labor will substantially
increase the amount of land planted in both cases; such an innovation could put pressure on land
availability in Eastern Province.
Results are shown in more detail in figure 5. Oxen cultivation, while reducing labor
constraints and allowing more land to be farmed, is also associated with greater returns to labor,
and increased labor use. Increased labor available for weeding and harvesting improves
household returns.
As noted above, cash on hand can be a critical constraint for Zambian smallholders. In
each of the above scenarios, we assume that households--either through own savings, own
finance, or borrowing--have enough cash on hand to conduct "normal" input purchases and
make needed expenditures in a timely way. This assumption may not hold, especially given the
relatively low returns to farming (as seen in our model results). We examine the impacts of more
limited cash availability and find that the representative household must have about ZK180,000
on hand at the start of the season to conduct its farming operations (figure 6). In order to reach
the baseline solution, shown above, the household needs about ZK240,000 (about $55) on hand
at the start of the season. At low levels of available cash, returns to cash on hand are quite
substantial (the graph in figure 6 is steeply sloped), which indicates that under the assumptions of
the model, returns to credit (and demand for credit) are likely to be robust.
Fertilizer Policy
We examined above the household impacts of changes in input and output prices (as a
part of the remoteness scenario). To examine the impacts of fertilizer policy changes, we reran
the Eastern Province model but varied the fertilizer price from 10 percent of current prices up to
23
more than 100 percent of these prices. The results are shown in figure 7. At very low prices, the
hand-hoe household, which grows hybrid maize and fertilized millet, uses almost 1000 kg of
fertilizer. The progressive decline in the objective function value as prices are raised is due to
increased costs associated with higher-priced inputs, not to a change in cropping patterns (a basis
change). When the price reaches 70 percent of current prices, the household produces less
hybrid maize and total fertilizer use declines to slightly under 700 kg. At current prices, slightly
over 300 kg is used, but the model shows that even a slight price increase will cause fertilizer use
to drop to zero.
These results are from a model with no food-security constraint; this suggests that
fertilizer policy can boost smallholder incomes only if they have confidence in the market.
However, the elasticity of response is not high; a 20 percent decline in fertilizer prices is
associated with only a 6 percent increase in income. As the price moves from 80 to 70 percent of
current prices, a one percent decrease in the price of fertilizer is associated with a 0.3 percent
increase in household income. At the point where the price is reduced to 70 percent of current
prices, we reach a ceiling on fertilizer use. Below this point, additional fertilizer applications are
not economically rational and the only impact of increased subsidies are on household incomes
(increased incomes due to lower fertilizer prices). This finding sheds light on the impact of
heavily subsidized fertilizer. It might promote inefficient over-use of fertilizer and is no better to
the farmer than a pure income transfer.
Improved Road Infrastructure
Eastern Province model results under the remoteness scenario are shown in table 10.
Remoteness leads to less land under cultivation, lower returns per household member, and lower
returns to land. Net returns are about 10 percent lower under the remoteness scenario, indicating
important welfare costs associated with remoteness.
Improved infrastructure may also increase confidence in markets. Such confidence may,
over time, reduce the need to produce for own consumption. As was shown in figure 4, the food-
security constraint has an important returns-lowering impact on households. Removal of the food
security constraint increases well-being by about 20 percent and makes production on less land
area more feasible. Without the food-security constraint, Eastern households produce higher
(relative) return crops28, such as sweet potatoes, and increased amounts of groundnuts. The
combination of less remoteness and increased confidence in the market is associated with about a
25 percent increase in net returns for all households (hand-hoe and oxen, and across provinces)
compared to remote, food-security-constrained households. These results provide insight into
the benefits to household well-being of investments that allow smallholders to purchase (rather
than produce) maize for their food security requirements.
HIV/AIDS and reduced labor
Model results under this scenario are shown in table 11. Reductions in available
household labor, as might occur with an HIV/AIDS, lead to much less land being cultivated and
28Note the subtle difference between high-value crops and crops that have higher returns than other low-value crops.
24
more production of staple foods relative to other products. In addition, in the hand-hoe
household, purchased inputs increase as a share of total value produced, indicating some
substitution of purchased inputs for scarce labor. Decreased amounts of labor available to the
household, shown in figure 5, lead to progressively lower returns and constrain the household to
cultivate less land. This leads to increased production of food staples that require less labor
and/or labor spread more evenly or more flexibly through the year (e.g., cassava). The opposite
is true when labor is more abundant.
Beyond the Model: Moving into Higher Value Crop Activities and Conservation Farming
To complement findings from the smallholder model we will briefly examine other
studies that shed some light on the potential for agricultural-led poverty-reducing growth for
smallholders based on the adoption of higher value crops and of conservation farming (CF).
Moving into Higher Value Crop Activities
Given current yields and the single-season cropping associated with rain-fed agriculture,
smallholder agriculture can not be expected to make minor contributions to rural poverty
reduction. Our smallholder models show net returns of $0.29 per person per day with hand-hoe
technologies and $1.06 per person per day with oxen technology, far below the internationally
accepted minimums of $1 per person per day. 29
These numbers may be compared with recent financial analyses of returns to different
smallholder crops (Keyser, 2002, see table 12). Crops grown on relatively small landholdings,
while contributing to food security and supplementing family income, can not be expected to
make substantial contributions to poverty reduction. An alternative is clearly needed.
Higher-valued crops yield higher net incomes but do not provide enough income by
themselves to reduce poverty much (table 13). Budgets from Keyser (2002) for commercially
oriented emergent and large-scale farmers show more promise, but (as will be noted below)
initial investments and operating costs for such enterprises require substantial outlays.
Most traditional smallholder crops, such as local maize, groundnuts and cotton, might
have low input costs, but they also have relatively low returns to labor (see table 14). Higher-
value crops like paprika and tobacco have higher cash input and labor demands. Furthermore,
many crops experienced stagnant or falling prices from 1994 to 2001, while input costs tended to
rise. In any case, returns to labor have remained low and in many cases declined, even in cases
where there might have been improvements in yields. High cash and labor requirements are a
major constraint for smallholders.
Paprika represents an example of a new smallholder crop that proliferated quickly. It was
first introduced into Zambia in the early 1990s and quickly developed as a smallholder crop and
important agricultural export. Development of this sector was only possible due to investment
decisions made by a few entrepreneurs. These entrepreneurs identified a specific market
29Actually we make fairly optimistic assumptions about maize yields in both cases.
25
potential and made long-term investments to sustain their operations. This has required
substantial investments in farmer extension, input supply, marketing support, laboratory
equipment, processing facilities, construction of rural depots and negotiation with potential
buyers around the world. Opportunities in other areas exist, but success still depends on
individual entrepreneurs who are willing to respond to market signals and accept the risk and
high cost of investing in new areas (Keyser, Helsop, Abel, 1999, p. 39-40). Producers
experienced declining returns to labor despite improvements in average yields. These are not
good omens for increasing paprika production.
Keyser (2002) and Mwape (2003) claim that commercial agricultural producers in
Zambia can attain higher yields than small-scale producers. Higher returns per hectare,
combined with the fact that these returns are spread over more hectares demonstrate that farming
can be a profitable enterprise. Attainment of "commercial" status requires knowledge of
advanced farming techniques and business and financial management skills. Financial barriers
to entry can be daunting: table 15 presents estimates of the investment costs needed to enter
selected high-value, commercial agricultural enterprises. The combination of huge outlays, the
need to spread investments over large land areas and the technical skill required to produce and
market at a commercial level creates a substantial barrier to entry into commercial farming.
Even a smaller-scale commercial maize farmer require substantial financial resources: a basic,
small-scale commercial farmer with a single 60-horsepower tractor and basic cultivation
implements (plows, harrows, cultivators, fertilizer spreader and maize sheller) would incur more
than $100,000 in fixed investments.
Conservation Farming
Conservation farming (CF) in Zambia is a locally adopted variant of traditional minimum
tillage technologies adopted in many parts of Sub-Saharan Africa (Bwalya, 1999; Haggblade and
Tembo, 2003). CF has gained popularity in the 1990s in response to market liberalization and
the perceived needs to increase fertilizer efficiency, better conserve and manage water resources,
increase productivity, and also to spread labor more evenly over the year. As applied in Zambia,
CT involves a package of several key practices: dry-season land preparation using minimum
tillage (rather than plowing after the first rains), crop residue retention (instead of burning), seed
and fertilizer application in fixed planting stations (rather than spreading), and nitrogen fixing
crop rotations and fallows (rather than continuous production of crops such as maize and cotton).
CF technologies and implements have been developed for hand-hoe and oxen land preparation. It
is important to note that minimum tillage is not synonymous with "low-input" agricultural
production. In many cases there is need for increased labor and outlays on improved seeds and
fertilizers. This is a reason behind the fact that the highest adoption rates of CF in Zambia have
actually been by commercial and emergent farmers.
In 1998, MACO officially formally accepted CF as an official policy of GRZ, and
increased promotion efforts. One of the emerging "success stories" has been the adoption of CF
by many smallholder cotton producers participating in the outgrower "Dunavant Distributor
System". In fact, there is evidence that "CF farmers often receive extra extension support as well
as input packages of high-yielding variety (HYV) seeds and fertilizers to which most
conventional farmers have not had access in the decade and a half following the collapse of
Zambia's input supply and credit systems. Even under conventional tillage, higher fertilizer and
26
HYV seed use will increase output (Haggblade and Tembo, 2003, p.3)." Thus, it has been
difficult to assess the impacts of changes in tillage practices alone versus the package of different
tillage practices and use of improved inputs and increased input use. A recent review of CF in
Zambia (see Haggeblade and Tembo, 2003) notes that it is hard to estimate the number of
smallholders that have adopted CF, because many adopt some components and not others.30
More information is needed about conservation farming to assess its benefits and costs in the
smallholder household farming system.31
30Haggeblade and Tembo (2003, p.80) estimate that "between 20,000 and 75,000 Zambian farmers currently benefit
from increased yield and incomes under conservation farming."
31Although some donors and NGOs advocate conservation tillage, "little published data is yet available on the
economics of conservation tillage compared to alternatives... More research is also needed into intra-household
effects of adopting conservation tillage technology, given that non-plowing field operations are generally carried out
by women and children (Copesake, 1997, p.31)."
27
28
4. SUMMARY AND CONCLUSIONS
Smallholder agriculture has some poverty reducing potential in Zambia, but faces major
constraints. A major priority to benefit poor producers should be labor-saving technologies and
better-functioning input and output markets. This would allow smallholders to expand their land
under cultivation. This is important because in most cases smallholders are not land constrained
(like smallholders in many other countries in the world). Access to credit is also important
because movement to higher-productivity farming systems usually requires increased
investments in inputs and working capital. Capacity building and improved research and
extension services are also critical. Remoteness is a problem that can be overcome through
investments in infrastructure, but complementary assets are lacking in low-productivity remote
areas. Also, transport services need to be improved to lower the costs of transportation. Deeply
rooted structural problems that characterize Zambia's dualistic agricultural sector further
exacerbate the problem. Issues such as land reform, liberalization of fertilizer markets, and
improving infrastructure need to be part of a broader rural development strategy. The rural poor
lack not only land, but other productive resources that would allow them to respond to policies
and investments that are aimed at stimulating agricultural growth. Recent innovations in contract
farming and conservation farming for smallholders point to the potential when a more holistic
technology-technical assistance-credit-marketing approach is adopted. Also instructive are the
experiences from resettlement schemes for ex-mine workers (see box 7). However, numerous
constraints persist with respect to smallholder agriculture and it will be difficult to overcome
them in the short term. This will require imaginative short term policy and investments that can
stimulate more efficient and equitable production systems without reverting to a more centrally
managed agricultural sector.
Box 7: Resettlement Schemes: Urban to Rural Migration
Zambia has tried to encourage some urban residents, notably persons who were formerly employed
in the copper mining sector to move to rural areas. Reports show that residents in resettlement
schemes have had some success in investing in improved technologies. Many, however, bring
human and financial capital with them (SGS Zambia, 1999). Residents who return to villages after
spending time working in urban areas have accumulated human and financial capital that can be
applied to more innovative agricultural practices. This points to some assets "necessary" for the
success of emergent farmers and higher value agriculture.
Looking Ahead: Tapping Smallholder Potential
Market and trade liberalization has created opportunities for Zambian farmers. Elimination of
pan-territorial maize pricing and fertilizer subsidies makes location an important factor in
determining comparative advantage. Suitable farming systems must be identified depending on
agro-ecological conditions, access to infrastructure and markets, and socioeconomic conditions.
In addition, more realistic expectations about the possibilities for smallholder transformation are
needed. As Whiteside (1998, p.1) noted: "Donors and governments need to recognize that to
achieve sustainable increases in agricultural productivity will take decades, not years."
Improved research and extension based on the assets held by smallholders and the new policy
regime are needed, including a "basket of choices" that consider smallholder farming systems
and off-farm opportunities. Investments in transport and communication infrastructure can help
lower transaction costs. There is a pressing need for more technical assistance for smallholders
to improve business skills and group formation.32 See box 8.
Box 8: Needs for Sustainable Smallholder Farming Systems
To achieve more sustainable farming systems for Zambian smallholders, Copestake (1997), Ruiske
et al. (1997) and Saasa (2000) highlight the need for:
� Appropriate technologies and farming systems that are suited to diverse agro-ecological
conditions, for smallholders in remote and risky areas.
� Technological innovations to raise labor productivity.
� Farming systems that are ecologically and financially sustainable.
� Stronger credit markets.
� Better skills and information to participate in domestic and international markets.
� Lower transport-related and other types of transactions costs.
� Improved storage to assure timely application of inputs and to capture better prices for outputs.
� Aggregated smallholder activities, to reap economies of scale, lower transaction costs and
improve bargaining power with the private sector.
� Improved links between smallholders and commercial input and output traders, including
contract farming and out-grower schemes.
.
There is need for short-term interventions to help poor smallholders make this
transformation, since the constraints they face will take time to overcome. Targeted short-term
interventions for rural poverty reduction and social protection, such as public works, soil fertility
improvements, and other forms of safety nets and social funds are options.
Market and trade liberalization in Zambia were unsuccessful because of the hasty
elimination of the public marketing systems, private-sector weaknesses, inappropriate incentive
structures, and the lack of infrastructure in rural areas. Furthermore, rural Zambians who had
become dependent on public-sector marketing systems did not have the requisite human and
social capital to deal with the new realities. As noted in IDS (2003, 22-24) "... making the
playing field more level is an important prerequisite for effective pro-poor policies....It is
important to look at how the poor markets, and gain the capacity to engage with some strength in
markets, important contributing factors highlighted by the case studies include: access to capital,
gaining new skills (from marketing to business experience), building social and commercial
networks, the existence of NGO `facilitators,' and logistical support (from roads to mobile
phones. Thus, overall, adding a pro-poor component to market oriented policies is not an easy
game. Markets are highly politicized, the playing field is uneven, and, without regulation and
protection, poor communities are vulnerable to potential exploitation. Without attention to
improving the capacity of poor people to engage in markets � through active state support and
redistributive measures � the ideals of `pro-poor growth' and `private sector partnership' for
32See Parker (2003) for an interesting, and humbling account, of attempts by CLUSA to organize smallholders.
29
development will remain more rhetorical gloss than reality." This point to the need for a more
comprehensive and holistic approach to rural development is needed, not just an agricultural or
commodity-specific strategy. It is also critical to consider the dualistic nature of the agricultural
sector, and possibility to think in terms of specific smallholder strategies.
Suggestions for Further Research
Although the household model can contribute to the policy dialogue, a number of factors
limit its insights. These include the prevalence of non-price constraints, the importance of
household assets and bundles of assets in addition to cash flows, the risks associated with
fluctuating climatic conditions and prices, and consideration of fixed investment costs in addition
to variable input costs. Non-price, asset, risk and stock/flow cost factors cannot be easily
incorporated into a linear programming model. To incorporate these factors, we would need a
multi-period dynamic model; this would require additional data.
The model is a stylized model of representative households. It would be important to
match and map these representative households with actual households to better understand how
land and labor constraints affect outcomes. For example, model results indicate that
smallholders with up to five hectares should not be "land constrained" using prevalent
technologies and household labor. But which smallholder households actually have access to five
hectares, and where are they located? There might be areas of the country, especially along the
"line-of-rail," with higher population densities and land constraints. In such areas better-
developed land and labor markets could influence cropping decisions.
Model results and additional analysis indicate that many smallholders are trapped in a
"low-level equilibrium." Breaking out of this trap will require adopting new technologies and
cropping patterns, relatively large investments in fixed assets and working capital. Additional
modeling of smallholder adoption of technologies, and subsequent returns from adoption would
be beneficial. This line of research will be more fruitful than focusing on how smallholders can
survive using existing technologies and cropping patterns. More research needs to be carried out
to investigate the economics of adopting higher value crops and/or for the adoption of
conservation farming for staple and higher value crops. There is a need to explicitly incorporate
risks (e.g., climate and price risks) into the modeling of smallholder agriculture and to better
understand how the existence or absence of risks and risk management instruments impact the
transformation from subsistence-oriented smallholders to market-oriented emergent farmers.
We also need research that explores the linkages and labor generating capacity of larger
commercial farms and emergent farms. For example, how many jobs might a 1000ha farm
enterprise generate � both on- and off-farm? What are viable economic units of production for
emergent farmers? More also needs to be known about rural labor markets and rural-urban labor
markets to better understand how off-farm employment in agricultural and non-agricultural
activities might contribute to poverty reduction among smallholders.
30
31
REFERENCES
Alwang, J. and P.B. Siegel. (1999) "Labor Shortages on Small Landholdings in Malawi:
Implications for Policy Reforms," World Development. Vol. 27, No. 8, pp. 1461-1475.
Alwang, J., P.B. Siegel and S. Jorgensen (1996) "Seeking Guidelines for Poverty Reduction
Policies for Zambia," World Development. Volume 24, No. 11, pp. 1711-1723.
Alwang, J. and P.B. Siegel (1994). "Rural Poverty Assessment." Volume III of the "Zambia
Poverty Assessment", Report No. 12985-ZA. The World Bank: Washington, D.C.
Baffes, J. (2004) "Cotton Market Setting, Trade Policies, and Issues." Policy Research Working
Paper Number 3218. The World Bank: Washington, D.C. see www.worldbank.org/research
Boughton, D., D. Tschirley, H. de Marrule, A. Osorio, and B. Zulu (2002) "Cotton Sector
Policies and Performance in Sub-Saharan Africa: Lessons Behind the Numbers in
Mozambique and Zambia." Flash ... No.34E. Department of Statistics and Department of
Policy Analysis, MADER-Directorate of Economics. See
www.aec.msu.edu/agecon/fs2/mozambique
Bwalya, M. (1999) "Conservation farming with animal traction in smallholder farming systems:
Palabana experiences." In Conservation Tillage with Animal Traction. Edited by P.G.
Kaumbutho and T.E. Simaklenga. Resource Book of the Animal Traction Network for
Eastern and Southern Africa (ATNESA): Harare.
Copestake, J.G. (1997) "Encouraging Sustainable Smallholder Agriculture in Zambia."
Agricultural Services Reform in Southern Africa Report R6452CA. Centre for
Development Studies, University of Bath.
CSO (2001) "Preliminary Report of Census 2000." Central Statistics Office. Government of
Zambia: Lusaka.
Deininger, K. and P. Olinto (2000) "Why Liberalization Alone Has Not Improved Agricultural
Productivity in Zambia: The Role of Asset Ownership and Working Capital Constraints."
Policy Research Working Paper Number 2302. The World Bank: Washington, D.C.
see www.worldbank.org/research
Devereux, S. (2000) "Social Safety Nets for Poverty Alleviation in Southern Africa."
Research Report for the DFID, ESCOR Report R7017. The Institute of Development
Studies (IDS): University of Sussex
Doss, C.R. (2002) "Men's Crops" Women's Crops? The Gender Patterns of Cropping in Ghana."
World Development. Vol. 30, No. 11, pp. 1987-2000.
Evans, D. (2001) "Identifying Winners and Losers in Southern Africa from Globalisation:
Integrating Findings from GTAP and Poverty Case Studies on Global Trade Policy
Forum." IDS Working Paper 140. Institute for Development Studies: Sussex.
FAOSTATS-Agriculture website. See www.fao.org
FAO/WFP (2002). Special Report: Crop and Food Supply Assessment Mission to Zambia.
http://www.fao.org/docrep/005/y6997e/y6997e00.htm#P101_11689.
Francis, P.A., J.T. Milimo, G.A. Njobvu, and S.P.M. Tembo (1997) "Listening to Farmers:
Participatory Assessment of Policy Reform in Zambia's Agriculture Sector." The World
Bank: Washington, D.C.
Government of the Republic of Zambia (GRZ) (2000) "Agricultural Statistics Bulletin
1999/2000." Ministry of Agriculture, Food and Fisheries: Lusaka.
Government of the Republic of Zambia (GRZ) (2001) "Agricultural Commercialization
Programme (ACP) 2002-2005." Ministry of Agriculture, Food and Fisheries: Lusaka.
Government of the Republic of Zambia (GRZ) (2002) "Developments in Fertilizer Marketing in
Zambia: Commercial Trading, Government Programs and the Smallholder Farmer."
Ministry of Agriculture and Cooperatives. Food Security Research Project. Lusaka.
Haggblade, S. and G. Tembo (2003) "Conservation Framing in Zambia." EPTD Discussion
Paper No. 108. International Food and Policy Research Institute (IFPRI): Washington, D.C.
see www. ifpri.org
Haggblade, S., editor (2004) Building on Successes in African Agriculture. Focus 12 2020
Vision. International Food and Policy Research Institute (IFPRI): Washington, D.C. see
www. ifpri.org
Howard, J.A. and C. Mungoma (1996) "Zambia's Stop-and-Go Revolution: the Impact of
Policies and Organization on the Development and Spread of Maize Technology." ." MSU
International Development Working Paper No. 61. Michigan State University: East
Lansing.
Institute for Development Studies (IDS) (2003) "The Rural Poor, the Private Sector and Markets:
Changing Interactions in Southern Africa." Report prepared for the Sustainable Livelihoods
in Southern Africa (SLSA) Program. See www.ids.uk/slsa
Jansen, D. (1977) Agricultural policy and performance in Zambia : history, prospects, and
proposals for change. Institute of International Studies, University of California: Berkeley.
32
Jayne, T.S, M. Mukumbu, M. Chisvo, D. Tschirley, M.T. Weber, B. Zulu, R. Johansson, P.
Santos, and D. Soroko (1999) "Successes and Challenges of Food Market Reform:
Experiences from Kenya, Mozambique, Zambia, and Zimbabwe." MSU International
Development Working Paper No. 72. Michigan State University: East Lansing.
Jayne, T.S., T. Yamano, M. Weber, D. Tschirley, R. Ben�fica, D. Neven, A. Chapoto, and B.
Zul� (2001) "Smallholder Income and Land Distribution in Africa: Implications for
Poverty Reduction Strategies." MSU International Development Working Paper No. 24.
Michigan State University: East Lansing.
Keyser, J.C. (2002) "Zambia Financial Crop Models: Comparison of 1994-5 and 2001-2 Seasons
and Zimbabwe 2000-02 Seasons." Report prepared for The World Bank: Washington, D.C.
mimeo.
Keyser, J.C., T. Helsop, and J. Abel (2001) "Trade and Investment Opportunities in
Agriculture." Prepared for USAID Zambia Trade and Investment Enhancement Project
(ZAMTIE) under contract # 690-C-00-00-00283-00. Lusaka.
Kherallah, M., C. Delgado, E. Gabre-Madhin, N. Minot, and M. Johnson (2000) "The Road
Half-Traveled: Agricultural Market Reform in Sub-Saharan Africa." International Food
Policy Research Institute: Washington, D.C.
Loy, J.P. and R. Wichern (2000) "Integration of Zambian Maize Markets." Quarterly Journal of
International Agriculture. 39(2):173-198.
Masters W.A., T. Bedingar, and J.F. Oehmke (1998) The Impact of Agricultural research in
Africa: Aggregate and Case Study Evidence, Agricultural Economics, 19: 81-86.
Milimo, J.T., T. Shilito, and K. Brock (2000) "The Poor of Zambia Speak." Published by the
Zambia Social Investment Fund. Lusaka.
McCulloch, N. B. Baulch, and M. Cherel-Robson (2000) "Globalization, Poverty and Inequality
in Zambia During the 1990s." Paper presented at the OECD Conference on Poverty and
Income Inequality in Developing Countries: A Policy Dialogue on the Effects of
Globalization. November 30-December 1, 2000. see http://www.ids.ac.uk
Mwape, F. (2003) "Rural Household Models in Zambia". Consultant Report prepared for Social
Development Department. The World Bank: Washington, D.C. mimeo.
Mwape, F. (2004) "Institutional Analysis of Fertilizer in Zambia." Background Report for
Zambia PSIA. Prepared for Social Development Department. The World Bank:
Washington, D.C. mimeo.
Nweke, F., S. Haggblade, and B. Zulu (2004) Building on Successes in African Agriculture:
Recent Growth in African Cassava. Brief 4. In S. Haggblade, editor (2004) Building on
33
Successes in African Agriculture. Focus 12 2020 Vision. International Food and Policy
Research Institute (IFPRI): Washington, D.C. see www. ifpri.org
Parker, S. (2003) "CLUSA Zambia Rural Group Business Program." Paper presented at "Paving
the Way Forward for Rural Finance: An International Conference on Best Practices.
International Trade Center, Ronald Reagan Building, Washington, D.C. June 2-4.
PHS (1999) Population and Health Survey. Central Statistics Office. Government of Zambia:
Lusaka.
Pillai, V.K., T.S. Sunil, and R. Gupta (2003) "AIDS Prevention in Zambia: Implications for
Social Services." World Development. Vol. 31, No. 1, pp. 149-161.
Ruiske, J.,T. Reardon, J. Howard, and V. Kelly (1997) "Developing Cereal-Based Demand for
Fertilizer Among Smallholders in Southern Africa: Lessons Learned and Implications for
Other Africa Regions." Policy Synthesis #30 prepared for USAID Bureau for Africa Office
of Sustainable Development. Michigan State University: East Lansing.
Saasa, O. (2000) "Small-scale food processing sector in Zambia." CTA Working Document No.
8015. Report produced by Technical Centre for Agricultural and Rural Cooperation (CTA):
Wageningen, The Netherlands.
Seshamani, V. (1998) "The Impact of Market Liberalization on Food Security in Zambia," Food
Policy 23(6), 539-551.
Seshamani, V., H. White, N. Chisupa, and J. Leavy (2002) "Northern Province Field Report."
Draft report prepared for DFID. Seewww.ids.uk. mimeo
SGS Zambia Ltd. (1999) "The Study of Duality in Zambia: Final Report." Submitted to USAID
Mission in Zambia: Lusaka.
Skjonsberg, E. (1979) Change in an African Village: Kefa Speaks. Kumarian Press: West
Hartford, CT.
Skjonsberg, E. (1994) "Kefa Revisited: A Report on Change in Kefa Village, Eastern Province,
Zambia." Background Report for Zambia Poverty Assessment. Prepared for Population and
Health Division, Southern Africa Department, The World Bank: Washington, D.C. mimeo.
Skjonsberg, E. (2003) "A Report on 25 Years of Rural Change in Kefa Village, Eastern
Province, Zambia." Background Report for Zambia PSIA. Prepared for Social
Development Department. The World Bank: Washington, D.C. mimeo.
Smale, .M and T.S. Jayne (2004) "Maize Breeding in East and Southern Africa, 1990-2000." In
S. Haggblade, editor (2004) Building on Successes in African Agriculture. Focus 12 2020 Vision.
International Food and Policy Research Institute (IFPRI): Washington, D.C. see www.
ifpri.org
34
Yanggen, D., V. Kelley, T. Reardon, and A. Naseem (1998) "Incentives for Fertilizer Use in
Sub-Saharan Africa: A Review of Empirical Evidence on Fertilizer Response and
Profitability." MSU International Development Working Paper No. 70. Michigan State
University: East Lansing.
Whiteside, M. (1998) "Encouraging Sustainable Smallholder Agriculture in Southern Africa in
the Context of Agricultural Services Reform." Natural Resource Perspectives Number 36.
Overseas Development Institute (ODI): London.
World Bank (1994) "Zambia Poverty Assessment", Report No. 12985-ZA. The World Bank:
Washington, D.C.
World Bank (2002a) "Zambia: Poverty Reduction Strategy Paper and Joint Staff Assessment."
Report No. 24035-ZA. The World Bank: Washington, D.C.
World Bank (2002b) "Zambia Agricultural Sector Investment Program: Implementation
Completion Report." Report No. 24444-ZA. The World Bank: Washington, D.C.
World Bank (2003a) A User's Guide to Poverty and Social Impact Analysis. Poverty Reduction
Group and Social Development Department. Washington, D.C. see
www.worldbank.org/psia
World Bank (2003b) "Zambia: The Challenge of Competitiveness and Diversification" Report
No. 25388-ZA Washington, D.C.
World Bank (2004) "Zambia Road Rehabilitation and Maintenance Project in Support of First
Phase of the ROADSIP II Program." " Report No. 27891-ZA Washington, D.C.
35
Table 1: Typology of Agricultural Producers in Zambia
Approx. Approx Farm Technology, Market Location Major
# of Producers Size Cultivation Orientation Constraints
Practice
Small-Scale 800,000 hhs < 5ha Hand hoe, Staple foods, Entire country Remoteness,
Producers (with majority minimal inputs, primarily home seasonal labor
cultivating 2 or household labor consumption constraints, lack
less ha of rain- of input and
fed land) output markets
Emergent 50,000 hhs 5 - 20 ha Oxen, hybrid Staple foods Mostly line-of- Seasonal labor
Farmers seed and and cash crops, rail (Central, constraints, lack
fertilizer, few primarily Lusaka, of credit, weak
with irrigation, market Southern market
mostly orientation Provinces), information
household labor some Eastern,
Western
Provinces
Large-Scale 700 farms 50 - 150ha Tractors, hybrid Maize and cash Mostly Central, High cost of
Commercial seed, fertilizer, crops Lusaka, credit,
Farms some irrigation, Southern indebtedness
modern mang., Provinces
hired labor
Large 10 farms 1000+ ha High Maize, cash Mostly Central, Uncertain
Corporate mechanization, crops, vertical Lusaka, policy
Operations irrigation, integration Southern environment
modern mang., Provinces
hired labor
Produced by authors. Adapted from World Bank (2003b, p. 66-67), Francis, et al., (1997, p.13).
Table 2: Estimated Shares (percent) of Zambian National Production by Typology of Producers
Small-Scale Emergent Commercial Corporate
Maize 60 15 25
Sorghum 90 8 2
Soybean 20 10 70
Wheat 5 30 45
Groundnuts 85 10 5
Cotton 98 2
Coffee 5 45 50
Sugarcane 40 60
Tobacco 60 40
Milk 20 30 50
Poultry 10 20 20 50
Source: World Bank (2003b, p.67).
36
Table 3: Major Agro-Ecological Zones of Zambia
Provinces % Share of Rainfall Growing Major Soil Quality, Livestock
Covered Rural Season Agricultural Agricultural Potential
Population Activities Potential
Region I Southern 48 600- 80-120 Maize limited Soils: Shallow Limited by
parts of 800mm days by rainfall. Sands existence of tse-
Western and Sorghum, tse fly and
Southern millet, Ag Potential: trypanosomiasis
Province sunflower, Poor
cassava,
cotton,
tobacco.
Livestock
limited by tse-
tse fly.
Region II Most parts of 43 800- 100-140 Maize, Soils: Absence of
Central, 1000mm days groundnuts, Moderately tse-tse fly and
Eastern, and wide leached sandy trypanosomiasis
Lusaka, range of crops loams
Southern and livestock.
Provinces Ag Potential:
Good
Region III Northern, 9 1100- 120-150 Maize, Leached and Limited by
Luapala, 1700mm days bananas, acidic sands existence of tse-
Copperbelt, coffee, tea. tse fly and
Northwestern Limited by Ag Potential: trypanosomiasis
Provinces high acid Moderate
soils.
Produced by authors. Adapted from: Francis (et al., 1997), p.5; Copestake (1997, p.9); Milimo, Shilito, and Brock
(2000, p.11); World Bank (2003, p.68).
Table 4: Distribution of Rural Households, Land per Household and Agricultural Activities, by Province
Number of Rural Ha per % Households Engaged in
Province Households Household Crops Livestock Poultry
Central 70916 3.23 94.3 29.1 80.7
Copperbelt 34540 3.18 99.4 18.0 67.7
Eastern 188592 2.20 98.9 45.4 67.7
Luapula 106661 2.61 98.1 17.9 62.9
Lusaka 17265 1.98 100 22.4 62.2
Northern 136538 6.54 97.6 28.5 67.4
North-western 50379 1.70 99.4 17.6 48.0
Southern 110810 2.40 94.9 48.7 84.4
Western 102743 1.75 99.5 18.7 64.6
Zambia 818445 3.05 97.8 31.3 68.6
Source: CSO [2001], Preliminary Report of Census 2000, 1999/2000 PHS Data, p.15.
37
Table 5: Major Smallholder Crops in Zambia 2000 � 2002 and Distribution by Province
Crop Average Area Major Producing Provinces and National Share of Production
Harvested (ha) (%)
Maize 500,000 Eastern (32%) Southern (29%) Central plus
Copperbelt (22%)
Cassava 165,000 Northern (50%) Luapula (37%)
Groundnuts 135,000 Eastern (32%) Northern (20%) Southern (14%)
Millet 65,000 Northern (56%) Western (18%) Eastern plus
Central (17%)
Seed cotton 55,000 Eastern (64%) Northern (4%) Southern (3%)
Sorghum 35,000 Northern (21%) Southern (19%) Central & Western
(14% each)
Mixed beans* 45,000
Sunflower Seed 18,000 Eastern (32%) Central (16%) Southern (12%),
Northern (5%)
Sweet Potato* 3,500
Notes: Numbers are rounded-off approximations. Mixed beans and sweet potato are grown in all provinces.
Sources: Government of Zambia (2000); FAO-Agstats (website).
Table 6: Estimated Composition of Zambia's Agricultural GDP (percent)
1994-1995 2000-2001
Average Share Average Share
Maize 26 19
Other Cereals (sorghum, millet, wheat, rice) 8 7
Roots and Tubers 5 14
Oilseeds and Legumes (groundnuts, beans, sunflower, soybean) 8 7
Cotton 15 19
Fruit/Vegetables/High-value Cash Crops 8 14
Livestock Products 30 20
Total 100 100
Source: World Bank (2003, p.69)
Table 7: Major Crops in Southern, Eastern and Northern Provinces (in hectares)
Crops Eastern Province Southern Province Northern Province
Maize 184,900 166, 020 25,411
Groundnuts 21,903 9,456 13,976
Sorghum/Millet 7,883 8,020 42,691
Sunflowers 8,388 1,519 610
Cotton 23,623 2,348 --
Sweet Potatoes 25,134 4,545 1,826
Cassava* ?? ?? 109,477
Source : FAO-Agstats (website).
Note: (*) Data for cassava in Northern Province are estimates.
38
Table 8: Baseline model results
Eastern Province Southern Province Northern Province
Item Hand-Hoe Oxen Hand-Hoe Oxen Hand-Hoe Oxen
1 2 3 4 5 6
OBJ FCN (ZK) Net
Value of Product 2,688,791 10,009,210 2,104,280 8,744,505 2,648,482 7,817,377
Returns (ZK) per 448,132 1,668,202 350,713 1,457,418 441,414 1,302,896
household member
% Staple Food
Value of OBJ FCN 68.4 38.0 97.0 65.2 68.0 23.0
Food Security Maize Maize Maize Maize Maize Maize
Const. & (Shadow (-711) (-.--) (-306) (-.--) (-699) (-243)
Price) ZK/kg Groundnuts Groundnuts Groundnuts Groundnuts Groundnuts Groundnuts
(-3703) (-4516) (-2140) (-3592) (-3594) (-3593)
Cash on Inputs 245,070 1,161,600 244,810 887,750 435,300 1,609,100
(ZK)
% Cash Inputs of
Gross Value of 8.3 10.3 10.3 9.2 14.1 17.0
Production
Land (ha) 3.16 4.16 2.78 4.72 2.74 3.47
Crops (Ha)
LM1 0.47 1.67
LM2 1.41 1.96
HM3 0.57
HM4 0.23
LM5 2.00 0.83
LM6 3.13
HM7 0.87 1.59
HM8 0.25
GR1 0.50
GR2 0.59 0.59
GR3
GR4
GR5 0.40
GR6 0.47 0.47
MIL1 0.69
MIL2
SP1
SP2 1.76 0.57
MB1 966,599 2,252,846
MB2 275 460
CT1
CT2
CT3
CT4
Returns to Land 850,883 2,406,060 756,935 1,852,649 966,599 2,765,151
(ZK/ha)
Labor Days, total 364 445 283 444 275 345
Note: See Annexes A and B for details on cropping activities, assumptions about inputs, yields and prices.
39
Table 9: Model results with land constraint, output and input price adjustments
Eastern Province
Item Hand-Hoe Oxen
1a 2b 3c 4d 5e 6f
OBJ FCN (ZK) Net
Value of Product 1,656,142 3,039,900 3,659,151 4,504,308 7,274,920 12,191,686
Returns (ZK) per
household member 276,024 506,650 609,859 750,718 1,212,487 2,031,948
% Staple Food Value
of OBJ FCN 84.5 0 38.3 31.1 0 27.6
Food Security Maize Maize Maize Maize
Constraint & (-8E-4) n/a (-711) (-840) n/a (-.--)
(Shadow price) ZK/Kg Groundnuts Groundnuts Groundnuts Groundnuts
(-0.01) (-3703) (-5453) (-4516)
Cash on Inputs (ZK) 175,080 152,000 294,150 545,680 1,112,000 1,495,200
% Cash Inputs of
Gross Value of 9.5 4.8 7.4 10.8 13.2 10.9
Production
Land (ha) 2.17 2.00 3.85 2.00 2.00 4.76
Crops (Ha)
LM1 1.67 0.18
LM2 1.76
LM5 0.83 2.00
GR1 0.50
GR2 0.588
GR5 0.40 0.40
MIL1 2.00 1.33
SP2 0.77 2.00 2.36
Returns to Land 763,199 1,519,950 950,429 2,252,154 3,637,460 2,561,279
(ZK/ha)
Labor Days, total 256 293 459 221 235 515
aLand availability restricted (2 ha) for hand-hoe cultivation. Solution is infeasible; Land availability restricted for
b
hand-hoe cultivation, without the food-security constraint; Model results with household labor increased by 20
c
percent (for hand hoe); Model results with 2 ha land constraint. Land constraint is binding; Results after removing
d e
the food-security constraint for oxen technology while limiting land to 2 ha; fModel results with household labor
increased by 20 percent (for oxen). Cotton and sorghum are not shown, because these activities never enter solution.
40
Table 10: Model results under remoteness scenario, Eastern Province
Item Hand-Hoe Oxen
333 334
OBJ FCN (ZK) Net Value of Product
2,443,699 9,216,519
Returns (ZK) per household member
407,283 1,536,086.5
% Staple Food Value of OBJ FCN
48.7 13.0
Food Security Constraint Maize Maize
& (Shadow price) ZK/kg (-809) (-13)
Groundnuts Groundnuts
(-3645) (-5817)
Cash on Inputs (ZK) 276,320 1,740,100
% Cash Inputs of Gross Value of Production
10.1 15.8
Land (ha) 3.16 3.46
Crops (Ha)
LM1 0.47
LM2 1.41
LM5 0.83
GR2 0.59
GR5 0.40
MIL1 0.69
SP2 2.23
Returns to Land (ZK/ha) 773,322 2,663,734
Labor Days, total 364 393
Source: Model results.
33Model results with price of sunflowers lowered by 30 percent and maize by 15 percent, while the prices of
fertilizer, hybrid seed and fixed variable inputs are raised by 30 percent (for hand hoe).
34Model results with price of sunflower lowered by 30 percent and maize by 15 percent, while the prices of
fertilizer, hybrid seed and fixed variable inputs are raised by 30 percent (for oxen).
41
Table 11: Labor decreased by 20 percent (columns 2 and 4).
Columns 1 and 3 show results from the baseline model, Eastern Province
Eastern Province
Item Hand hoe Oxen
1a 2b 3 4
OBJ FCN (ZK) Net Value of Product
2,688,791 1,718,431 10,009,210 7,799,572
Returns (ZK) per household member
448,132 286,405 1,668,202 1,299,929
% Staple Food Value of OBJ FCN
52.01 81.47 33.54 39.34
Food Security Constraint Maize Maize Maize Maize
& (Shadow price) ZK/kg (-711) (-711) (-.--) (-.--)
Groundnuts Groundnuts Groundnuts Groundnuts
(-3703) (-3703) (-4516) (-5392)
Cash on Inputs (ZK) 245,070 195,990 1,161,600 861,790
% Cash Inputs of Gross Value of
Production 8.3 10.2 10.3 9.9
Land (ha) 3.16 2.47 4.16 3.46
Crops (Ha)
LM1 0.47 0.77
LM2 1.41 1.06
LM5 2.00 1.83
LM6
HM7 0.014
GR2 0.59 0.588
GR5 0.40 0.40
MIL1 0.69 0.05
SP2 1.76 1.22
Returns to Land (ZK/ha) 850,883 695,721 2,406,060 2,254,212
Labor Days, total 364 268 445 367
aBaseline model results for both oxen and hand-hoe cultivation. A 20 percent decrease in labor availability results in less land
b
being cultivated, lower objective function value, and increase in LM1 but lowering of LM2, no change in GR2 and a huge drop in
land allocated to millet.
42
Table 12: Net incomes by crop, financial model
Crop Net income Per farm (3 ha) Per person, 6 Per person per
($/ha) ($) persons per HH day
($/person) ($/person/day)
Local maize 52 156 26 0.07
Hybrid maize 75 225 38 0.10
Groundnuts 149 447 74 0.20
Source: Keyser, 2002, using best available technology and management.
Table 13: Net incomes by crop, financial model
Crop Net income Per farm (3 ha) Per person, 6 Per person per
($/ha) ($) people per HH day ($)
($)
Cotton 89 267 44 0.12
Burley tobacco 579 1737 290 0.79
Paprika 182 546 91 0.25
Source: Keyser, 2002, using best available technology and management.
Table 14: Comparison of Yields, Inputs, Labor, and Returns of Different Smallholder Crops 1994/5 and 2001/2
Crop Season Mgt Yield Farmgate Gross Seed, Fert, Transport, Labor Gross
Level (kg/ha) Price Rev Agri-chem Tractor, Days Profit per
($/kg) ($) Costs ($/ha) Etc. Labor Day
($/day)
Maize 94/5 Low 700 0.12 82 -- 20 60 0.86
(local)
Maize 01/2 Low 700 0.12 86 -- 20 60 0.91
(local)
Maize 94/5 Med 1,800 0.12 212 89 54 75 0.23
(hybrid)
Maize 01/2 Med 2,000 0.12 244 78 58 78 0.60
(hybrid)
Maize 94/5 High 3,250 0.12 382 158 93 95 0.36
(hybrid)
Maize 01/2 High 3,500 0.12 427 138 97 98 0.79
(hybrid)
Grdnuts 94/5 Low 350 0.56 196 -- 12 90 1.28
Grdnuts 01/2 Low 350 0.58 203 -- 11 90 1.36
Grdnuts 94/5 High 650 0.56 363 193 25 110 0.60
Grdnuts 01/2 High 650 0.58 378 128 24 110 1.37
Cotton 94/5 Med 600 0.47 282 12 -- 113 2.18
Cotton 01/2 Med 800 0.20 156 38 -- 126 0.69
Cotton 84/5 High 800 0.48 386 34 5 147 2.04
Cotton 01/2 High 1,100 0,20 220 55 4 167 0.62
Paprika 94/5 Med 700 0.90 630 217 13 220 1.29
Paprika 01/2 Med 800 0.70 560 232 13 230 0.85
Tobacco 94/5 Med 800 1.24 988 192 51 270 1.17
(burley)
Tobacco 01/2 Med 1,000 0.98 977 161 61 295 1.19
(burley)
Tobacco 94/5 High 1,200 1.32 1,588 285 74 325 1.91
(burley)
Tobacco 01/2 High 1,500 1.02 1,535 251 90 350 1.83
(burley)
Source: Keyser (2002).
Notes: Labor days include household and hired labor
43
Table 15: Estimated Costs of Entry into Commercial Farming for Selected Enterprises
Total Annualized Assumed Annualized Increased output Value of
Investment Investment Area (ha) Investment (% change) increased
Cost ($) Cost ($) Cost ($/ha) output ($)
Commercial 656,000 77,370 600 129 57% 325
Maize (to 5,500 kg/ha)
Greenhouse 287,500 25,832 2 12,916 -- --
Center Pivot 187,163 11,840 80 148 -- --
Irrigation
Drip Irrigation 72,791 1820 40 163 -- --
Source: Keyser (2002)
44
Figure 1: Maize land planted and total production, 1993-2002.
1,800,000 800,000
1,600,000 700,000
1,400,000
600,000
1,200,000
500,000 )se
T) ar
M( 1,000,000
oni ecth( Maize area planted
400,000 det Maize production
oductrP 800,000 an
pl
300,000 earA
600,000
200,000
400,000
200,000 100,000
0 0
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
Year
Source: FAO Agstats
45
Figure 2: Land planted to various crops, 1993-2002
160,000
140,000
120,000
100,000
Millet
ectares) Sorghum
(h Groundnuts
80,000
tedn Soybeans
pla Sunflower
earA Seed Cotton
60,000
40,000
20,000
0
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
Year
Source: FAO Agstats
46
Figure 3: Production of various crops, 1993-2002
160,000
140,000
120,000
)T 100,000
M( Millet
oni Sorghum
Groundnuts
uct 80,000
od Soybeans
prlatoT Sunflower
Seed Cotton
60,000
40,000
20,000
0
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
Year
Source: FAO Agstats
Figure 4. Impacts of landholding on net returns
Variation in Net Returns by Landholding, Eastern Province
12.00
10.00
8.00
acha)wK
Oxen(FSC)
illion Oxen(No FSC)
6.00
(M Hand Hoe (FSC)
sn Hand-Hoe(No FSC)
ur
Ret
4.00
Net
2.00
0.00
1.90 2.10 2.30 2.50 2.70 2.90 3.10
Landholding (Ha)
Source: Household model results.
47
Figure 5: Impacts of increased labor availability on net returns and acreage under cultivation, hand-hoe and
oxen technology households.
Net Returns and Acreage, by Household Labor Availability
16.00
14.00
12.00
ha)
acwK 10.00
onlili Land (Hand-hoe)
M( Net Returns (Hand-hoe)
8.00
Land (Oxen)
turnseR Net Returns (Oxen)
a),H( 6.00
Land
4.00
2.00
0.00
500 550 600 650 700 750 800
Man-day Equivalents
Source: Household model results
Figure 6: Impacts of cash constraints on model solution, Eastern Province.
Relationship between cash available and objective function, Eastern Province
3,000,000
2,500,000
2,000,000
)KZ
M(
snruteRt 1,500,000 Obj Fcn Value (ZMK)
Ne
1,000,000
500,000
0
150,000 170,000 190,000 210,000 230,000 250,000 270,000 290,000 310,000
Cash available (in September)
48
Source: Household model results
Figure 7: Impacts of fertilizer price changes on model solution, Eastern Province.
Net returns vs fertilizer price increases (Hand hoe)
4,000.00
3,500.00
3,000.00
2,500.00
returns 2,000.00
Net
1,500.00
1,000.00
500.00
0.00
10 20 30 40 50 60 70 80 90 100 110 120
Percent of original price
Fert quantity used (Kilos) Obj Fcn V
49
ANNEX
ZAMBIA SMALLHOLDER MODEL: STRUCTURE, ASSUMPTIONS AND DATA
Model Structure:
As part of the World Bank's 1994 Zambia Poverty Assessment, a household model was designed to better
understand the economic opportunities and constraints of "representative" smallholder farmer households.
This model was used to measure the impacts of changes in policies and investments on crops produced
and marketed, technologies and cultivation practices, labor and land allocation and constraints and
incomes (see World Bank, 1994; Siegel and Alwang, 1994; Alwang, Siegel and Jorgensen, 1996). The
model was updated to reflect conditions in 2003 and regionalized to reflect conditions in Eastern,
Southern, and Northern provinces.
A deterministic single-period LP model was used and the models' parameters are assumed fixed for the
duration of the planning horizon, which is a single year. All farm-level decisions (e.g., crop, land, labor
usage) are assumed to be made at the beginning of the planning horizon, when land-preparation and
planting decisions take place. Crop activities are divided into monthly periods to reflect the seasonality of
crop production and labor requirements. The model only considers incomes from on-farm agricultural
activities.35 By articulating the objectives, resource requirements and resource constraints facing
representative rural households, we predict their choices (crop production activities) and outcomes
(measured by net value of production, household income, and other indicators). The data used were taken
from various sources, notably the 1994 model and a paper prepared by Dr. Francis Mwape of University
of Zambia that updates many of the parameters (see Mwape, 2003).
The model maximizes returns to fixed assets, subject to the technologies and constraints. The "production
technologies" are basically crop budgets, as summarized below. The constraints apply to land, monthly
labor, food security and cash. The model includes the effects of household composition, access to land,
technology and inputs, and patterns of seasonality as they affect the availability of household labour and
household food security. Several scenarios of the model are analysed. These scenarios provide insights
into smallholder behaviour in response to changes in the economic and physical environment.
The LP model can be expressed as the choice of activities to maximize the net value of production from
j=1,...,n activities purchasing i=1,...,q inputs for these activities, such that
n n q
max V = p j Xj - c ijXj
j=1 j=1 i=1
subject to input requirements and constraints for i=1,...,m resources:
35Most smallholder households derive a majority of their income from agriculture activities. There is a general lack
of non-farm non-agricultural activities (SGS Zambia, Ltd., 1999; Milimo, Shilito, Brock, 2003; Skonsberg, 2003).
50
am
ijXj bi ,
j=1
and the non-negativity constraint:
Xj 0
where:
V = net value of farm production (gross value of production minus cost of purchased inputs),
Xj = level of the jth farm activity (e.g., hectares of hybrid maize using a given technology and
management practice), and n is the number of possible activities under consideration (e.g., different crops
using different technologies and management practices),
pj = per hectare value from the jth farm output (price times yield),
cij = cost of the ith purchased input used for 1 hectare of production of the jth activity,
aij = quantity of ith resource (e.g., labor) required for one hectare of the jth activity,
bi = amount of ith resource available to farm (e.g., household's total labor availability for a given month).
Solution of the LP model provides information on how changes in resource endowments, bi, change the
net value of production, V. If a resource constraint is binding, then it is possible to calculate the shadow
price of that scarce resource. The shadow price indicates the value to the household of an additional unit
of the scarce resource. Thus, the shadow price can be considered as the price the household would be
willing to pay for an extra unit of a scarce resource, such as labor. By examining shadow prices of the
binding constraints, insights can be gained about the relative importance of different constraints. For
example, by separating labor requirements by month, the value of labor inputs during different times of
the year can be computed. The model thus quantifies the value of each constraint, in terms of how much
it contributes to the objective function.
Objective Function of the Household
It is assumed that two objectives guide smallholders' decision-making. First, household behavior is
governed by a safety-first rule that compels households to produce a minimum amount of staple foods.
Second, it is assumed that, having met the first objective, the household maximizes net income from crop
production activities. The safety-first rule (i.e., the food-security objective) is modeled as a constraint.
That is, a constraint is used to "force" the household to produce enough staple food to feed itself before
other crop activities can be selected to maximize the net value of production. Resource allocation
decisions are also subject to seasonal labor constraints. In addition, household decisions depend on the
level of available technology, the amount of available cash and land, and the availability of input and
output markets (i.e., timeliness of delivery).
Types of Households
51
For the model it is assumed that traditional households use hand-hoe technology and emergent
households use oxen technology. It is thus possible to use the model to isolate differences in crop mix,
labor allocation, etc. based on use of hand-hoe or oxen technology. Oxen can be used by emergent
households during land preparation, weeding, harvest, and for post-harvest activities. It is common
practice in Zambia for several households to share use of oxen. Households with access to oxen are
assumed to contribute labor, upkeep costs of the oxen, and upkeep cost of oxen-related equipment. Oxen
activities are thus distinguished from hand-hoe activities by their higher returns, different labor
requirements, and higher input costs. It is assumed that both types of households depend solely on
household labor for crop production. The household labor endowment for crop production depends on
the composition of the household (i.e., the number of adult equivalent workers. See tables A and B.
52
ANNEX A:
Summary of Crop Production Activities by Province
Table A.1 Summary of information on crop production activities for Eastern Province
Crop Production Characteristics Labor Requirements in Labor Days/Ha
Code Crop Tech Variety Management Fert Yield Kg/Ha Prep Plant Weed Fert Guard Harvest P-Harvest Total
LM1 Maize Hand Local High No 1,200 34 8 20 0 10 12.4 12 96.40
LM2 Maize Hand Local Low No 1,020 30 7 18 0 10 10.54 10.8 86.34
HM3 Maize Hand Hybrid High Yes 2,200 38 6 22 8 12 17 17 120.00
HM4 Maize Hand Hybrid Low Yes 1,870 34 6 21 8 11 14 15.3 109.30
LM5 Maize Ox Local High No 2,400 17 6 22 0 12 18.4 16 91.40
LM6 Maize Ox Local Low No 2,040 15 6 19.8 0 10 16.56 14.4 81.76
HM7 Maize Ox Hybrid High Yes 4,400 18 6 26 8 11 27.2 25 121.20
HM8 Maize Ox Hybrid Low Yes 3,740 15 6 22 6 10 24.48 22.5 105.98
GR1 Groundnuts Hand Local High No 400 52 8 32 0 11 24 64 191.00
GR2 Groundnuts Hand Local Low No 340 40 8 25 0 10 21.6 57.6 162.20
GR3 Groundnuts Hand Hybrid High No 700 47 8 24 0 12 40 64 195.00
GR4 Groundnuts Hand Hybrid Low No 595 38 8 21.6 0 10 36 57.6 171.20
GR5 Groundnuts Ox Local High No 500 8 8 35.2 0 10 32 44 137.20
GR6 Groundnuts Ox Local Low No 425 8 8 27.5 0 9 28.8 39.6 120.90
GR7 Groundnuts Ox Hybrid High No 900 8 8 26.4 0 12 48 90 192.40
GR8 Groundnuts Ox Hybrid Low No 765 8 8 23.76 0 10 43.2 81 173.96
SUN1 Sunflower Hand Local Low No 800 40 5.2 36 0 0 10 6 97.20
SUN2 Sunflower Ox Local High No 1,600 16 5.2 36 1 0 13.6 16 87.80
MIL1 Millet Hand Local Low No 2,000 34 5.2 22.8 0 36 12.4 36 146.40
MIL2 Millet Ox Hybrid High Yes 2,800 3.2 5.2 20.52 1.2 36 27.2 36 129.32
SP1 Sweet Potatoes Hand Local Low No 4,000 60 4 24 0 0 16 20 124.00
SP2 Sweet Potatoes Ox Hybrid High No 12,000 8 12 21.6 0 0 36 40 117.60
COT1 Cotton Hand Hybrid High Yes 1,200 34 2 30 2 22 50 4.8 144.80
COT2 Cotton Hand Hybrid Low No 1,020 30 2 27 0 20 42.5 3.2 124.70
COT3 Cotton Ox Hybrid High Yes 1,400 11 2 20 2 22 50 4 111.00
COT4 Cotton Ox Hybrid Low No 1,190 9 2 16 0 19 42.5 3.2 91.70
53
Table A.2. Summary of information on crop production activities for Southern Province
Crop Production Characteristics Labor Requirements in Labor Days/Ha
Yield
Code Crop Tech Variety Manage. Fert Kg/Ha Prep Plant Weed Fert Guard Harv. P-Harv. Total
LM1 Maize Hand Local High No 1,200 34 5.2 22.8 0 10 12.4 12.8 97.20
LM2 Maize Hand Local Low No 1,020 32 5.2 18 0 9 10.54 11.52 86.26
HM3 Maize Hand Hybrid High Yes 2,000 36 5.2 24 8 12 17 16 118.20
HM4 Maize Hand Hybrid Low Yes 1,700 33 5.2 22 8 11 15.3 14.4 108.90
LM5 Maize Ox Local High No 2,400 17 5.2 25.3 0 11 18.4 15.6 92.50
LM6 Maize Ox Local Low No 2,040 16 5.2 19.8 0 10 16.56 14.04 81.60
HM7 Maize Ox Hybrid High Yes 4,400 17 5.2 27 7 11 27.2 25 119.40
HM8 Maize Ox Hybrid Low Yes 3,740 15 5.2 24 7 10 24.48 22.5 108.18
GR1 G'nuts Hand Local High No 400 32 8 32 0 11 24 64 171.00
GR2 Groundnuts Hand Local Low No 340 30 8 25 0 10 21.6 57.6 152.20
GR3 Groundnuts Hand Hybrid High No 700 34 8 21.6 0 12 40 64 179.60
GR4 Groundnuts Hand Hybrid Low No 595 33 8 19.44 0 10 36 57.6 164.04
GR5 Groundnuts Ox Local High No 500 8 8 35.2 0 10 32 44 137.20
GR6 Groundnuts Ox Local Low No 425 8 8 27.5 0 11 28.8 39.6 122.90
GR7 Groundnuts Ox Hybrid High No 900 8 8 24 0 12 48 90 190.00
GR8 Groundnuts Ox Hybrid Low No 765 8 8 21.38 0 10 43.2 81 171.58
SUN1 Sunflower Hand Local Low No 400 40 5.2 48 0 0 10 6 109.20
SUN2 Sunflower Ox Local High No 1,600 16 5.2 48 0.8 0 13.6 16 99.60
MIL1 Millet Hand Local Low No 2,000 34 5.2 22.8 0 36 12.4 26 136.40
MIL2 Millet Ox Local High Yes 2,800 3.2 5.2 20.52 1.2 36 27.2 36 129.32
Sweet
SP1 Potatoes Hand Local Low No 4,000 60 4 24 0 0 16 20 124.00
Sweet
SP2 Potatoes Ox Hybrid High No 12,000 8 12 21.6 0 0 44 60 145.6
SOR1 Sorghum Hand Local Low No 2,000 34 5.2 22.8 0 36 12.4 12 122.40
SOR2 Sorghum Ox Local High Yes 2,800 3.2 5.2 68.4 1.2 36 27.2 10.4 151.60
COT1 Cotton Hand Hybrid High Yes 1,200 34 2 30 2 22 50 4.8 144.80
COT2 Cotton Hand Hybrid Low No 1,020 30 2 27 0 20 42.5 3.2 124.70
COT3 Cotton Ox Hybrid High Yes 1,400 11 2 20 2 22 50 4 111.00
COT4 Cotton Ox Hybrid Low No 1,190 9 2 16 0 19 42.5 3.2 91.70
54
Table A.3 Summary of information on crop production activities for Northern Province
Crop Production Characteristics Labor Requirements in Labor Days/Ha
Yield
Code Crop Tech Variety Manage. Fert Kg/Ha Prep Plant Weed Fert Guard Harv. P-Harv. Total
LM1 Maize Hand Local High No 1,200 32 5.2 22.8 0 10 12.4 2.8 85.20
LM2 Maize Hand Local Low No 1,020 32 5.2 18 0 9 11.16 2.52 77.88
HM3 Maize Hand Hybrid High Yes 3,000 34 5.2 22.8 5.2 10 18.4 5.6 101.20
HM4 Maize Hand Hybrid Low Yes 2,550 33 5.2 20.5 5.2 7 16.56 5.04 92.52
LM5 Maize Ox Local High No 2,400 15 5.2 30.4 0 9 18.4 5.6 83.60
LM6 Maize Ox Local Low No 2,040 14 5.2 27.4 0 8 16.56 5.04 76.16
HM7 Maize Ox Hybrid High Yes 4,400 17 5.2 34.2 9.2 10 27.2 10.4 113.20
HM8 Maize Ox Hybrid Low Yes 3,740 15 5.2 30.78 9.2 8 24.48 9.36 102.02
GR1 Groundnuts Hand Local High No 400 32 8 32 0 8 24 46 150.00
GR2 Groundnuts Hand Local Low No 340 30 8 27.2 0 8 20.4 39.1 132.70
GR3 Groundnuts Hand Hybrid High No 700 34 8 27 0 8 40 65 182.00
GR4 Groundnuts Hand Hybrid Low No 595 33 8 22.95 0 7 33 55.25 159.20
GR5 Groundnuts Ox Local High No 500 8 8 24 0 8 32 55 135.00
GR6 Groundnuts Ox Local Low No 425 8 8 20.4 0 7 28.8 46.75 118.95
GR7 Groundnuts Ox Hybrid High No 900 8 8 24 0 9 46 92 187.00
GR8 Groundnuts Ox Hybrid Low No 765 8 8 20.4 0 8 41.4 78.2 164.00
CAS1 Cassava Hand Local Low No 2,600 60 4 24 0 0 16 20 124.00
CAS2 Cassava Ox Hybrid High No 4,000 60 12 48 12 0 44 108 284.00
MIL1 Millet Hand Local Low No 2,000 34 5.2 22.8 0 33 12.4 2.8 110.20
MIL2 Millet Ox Local High Yes 2,800 3.2 5.2 68.4 1.2 32 27.2 10.4 147.60
Sweet
SP1 Potatoes Hand Local Low No 4,000 60 4 24 0 0 16 20 124.00
Sweet
SP2 Potatoes Ox Hybrid High No 12,000 60 12 48 0 0 44 108 272.00
Mixed
MB1 Beans Hand Local Low No 300 40 4 24 0 0 16 14 98.00
Mixed
MB2 Beans Ox Local High Yes 800 40 4 48 2 0 44 38 176.00
PR1 Paddy Rice Hand Hybrid High No 1,600 34 5.2 22.8 0 0 12.4 2.8 77.20
PR2 Paddy Rice Ox Hybrid Low Yes 2,500 3.2 5.2 68.4 1.2 0 27.2 10.4 115.60
55
ANNEX B
Summary of model assumptions
Table B.1 Summary of model assumption for Eastern Province
Code Seed Fert Variable Costs Fixed Costs Total Costs Output Price Gross Returns Net Returns Net Returns
Kg/Ha K/Kg Kg/Ha K/Kg bags/Ha K/bag K/Ha K/Ha K/Kg K/Ha K/Ha K/Labor Day
LM1 20 700 24 1,500 4,050 54,050 700 840,000 785,950 8,153
LM2 20 700 20 1,500 4,050 48,650 700 714,000 665,350 7,706
HM3 20 5,000 300 1,480 36 1,500 6,540 604,540 700 1,540,000 935,460 7,796
HM4 20 5,000 300 1,480 31 1,500 6,540 596,440 700 1,309,000 712,560 6,519
LM5 20 700 27 1,500 4,050 58,550 700 1,680,000 1,621,450 17,740
LM6 20 700 23 1,500 4,050 52,475 700 1,428,000 1,375,525 16,824
HM7 20 5,000 300 1,480 88 1,500 6,540 682,540 700 3,080,000 2,397,460 19,781
HM8 20 5,000 300 1,480 75 1,500 6,540 662,740 700 2,618,000 1,955,260 18,449
GR1 80 2,200 5 1,500 4,050 187,550 2,200 880,000 692,450 3,625
GR2 80 2,200 4 1,500 4,050 186,425 2,200 748,000 561,575 3,462
GR3 80 4,500 7 1,500 6,540 377,040 2,200 1,540,000 1,162,960 5,964
GR4 80 4,500 6 1,500 6,540 375,465 2,200 1,309,000 933,535 5,453
GR5 80 2,200 6 1,500 4,050 189,050 2,200 1,100,000 910,950 6,640
GR6 80 2,200 5 1,500 4,050 187,700 2,200 935,000 747,300 6,181
GR7 80 4,500 10 1,500 6,540 381,540 2,200 1,980,000 1,598,460 8,308
GR8 80 4,500 9 1,500 6,540 379,290 2,200 1,683,000 1,303,710 7,494
SUN1 1 12,800 2 1,500 4,050 19,850 550 440,000 420,150 4,323
SUN2 1 12,800 8 1,500 6,540 31,340 550 880,000 848,660 9,666
MIL1 20 800 40 1,500 4,050 80,050 800 1,600,000 1,519,950 10,382
MIL2 20 800 300 1,480 40 1,500 6,540 526,540 800 2,240,000 1,713,460 13,250
SP1 2,800 36 80 1,500 4,050 224,850 350 1,400,000 1,175,150 9,477
SP2 2,800 70 240 1,500 6,540 562,540 350 4,200,000 3,637,460 30,931
COT1 18 638 200 150 16 1150 5014 64,898 1000 1,200,000 1,135,102 7,839
COT2 14 638 9 1150 5014 24,296 1000 1,020,000 995,704 7,985
COT3 26 638 200 150 13 1150 5014 66,552 1000 1,400,000 1,333,448 12,013
COT4 24 638 9 1150 5014 30,676 1000 1,190,000 1,159,324 12,643
56
Table B. 2. Summary of model assumptions for Southern Province
Fixed Total Output Gross Net Net
Code Seed Fert Variable Costs Costs Costs Price Returns Returns Returns
K/Labor
Kg/Ha K/Kg Kg/Ha K/Kg bags/Ha K/bag K/Ha K/Ha K/Kg K/Ha K/Ha Day
LM1 20 833 24 1,500 4,050 56,710 833 999,600 942,890 9,701
LM2 20 833 20 1,500 4,050 51,310 833 849,660 798,350 9,255
HM3 20 3,500 100 1,560 36 1,500 6,540 286,540 833 1,666,000 1,379,460 11,671
HM4 20 3,500 100 1,560 31 1,500 6,540 278,440 833 1,416,100 1,137,660 10,447
LM5 20 833 48 1,500 4,050 92,710 833 1,999,200 1,906,490 20,611
LM6 20 833 41 1,500 4,050 81,910 833 1,699,320 1,617,410 19,821
HM7 20 3,500 200 1,560 88 1,500 6,540 520,540 833 3,665,200 3,144,660 26,337
HM8 20 3,500 200 1,560 75 1,500 6,540 500,740 833 3,115,420 2,614,680 24,170
GR1 80 1,875 5 1,500 4,050 161,550 1,875 750,000 588,450 3,441
GR2 80 1,875 4 1,500 4,050 160,425 1,875 637,500 477,075 3,135
GR3 80 2,074 7 1,500 6,540 182,960 1,875 1,312,500 1,129,540 6,289
GR4 80 2,074 6 1,500 6,540 181,385 1,875 1,115,625 934,240 5,695
GR5 80 1,875 6 1,500 4,050 163,050 1,875 937,500 774,450 5,645
GR6 80 1,875 5 1,500 4,050 161,700 1,875 796,875 635,175 5,168
GR7 80 2,074 10 1,500 6,540 187,460 1,875 1,687,500 1,500,040 7,895
GR8 80 2,074 9 1,500 6,540 185,210 1,875 1,434,375 1,249,165 7,280
SUN1 5 800 8 1,500 4,050 20,050 800 320,000 299,950 2,747
SUN2 5 800 32 1,500 6,540 58,540 800 1,280,000 1,221,460 12,264
MIL1 20 800 40 1,500 4,050 80,050 800 1,600,000 1,519,950 11,143
MIL2 20 800 300 1,718 40 1,500 6,540 597,940 800 2,240,000 1,642,060 12,698
SP1 2,800 38 80 1,500 4,050 230,450 350 1,400,000 1,169,550 9,432
SP2 2,800 70 240 1,500 6,540 562,540 350 4,200,000 3,637,460 16,239
SOR1 20 667 40 1500 4050 77,390 667 1,334,000 1,256,610 10,266
SOR2 20 667 300 1,580 40 1500 6540 553,880 667 1,867,600 1,313,720 8,666
COT1 18 638 200 150 16 1150 5014 64,898 1000 1,200,000 1,135,102 7,839
COT2 14 638 9 1150 5014 24,296 1000 1,020,000 995,704 7,985
COT3 26 638 200 150 13 1150 5014 66,552 1000 1,400,000 1,333,448 12,013
COT4 24 638 9 1150 5014 30,676 1000 1,190,000 1,159,324 12,643
57
Table B.3. Summary of model assumptions for Northern Province
Fixed Total Output Gross Net Net
Code Seed Fert Variable Costs Costs Costs Price Returns Returns Returns
K/Labor
Kg/Ha K/Kg Kg/Ha K/Kg bags/Ha K/bag K/Ha K/Ha K/Kg K/Ha K/Ha Day
LM1 20 750 24 1,500 4,050 55,050 750 900,000 844,950 9,917
LM2 20 750 20 1,500 4,050 49,650 750 765,000 715,350 9,185
HM3 20 3,315 200 1,800 36 1,500 6,540 486,840 750 2,250,000 1,763,160 17,423
HM4 20 3,315 200 1,800 31 1,500 6,540 478,740 750 1,912,500 1,433,760 15,497
LM5 20 750 60 1,500 4,050 109,050 750 1,800,000 1,690,950 20,227
LM6 20 750 51 1,500 4,050 95,550 750 1,530,000 1,434,450 18,835
HM7 20 3,315 300 1,800 88 1,500 6,540 744,840 750 3,300,000 2,555,160 22,572
HM8 20 3,315 300 1,800 75 1,500 6,540 725,040 750 2,805,000 2,079,960 20,388
GR1 80 1,500 5 1,500 4,050 131,550 1,500 600,000 468,450 3,123
GR2 80 1,500 4 1,500 4,050 130,425 1,500 510,000 379,575 2,860
GR3 80 1,500 8 1,500 6,540 138,540 1,500 1,050,000 911,460 5,008
GR4 80 1,500 7 1,500 6,540 136,740 1,500 892,500 755,760 4,747
GR5 80 1,500 6 1,500 4,050 133,050 1,500 750,000 616,950 4,570
GR6 80 1,500 5 1,500 4,050 131,700 1,500 637,500 505,800 4,252
GR7 80 1,500 11 1,500 6,540 143,040 1,500 1,350,000 1,206,960 6,454
GR8 80 1,500 9 1,500 6,540 140,565 1,500 1,147,500 1,006,935 6,140
CAS1 2,500 40 52 1,500 4,050 182,050 500 1,300,000 1,117,950 9,016
CAS2 2,500 50 80 1,500 6,540 251,540 500 2,000,000 1,748,460 6,157
MIL1 20 800 40 1,500 4,050 80,050 800 1,600,000 1,519,950 13,793
MIL2 20 800 300 1,718 40 1,500 6,540 597,940 800 2,240,000 1,642,060 11,125
SP1 2,800 30 80 1,500 4,050 208,050 350 1,400,000 1,191,950 9,613
SP2 2,800 50 240 1,500 6,540 506,540 350 4,200,000 3,693,460 13,579
MB1 40 1,200 6 1500 4050 61,050 1200 360,000 298,950 3,051
MB2 60 1,500 1.4 31,500 16 1500 6540 164,640 1200 960,000 795,360 4,519
PR1 20 1,500 8 1500 4050 46,050 1900 3,040,000 2,993,950 38,782
PR2 20 1,500 300 1,718 13 1500 6540 571,440 1900 4,750,000 4,178,560 36,147
58
Africa Region Working Paper Series
Series # Title Date Author
ARWPS 1 Progress in Public Expenditure Management in January 1999 C. Kostopoulos
Africa: Evidence from World Bank Surveys
ARWPS 2 Toward Inclusive and Sustainable Development in March 1999 Markus Kostner
the Democratic Republic of the Congo
ARWPS 3 Business Taxation in a Low-Revenue Economy: A June 1999 Ritva Reinikka
Study on Uganda in Comparison with Neighboring Duanjie Chen
Countries
ARWPS 4 Pensions and Social Security in Sub-Saharan October 1999 Luca Barbone
Africa: Issues and Options Luis-A. Sanchez B.
ARWPS 5 Forest Taxes, Government Revenues and the January 2000 Luca Barbone
Sustainable Exploitation of Tropical Forests Juan Zalduendo
ARWPS 6 The Cost of Doing Business: Firms' Experience June 2000 Jacob Svensson
with Corruption in Uganda
ARWPS 7 On the Recent Trade Performance of Sub-Saharan August 2000 Francis Ng and
African Countries: Cause for Hope or More of the Alexander J. Yeats
Same
ARWPS 8 Foreign Direct Investment in Africa: Old Tales November 2000 Miria Pigato
and New Evidence
ARWPS 9 The Macro Implications of HIV/AIDS in South November 2000 Channing Arndt
Africa: A Preliminary Assessment Jeffrey D. Lewis
ARWPS 10 Revisiting Growth and Convergence: Is Africa December 2000 C. G. Tsangarides
Catching Up?
ARWPS 11 Spending on Safety Nets for the Poor: How Much, January 2001 William J. Smith
for
How Many? The Case of Malawi
ARWPS 12 Tourism in Africa February 2001 Iain T. Christie
D. E. Crompton
ARWPS 13 Conflict Diamonds February 2001 Louis Goreux
ARWPS 14 Reform and Opportunity: The Changing Role and March 2001 Jeffrey D. Lewis
Patterns of Trade in South Africa and SADC
59
Africa Region Working Paper Series
Series # Title Date Author
ARWPS 15 The Foreign Direct Investment Environment in March 2001 Miria Pigato
Africa
ARWPS 16 Choice of Exchange Rate Regimes for Developing April 2001 Fahrettin Yagci
Countries
ARWPS 18 Rural Infrastructure in Africa: Policy Directions June 2001 Robert Fishbein
ARWPS 19 Changes in Poverty in Madagascar: 1993-1999 July 2001 S. Paternostro
J. Razafindravonona
David Stifel
ARWPS 20 Information and Communication Technology, August 2001 Miria Pigato
Poverty, and Development in sub-Saharan Africa
and South Asia
ARWPS 21 Handling Hierarchy in Decentralized Settings: September 2001 Navin Girishankar A.
Governance Underpinnings of School Alemayehu
Performance in Tikur Inchini, West Shewa Zone, Yusuf Ahmad
Oromia Region
ARWPS 22 Child Malnutrition in Ethiopia: Can Maternal October 2001 Luc Christiaensen
Knowledge Augment The Role of Income? Harold Alderman
ARWPS 23 Child Soldiers: Preventing, Demobilizing and November 2001 Beth Verhey
Reintegrating
ARWPS 24 The Budget and Medium-Term Expenditure December 2001 David L. Bevan
Framework in Uganda
ARWPS 25 Design and Implementation of Financial January 2002 Guenter Heidenhof H.
Management Systems: An African Perspective Grandvoinnet
Daryoush Kianpour B.
Rezaian
ARWPS 26 What Can Africa Expect From Its Traditional February 2002 Francis Ng
Exports? Alexander Yeats
ARWPS 27 Free Trade Agreements and the SADC Economies February 2002 Jeffrey D. Lewis
Sherman Robinson
Karen Thierfelder
ARWPS 28 Medium Term Expenditure Frameworks: From February 2002 P. Le Houerou Robert
60
Africa Region Working Paper Series
Series # Title Date Author
Concept to Practice. Preliminary Lessons from Taliercio
Africa
ARWPS 29 The Changing Distribution of Public Education February 2002 Samer Al-Samarrai
Expenditure in Malawi Hassan Zaman
ARWPS 30 Post-Conflict Recovery in Africa: An Agenda for April 2002 Serge Michailof
the Africa Region Markus Kostner
Xavier Devictor
ARWPS 31 Efficiency of Public Expenditure Distribution and May 2002 Xiao Ye
Beyond: A report on Ghana's 2000 Public S. Canagaraja
Expenditure Tracking Survey in the Sectors of
Primary Health and Education
ARWPS 33 Addressing Gender Issues in Demobilization and August 2002 N. de Watteville
Reintegration Programs
ARWPS 34 Putting Welfare on the Map in Madagascar August 2002 Johan A. Mistiaen
Berk Soler
T. Razafimanantena
J. Razafindravonona
ARWPS 35 A Review of the Rural Firewood Market Strategy August 2002 Gerald Foley
in West Africa Paul Kerkhof
Djibrilla Madougou
ARWPS 36 Patterns of Governance in Africa September 2002 Brian D. Levy
ARWPS 37 Obstacles and Opportunities for Senegal's September 2002 Stephen Golub
International Competitiveness: Case Studies of the Ahmadou Aly Mbaye
Peanut Oil, Fishing and Textile Industries
ARWPS 38 A Macroeconomic Framework for Poverty October 2002 S. Devarajan
Reduction Strategy Papers : With an Application Delfin S. Go
to Zambia
ARWPS 39 The Impact of Cash Budgets on Poverty Reduction November 2002 Hinh T. Dinh
in Zambia: A Case Study of the Conflict between Abebe Adugna
Well Intentioned Macroeconomic Policy and Bernard Myers
Service Delivery to the Poor
ARWPS 40 Decentralization in Africa: A Stocktaking Survey November 2002 Stephen N. Ndegwa
ARWPS 41 An Industry Level Analysis of Manufacturing December 2002 Professor A. Mbaye
Productivity in Senegal
61
Africa Region Working Paper Series
Series # Title Date Author
ARWPS 42 Tanzania's Cotton Sector: Constraints and December 2002 John Baffes
Challenges in a Global Environment
ARWPS 43 Analyzing Financial and Private Sector Linkages January 2003 Abayomi Alawode
in Africa
ARWPS 44 Modernizing Africa's Agro-Food System: February 2003 Steven Jaffee
Analytical Framework and Implications for Ron Kopicki
Operations Patrick Labaste
Iain Christie
ARWPS 45 Public Expenditure Performance in Rwanda March 2003 Hippolyte Fofack
C. Obidegwu
Robert Ngong
ARWPS 46 Senegal Tourism Sector Study March 2003 Elizabeth Crompton
Iain T. Christie
ARWPS 47 Reforming the Cotton Sector in SSA March 2003 Louis Goreux
John Macrae
ARWPS 48 HIV/AIDS, Human Capital, and Economic April 2003 Channing Arndt
Growth Prospects for Mozambique
ARWPS 49 Rural and Micro Finance Regulation in Ghana: June 2003 William F. Steel
Implications for Development and Performance of David O. Andah
the Industry
ARWPS 50 Microfinance Regulation in Benin: Implications of June 2003 K. Ouattara
the PARMEC LAW for Development and
Performance of the Industry
ARWPS 51 Microfinance Regulation in Tanzania: June 2003 Bikki Randhawa
Implications for Development and Performance of Joselito Gallardo
the Industry
ARWPS 52 Regional Integration in Central Africa: Key Issues June 2003 Ali Zafar
Keiko Kubota
ARWPS 53 Evaluating Banking Supervision in Africa June 2003 Abayomi Alawode
ARWPS 54 Microfinance Institutions' Response in Conflict June 2003 Marilyn S. Manalo
Environments: Eritrea- Savings and Micro Credit
Program; West Bank and Gaza � Palestine for
Credit and Development; Haiti � Micro Credit
62
Africa Region Working Paper Series
Series # Title Date Author
National, S.A.
AWPS 55 Malawi's Tobacco Sector: Standing on One Strong June 2003 Steven Jaffee
leg is Better than on None
AWPS 56 Tanzania's Coffee Sector: Constraints and June 2003 John Baffes
Challenges in a Global Environment
AWPS 57 The New Southern AfricanCustoms Union June 2003 Robert Kirk
Agreement Matthew Stern
AWPS 58a How Far Did Africa's First Generation Trade June 2003 Lawrence Hinkle
Reforms Go? An Intermediate Methodology for A. Herrou-Aragon
Comparative Analysis of Trade Policies Keiko Kubota
AWPS 58b How Far Did Africa's First Generation Trade June 2003 Lawrence Hinkle
Reforms Go? An Intermediate Methodology for A. Herrou-Aragon
Comparative Analysis of Trade Policies Keiko Kubota
AWPS 59 Rwanda: The Search for Post-Conflict Socio- October 2003 C. Obidegwu
Economic Change, 1995-2001
AWPS 60 Linking Farmers to Markets: Exporting Malian October 2003 Morgane Danielou
Mangoes to Europe Patrick Labaste
J-M. Voisard
AWPS 61 Evolution of Poverty and Welfare in Ghana in the October 2003 S. Canagarajah
1990s: Achievements and Challenges Claus C. P�rtner
AWPS 62 Reforming The Cotton Sector in Sub-Saharan November 2003 Louis Goreux
Africa: SECOND EDITION
AWPS 63 Republic of Madagascar: Tourism Sector Study November 2003 Iain T. Christie
(E) D. E. Crompton
AWPS 63 R�publique de Madagascar: Etude du Secteur November 2003 Iain T. Christie
(F) Tourisme D. E. Crompton
AWPS 64 Migrant Labor Remittances in Africa: Reducing Novembre 2003 Cerstin Sander
Obstacles to Development Contributions Samuel M. Maimbo
AWPS 65 Government Revenues and Expenditures in January 2004 Francisco G. Carneiro
Guinea-Bissau: Casualty and Cointegration Joao R. Faria
Boubacar S. Barry
63
Africa Region Working Paper Series
Series # Title Date Author
AWPS 66 How will we know Development Results when we June 2004 Jody Zall Kusek
see them? Building a Results-Based Monitoring Ray C. Rist
and Evaluation System to Give us the Answer Elizabeth M. White
AWPS 67 An Analysis of the Trade Regime in Senegal June 2004 Alberto Herrou-Arago
(2001) and UEMOA's Common External Trade Keiko Kubota
Policies
AWPS 68 Bottom-Up Administrative Reform: Designing June 2004 Talib Esmail
Indicators for a Local Governance Scorecard in Nick Manning
Nigeria Jana Orac
Galia Schechter
AWPS 69 Tanzania's Tea Sector: Constraints and June 2004 John Baffes
Challenges
AWPS 70 Tanzania's Cashew Sector: Constraints and June 2004 Donald Mitchell
Challenges in a Global Environment
AWPS 71 An Analysis of Chile's Trade Regime in 1998 and July 2004 Francesca Castellani
2001: A Good Practice Trade Policy Benchmark A. Herrou-Arago
Lawrence E. Hinkle
AWPS 72 Regional Trade Integration inEast Africa: Trade August 2004 Lucio Castro
and Revenue Impacts of the Planned East African Christiane Kraus
Community Customs Union Manuel de la Rocha
AWPS 73 Post-Conflict Peace Building in Africa: The August 2004 Chukwuma Obidegwu
Challenges of Socio-Economic Recovery and
Development
AWPS 74 An Analysis of the Trade Regime in Bolivia August 2004 Francesca Castellani
in2001: A Trade Policy Benchmark for low Alberto Herrou-Aragon
Income Countries Lawrence E. Hinkle
AWPS 75 Remittances to Comoros- Volumes, Trends, October 2004 Vincent da Cruz
Impact and Implications Wolfgang Fendler
Adam Schwartzman
AWPS 76 Salient Features of Trade Performance in Eastern October 2004 Fahrettin Yagci
and Southern Africa Enrique Aldaz-Carroll
AWPS 77 Implementing Performance-Based Aid in Africa November 2004 Alan Gelb
Brian Ngo
Xiao Ye
AWPS 78 Poverty Reduction Strategy Papers: Do they December 2004 Rene Bonnel
matter for children and Young people made Miriam Temin
vulnerable by HIV/AIDS? Faith Tempest
AWPS 79 Experience in Scaling up Support to Local December 2004 Jean Delion
Response in Multi-Country Aids Programs (map) Pia Peeters
in Africa Ann Klofkorn Bloome
64
Africa Region Working Paper Series
Series # Title Date Author
AWPS 80 WHAT MAKES FDI WORK? A PANEL February 2005 Kevin N. Lumbila
ANALYSIS OF THE GROWTH EFFECT
OF FDI IN AFRICA
AWPS 81 Earnings Differences between Men and Women in February 2005 Kene Ezemenari
Rwanda Rui Wu
AWPS 82 The Medium-Term Expenditure Framework April 2005 Chukwuma Obidegwu
The Challenge of Budget Integration in SSA
countries
AWPS 83 Rules of Origin and SADC: The Case for change June 2005 Paul Brenton
in the Mid Term Review of the Trade Protocol Frank Flatters
Paul Kalenga
AWPS 84 Sexual Minorities, July 2005 Chukwuemeka
Violence and AIDS in Africa Anyamele
Ronald Lwabaayi
Tuu-Van Nguyen, and
Hans Binswanger
AWPS 85 Poverty Reducing Potential of Smallholder July 2005 Paul B. Siegel
Agriculture in Zambia: Jeffrey Alwang
Opportunities and Constraints
65