Potential impact of improved fallows on small farm livelihoods, Eastern Province, Zambia

Material Information

Potential impact of improved fallows on small farm livelihoods, Eastern Province, Zambia
Grier, Christina E
Publication Date:
Physical Description:
vii, 129 leaves : ill. ; 29 cm.


Subjects / Keywords:
Fallowing -- Zambia -- Eastern Province ( lcsh )
Rural development -- Economic aspects -- Zambia -- Eastern Province ( lcsh )
Farms, Small -- Zambia -- Eastern Province ( lcsh )
Food and Resource Economics thesis, M.S ( lcsh )
Dissertations, Academic -- Food and Resource Economics -- UF ( lcsh )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Thesis (M.S.)--University of Florida, 2002.
Includes bibliographical references (leaves 121-128).
General Note:
General Note:
Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.
Statement of Responsibility:
by Christina E. Grier.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact Digital Services ( with any additional information they can provide.
Resource Identifier:
002822774 ( ALEPH )
50512035 ( OCLC )
ANV1301 ( NOTIS )

Full Text

First, I would like to thank the farmers in Eastern Zambia who took the time out from their very busy days to talk to me about their farming practices. Without their help ths study would not have been possible. I would also like to thank Martha Mwale, my translator, without whom I would not have been able to conduct farmer interviews. She introduced me to some aspects of village culture that I may not have otherwise been able to observe and participate in without her. To all who opened their homes to me, fed me, and shared their life with me, I want to express my deepest, heartfelt thanks.
I would like to thank Donald Phiri and his family for sharing their home with me and feeding me when I was not in the village. Donald and his family introduced me to life in Chipata town and provided me with a surrogate family while so far from home. Their assistance was invaluable to my research in Eastern Zambia.
I would like to thank Dr. Kasim Masi and the staff at World Vision for
transporting me to the villages and making arrangements for my stay there. The staff there were always accommodating to my every need. I really appreciate all of their efforts. In addition, World Vision, through Dr. Christina Gladwin, funded my research and travel to Eastern Zambia, without this funding I would not have been able to conduct this research.
I would like to thank my committee for their help on this project. I really
appreciate the patience and direction that Dr. Peter Hildebrand and Dr. Clifton Hiebsch

provided during my master's career. Without their ideas and direction I do not think I would have been able to complete this thesis. I also owe a debt of gratitude to Dr. Chris Andrew, who was always my best cheerleader when times were tough. There is no way that I can repay these people for the kindness and direction they provided during my time in the master's program.
I want to thank my mom, dad, sisters, Amy and Amanda Grier, Aunt Patty and Uncle Don, and all members of my extended family for always standing by me and providing me with the much needed encouragement to stick it out when I was ready to throw in the towel. I want to thank my friends Grace Wong, Alicia and Steve Moore, Craig Perkins, Gus Woodruff, Tammy Vesper, Tom and Robin Morrissey, Mike and Susie Hetrick, Gracy and Mike Castine, Effie Philippakos, Merryl Malter, Charles Counts, Peter Alcorn, Kevin Blythe, Brad and Craig Plaketta, Steve Forguson, the rest of the members of the ridewet gang, and Gainesville friends for providing me with food when I forgot to eat, with company, laughs, and continual support over the years leading to the completion of this thesis. Words are not enough to express how much their support and encouragement has meant to me. I really would have not been able to accomplish what I have in my life without them. Their loyalty, encouragement, and belief in me have been and remain very precious to me. I will always value their friendship and love.

Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science
Christina Grier
May 2002
Committee Chair: Peter E. Hildebrand Major Department: Food and Resource Economics
Soil infertility, due to continuous cultivation, has been cited as one of the reasons productivity of small farms in Eastern Zambia has declined in recent years. Continuous cultivation on small farms without the use of fertilizer has depleted soil nutrients, resulting in reduced maize yields. Lower yields have led to lower farmer incomes and, in some cases, the inability to feed the household throughout the entire year. Improved fallow technologies (IFT's) are a proposed fertilizer alternative used to replenish nutrients removed from the soil. It has been found that farmers who adopt IFT's realize better maize yields and are, theoretically, able to maintain household food stocks throughout the year.

Using ethnographic linear programming, this thesis attempts to evaluate the effects of IFT adoption on households in Zambia. In particular, it distinguishes the effects of adoption on two different household types, female-headed and male-headed, and describes changes in household resource allocation and income generation with the adoption of improved fallow technologies.

ACKNOWLEDGMENTS ...................................................................... ii
ABSTRACT ...................................................................................... iv
I INTRODUCTION ............................................................................ I
Chapter Overview .............................................................................. I
2 LITERATURE REVIEW ..................................................................... 12
Chapter Overview ............................................................................... 12
Farming Systems ................................................................................ 12
Sustainable Livelihoods Approach ........................................................... 16
Food Security and Nutrition .................................................................. 22
Gender ............................................................................................ 23
Soil Fertility and Improved Fallows ......................................................... 25
3 FARMING SYSTEMS DESCRIPTION OF STUDY AREA ............................ 29
Chapter Overview ............................................................................... 29
Administration and Demographics ........................................................... 30
Topography and Climate ....................................................................... 35
Soils ............................................................................................... 36
Farming and Livelihood Systems .............................................................. 37
4 DATA COLLECTION AND ANALYSIS .................................................. 49
Chapter Overview ............................................................................... 49
Methods of Data Collection .................................................................... 49
Linear Programming ............................................................................ 51
Results ........................................................................................... 61
Conclusion ....................................................................................... 64

5 CONCLUSION AND RECOMMENDATIONS ......................................... 65
Linear Programming ............................................................................. 66
Recommendations for Further Study ........................................................ 67
A TIMELINE OF GOVERNMENT POLICIES .............................................. 69
Maize ............................................................................................. 73
Soil Fertility and Improved Fallows ......................................................... 76
C QUESTIONNAIRE ............................................................................ 82
D LAND AND CROP PRODUCTION, AND OTHER INCOME GENERATING ACTIVITY TABLES ......................................................................... 94
E LP MODEL RESULTS ...................................................................... 107
REFERENCES ................................................................................... 121
BIOGRAPHICAL SKETCH .................................................................... 129

Chapter Overview
There is a need in Sub-Saharan African to feed the increasing population while at the same time conserve natural resources in order to meet future needs. In order to combat food insecurity, researchers have focused on increasing agricultural yields by improving present farm management and soil fertility practices. Poor soil fertility is the biophysical root cause of declining yields in agriculture (Sanchez et al. 1997). Improved fallow technologies (IFTs) are one of the methods currently being researched to improve soil fertility. Economic cost/benefit analyses have been done on the effects of adopting IFTs from the household perspective (Franzel 1999; Kwesiga and Beniest 1998; Opio 2001; Sanchez et al. 1997; DeWolf and Rommelse; Montagnini and Mendelsohn 1997). Decision studies have been conducted in order to understand and be able to predict why a household chooses to adopt certain technologies (Gladwin et al. 2000; Peterson 1999b). This study examines different households and how outside factors, such as economic policies, affect a farmer's decision to adopt IFTs.
For the past two decades the focus of development workers on Sub-Saharan
Africa has been on economic development with a particular emphasis on food security. A person is considered food secure if he or she has the "physical and economic access to an adequate food supply without undue risk of losing such access" (Thomson and Metz 1997). In Africa a large portion of the rural population is comprised of small farm

households that not only produce their own food, but also supply food for urban areas. Low purchasing power of farm households, inadequate marketing and distribution systems, and poor infrastructure limits access to food. Without the availability of needed resources such as land, credit, and labor, households have difficulty producing food or generating cash in order to buy food. Less food equals reduced health and strength to work and, ultimately, poverty. Access to a sufficient amount of food directly influences the health and productivity of laborers in the economy.
The area of focus in this study is Eastern Zambia. Of the 9.6 million people living in Zambia, more than 60% derive their livelihood from agriculture and approximately
5 1% of those are rural, small-scale farmers'I (Saasa et al. 1999). Seventy-three percent of the population lives below the poverty level and 27% of the children suffer from malnutrition (The World Bank Group 2001). In Zambia, small-scale farmers produce the majority of maize, an important food staple for the entire population. Their maize production is extremely vulnerable to variations in climatic conditions; droughts in -1992, 1995, and 1998 destroyed many crops and devastated the maize supply. Continuous cropping has depleted the soil of nutrients and made it difficult for farmers to sustain ample harvests. Combined with a series of shifts in economic policy since independence in 1964, farmers have had difficulty managing their farms efficiently and gaining access to needed resources. These obstacles have severely affected the country's food crop production, thus threatening the food security and sustainability of many households.
Structural adjustment is an economic stabilization process whereby policies are designed to make the economy more productive through a more efficient use of available
Small-scale farmers are characterized as cultivating on 5 hectares or less.

resources. It relies on the intensive use of the private sector, with decentralization of markets, and the removal of exchange rate, price controls, and other biases against exports. By removing market and price distortions, it is believed that income distribution and thus, adequate food intake will improve. Structural transformation is an essential component of long-term development (Tomich et al. 1995). It occurs when economic activity and the distribution of labor shift from traditional, labor-intensive agriculture to modem, capital intensive industry and services (Tomich et al. 1995; O'Brien 1991). Transformation also changes the structure of demand. As income rises and income elasticities change, the demand for different commodities changes with it and food consumption declines as a share of expenditure. Consumption then increases in processed food products, manufactured goods, and modem services. Characteristics of developing countries include high population growth rates of 2.5 to 4 percent per year, fifty or more percent of the labor force in agriculture, low labor productivity in agriculture, the inequitable distribution of land, poor infr-astructure, and politically fragile governments (Tomich et al. 1995). For countries with abundant rural labor to reach their structural transformation point, which is marked by a relative decline in the agricultural sector, they must increase productivity in agriculture, decrease the absolute size of their agricultural labor force, and increase the labor force in the non-agricultural sector.
Zambia is a developing country with a population growth rate of 2.5% from 1993 to 1999, and, as previously stated, more than 60% of the population derives their livelihood from agriculture (World Bank 1999; Saasa et al. 1999). When farmers are unprepared for shocks, such as drought, floods, sickness, or subsidy removals, their productivity suffers. Low productivity limits the amount of cash income that can be

generated, thereby restricting the fanner from purchasing needed agricultural inputs, such as tools, seeds, and fertilizer. Lower cash income inhibits economic growth and results in reduced commodity flows from the agricultural sector to the manufacturing and service sectors. Stagnated growth reduces the demand for laborers in the non-agricultural sectors and lowers wages overall. Fewer jobs in the manufacturing sector encourage rural people to stay in subsistence agriculture.
From the time Zambia gained its independence from British colonial rule in 1964, various economic structural adjustment policies aimed at stabilizing the economy have been implemented, modified, and abandoned by the government (see Appendix A for a historic timeline of economic policies). Each had a direct impact on the agricultural sector by undermining productivity and efficiency and these frequent changes have damaged the trust Zambian citizen's may have had in the government (Jansen and Rukovo 1992; Valdes and Muir-Leresche 1993; Geisler 1992; Copestake 1997; Wood et al. 1990; Wichern et al. 1999). Under the Kaunda regime, from 1964 to 1991, policies were designed to help the sector increase its productivity, reach self-sufficiency in primary food crops, and increase the production of exportable agricultural commodities. From 1964 to 1982, measures such as cooperatives, state farms, and parastatal enterprises were used to achieve the country's crop production goals. Copper revenues from the government controlled mining industry funded most of the programs. In 1975, when the copper price collapsed, government expenditures were about 41 % of the GDP, 12% of which were maize subsidies (Wood et al. 1990, Valdes and Muir 1993). As a result of reduced copper revenues, the government began to borrow from external sources to fund their programs. By the end of 1982, external liabilities amounted to $4.5 billion,

approximately twenty percent of the GDP. Deficits of this magnitude could not be maintained and the government, under pressure of the International Monetary Fund (IMF), had to relinquish controls over agricultural marketing and prices. In 1986, all maize prices were deregulated and the price of maize was allowed to increase. Residents of the Copperbelt region rioted in protest and President Kaunda reacted by reinstating consumer maize subsidies. Although crop production from 1964 to 1989 increased an average rate of 2.5% per year, it still could not keep pace with the 3.7% annual population growth rate. In response, during the period from 1986 to 1991, the government announced two additional reform packages each with similar aims of increasing productivity and enhancing growth.
In 1991, Frederick Chiluba, of the Movement for Multi-Party Democracy (MMD), was elected with a commitment to stabilize the economy and promote economic growth. His plans were to reduce government spending through the removal of subsidies, balancing Zambia's budget, reducing civil service, and encouraging growth in the private sector through privatization of state-run enterprises (Jansen and Rukovo 1992). To date the agricultural marketing sector has been decentralized, fertilizer and maize subsidies have been removed, the milling industry has been privatized, fuel prices and interest rates have increased, and the Zambian Kwacha has been devalued. The elimination of price controls has caused seed and fertilizer prices to increase to a point that makes them unaffordable to many small farmers. The issue is complicated by the fact that the soils in the Eastern province are nutrient depleted from continuous use and farmers have trouble supplementing the needed nutrients without fertilizer. Consequently, small farm

production has decreased leaving barely enough maize to feed the household for an entire year.
Farmers have responded by planting more local rather than hybrid maize and/or shifting more of their production to cash crops such as cotton and tobacco (Peterson 1999a). Farmers also compensate for lower yields by finding ways to generate income in order to buy food to supplement the year's weak harvest. However, off-farm work detracts from the already limited labor supply in the household taking away from the quantity and quality of work that is needed in the fields. The decreased soil fertility, due to naturally poor soils or continuous cropping, is perhaps the greatest obstacle to increasing farm output in Eastern Zambia. Agricultural subsidies provided farmers with easy access to needed inputs, such as hybrid seed and fertilizer, allowing them to continuously cultivate the same field for several years. Farmers have been continuously cultivating the same fields since the mid-1950s (Priestley in Peterson 1999a) and maize yields without the use of fertilizer are 2.9 tons per hectare lower than with the use of fertilizer (Kwesiga and Beniest 1998).
In an effort to understand the issues surrounding the dwindling agricultural production and productivity in Zambia, the International Center for Research in Agroforestry (ICRAF), the World Vision Zambia Integrated Agroforestry Project (ZIAP), and the University of Florida have joined forces as part of a USAID-funded Collaborative Research Support Program (CRSP) for "Gender and Soil Fertility in Africa." ICRAF researchers have been working to improve the traditional fallow practice by testing plant species that return nutrients to the soil in a shorter time or on a smaller area. Improved fallows technologies (IFTs) and soil conservation techniques have been proposed as an

economically viable alternative to fertilizer use to replenish nutrients in the soils. IFTs have a large initial labor requirement, some species more so than others, but once planted require little attention over the next two seasons. Farmers who adopt the IFTs realize increased maize yields in fields where the improved fallows have been planted.
The species that have been tested at Msekera Research Station and on-farm include Sesbania sesban, Tephrosia vogeifi, Gliricidia sepium and Cajanus cajun. ICRAF scientists have found that by allowing these trees to grow on land with nutrientdepleted soils for two to three years and then returning the top growth to the field, essential nutrients will be restored, soil organic material will be increased, and soil structure will improve. After the fallow period, the trees are cut down and maize is planted in the same field. Farmers who have adopted this technology have seen improvement in their maize yields for up to two to three years after the fallow. At times, under certain conditions, yields can be greater than if fertilizer were used (Kwesiga, Sanchez, Franzel). The improved fallow trees also help to control soil erosion, suppress the growth of weeds, and provide fuel wood for the household. The goal of ZIAP is to "improve household food security, through increased agricultural productivity (yields) and increased market participation" (Peterson 1999a). World Vision plans to achieve their goal by providing rural farmers with soil management and agricultural production techniques, including, but not limited to, improved fallows. Conceptual Approach of Sustainable Livelihoods
Webster's Dictionary (1986) defines a system as a "regularly interacting or
interdependent group of elements forming a unified whole" and a livelihood is a "means of support or subsistence." A livelihood system is one in which a household combines

different resources (material and social) and activities as a way to make a living (DFID 1999). Livelihood strategies are developed in reaction to changes in the political, social, cultural, or biophysical environment. In order to determine the best way to address problems of hunger and poverty among households, an in-depth understanding of household livelihood systems and strategies is essential.
The sustainable livelihoods approach attempts to analyze and understand the livelihoods of people. By recognizing that there are multiple influences on people, the framework can be used as an aid to identify how people combine their limited resources to attain a sustainable existence. "A livelihood is sustainable when it can cope with and recover ftrm stresses and shocks and maintain or enhance its capabilities and assets both now and in the future, while not undermining the natural resource base" (DFII) 1999). Resources are the natural, financial, human, physical, and social capital that is available (Scoones 1998; DFID) 1999; Ashley and Carney 1999).
By understanding the effects of household composition on livelihood strategies, researchers are better prepared to make recommendations for enhancing growth and well being. Household composition, in this case, influences available farm labor and consumption requirements for household members. For instance, a newly established household consisting of a few people may be limited on labor but have less food and money requirements than a mature household with eleven members. Purpose of Research
This study aims to determine whether the adoption of improved fallows is a
sustainable practice for households in Eastern Zambia. The main purpose of IFTs is the maintenance of soil fertility, which involves returning nutrients to the soil that were

removed by harvests, runoff, erosion, and leaching. An additional goal of the improved fallows is to help farmers increase agricultural productivity without having to use large amounts of expensive fertilizer. By improving productivity, farmers will be more equipped to sustain themselves and their families. Improvements in productivity will enable a farmer to generate extra income in times when there are harvest surpluses to sell in the market.
Using a linear programming model patterned after a typical household in the
Eastern Province of Zambia, this analysis looks at whether adopting improved fallows as a soil conservation technique is a productive and efficient strategy for farmers given their particular constraints and resources. Linear programming is a tool that is useful for analyzing a typical family farm system. Designing a linear programming model (LP) involves identifying the production, income, and consumption patterns of farm households with the purpose of maximizing or minimizing one or more household objectives, for example maximizing annual discretionary cash and/or meeting food consumption requirements for the year. Once a base model is created that adequately simulates the household systems, alternatives to production, such as the improved fallows, can be incorporated and examined.
Models of households disaggregated into female-headed households and maleheaded households will be used for analysis. Examples of farms that utilize improved fallow technologies will be compared to farms that do not use the technology. The objective of both models will be the same, to maximize end of year cash available for discretionary spending. Both models will include on- and off-farm activities, such as the production and sale of crops and other income generating activities, and constraints,

including land, labor, cash, and consumption requirements of small-scale farms in Eastern Zambia. It examines how household composition affects what activities are adopted in order to survive and how changes at the macro level may alter household strategies. By examining the macro and micro effects on the household this research can more easily identify how and why certain households adopt different strategies in order to survive. Limitations of the Study
1. Th e project is in the early stages of development and only a limited number of
farmers that have tested and adopted the technology could be interviewed. Therefore,
there is little data on actual on-farm yields of maize after the adoption of the IFTs.
2. The sample size of field collected data is small and had to be combined with ICRAF
and World Vision data in order to get a complete description of the farming system
and to augment data collected in the field.
1. Describe changes in resource allocation and income generation with the adoption of
improved fallow technologies.
2. Assess the potential for adoption of improved allows by household type, i.e. femaleheaded households and male-headed households. Examine how household
composition affects livelihood strategies.
1. The use of improved fallow technologies in Eastern Zambia will increase small-farm
maize yields and improve livelihood systems for farmers.
Hypothesis one states that if farmers in Eastern Zambia adopt improved fallow technologies their productivity will increase and they will be better off economically. The main concern of small-scale farmers is to allocate scarce resources in order to sustain the home and family. With over 60% of the population engaged in agriculture, and over
5 1 % of this number farming on a subsistence level, small-scale farmers provide a significant proportion of the total agricultural production and represents more than 40%

of household consumption. An understanding of small-scale farming systems is necessary to understand why farmers are and are not adopting the improved fallows. Further, a complete analysis of the system will provide an aid to policymakers to implement policies that are relevant and favorable to this large rural sector.
2. Since labor is a limiting factor in most female-headed households, the IFT that has the smallest labor requirements will be the one that is most readily adopted.
IWhile there is enough arable land in Zambia for farmers to farm, most
households, especially female-headed households, do not have the available labor to plant and cultivate larger areas. With this in mind, labor-intensive technologies may not be
-appropriate for some households to adopt and the improved fallow species that requires the least amount of labor may be the most readily adopted.
The information gained from this study will be given to UF Soils CRSP and World Vision, the NGO that is promoting the IFTs. With it, the researcher hopes that World Vision will better target certain technologies to the appropriate households.
The contents of this thesis include a literature review, Chapter 2 that describes farming systems and the sustainable livelihood concept as it relates to gender, nutrition and food security. A description of the four improved fallow species considered in the project and their effects on soil fertility and maize yields is also included. Chapter 3 provides a description of the farming systems of the farmers in the villages that were visited. Chapter 4 presents methods of data collection and analysis through linear programming. Included in this chapter is an explanation of linear programming and its uses and the results from the model. Conclusions and recommendations from the research are presented in Chapter 5.

Chapter Overview
This chapter presents a literature review beginning with a description of a generalized subsistence farming system in a developing country and the problems farmers encounter. It describes the concept of sustainable livelihoods as it relates to agric ultural productivity and the influence of this approach on food security, nutrition and the sustainability of a household. Since women play a vital role in food production in Sub-Sahiaran Africa, a particular emphasis is placed on gender differences in labor and productivity. Finally, soil fertility management practices, with a specific emphasis on improved fallow technologies, are described.
Farming Systems
A system is a whole that is made up of many interrelated and interacting parts. There are many different types of systems: from ecosystems, farming, and biological, to social, cultural, or business. Within every system there also exist sub-systems each having a different level of complexity. For instance, Hart, in Hildebrand (1986), explains that there is a hierarchical relationship between agricultural systems in a region. "The regional agricultural system includes the farms in the area; the marketing, credit, and information systems; and the infrastructure that ties the regional subsystems together" (Hart 1986). The farm system consists of agroecosystems, in which the production of crops or animals or both takes place. Agroecosystems are similar to natural ecosystems

in that they are "composed of a biotic community of plants, animals, and microorganisms and the physical environment in which the community functions" (Hart 1986). The difference is that at least one of the plants or animals within the agroecosystem has some value to and is managed by the farmer. Within every crop agroecosystem exist separate subsystems of crops, soil, weeds, microorganisms, and insects. The crop subsystem is comprised of the crop populations that process energy, nutrients, and water to produce crop yields (Hart 1986). The animal agroecosystemn includes domesticated animals, soil, weeds, microorganisms, insects, and pasture subsystems. The animal subsystem is an "arrangement of animal populations that processes energy and material inputs to produce meat and animal products" (Hart 1986).
The farming system also includes household's livelihood strategies, or the socioeconomic subsystem, where household decisions and farm production are closely related: a home first and business second (Hildebrand 1986a). The household members at the center make decisions about the allocation of resources the household possesses in order to attain particular goals (Norman 1986). According to Scoones (1998), "a livelihood system comprises the capabilities, assets (both material and social resources) and activities required for a means of living." Exogenous factors from other systems, such as the ecosystem, government and economic systems, and social systems help to determine how the small-scale farm is managed (Norman 1986). Figure 2-1, taken from Hart (1986), illustrates the hierarchical relationship between these factors in an agricultural system; taken together it represents a livelihood system.

Hart (1986) explains that when applying an agricultural systems approach to a specific case, one need not focus on the entire hierarchy, as it may not be "necessary or practical." However, analyses of the levels immediately above and below are wan-anted

A Region
Market, Farm NonCredit, Systems agricultural
Information Systems
I %
/" Farm System
Socio-economic Subsystem Crop Animal
Agroecosystejm Agroecosystem
-- Crop Aeroecosvstem Animal Aeroecosstem ..
Soil Weed Ie Soil Weed Insects
Sub- Subsystem Insects Sub- Subsystem
sys- sys- Micro:organisms
em Crop Micro- tem Pasture
Subsystem organisms Subsystem Animal
Crop Sym Subsystem
sAnimal System CroD 1 Animal 1
Croo 2 --+ Animal 2
Croo N Animal N
It 11 I I
A Crop An Animal
Figure 2-1: Hierarchical Relationship Between Agricultural Systems

in order to measure the inputs and outputs of each system and understand how each operates. Farm productivity is an important focus because a farming system is a smaller system within the larger economic system. Its productivity affects the efficiency of the entire livelihood system. If farmers in a country can only produce a limited amount of a product, then the availability of that product within the country's economic system becomes limited as well (Hildebrand 1986b).
Sustainable Livelihoods (SL) Approach
People have many different methods of meeting their livelihood objectives and access to a variety of resources helps them realize their goals. "A livelihood is sustainable when it can cope with and recover from stresses and shocks', maintain or enhance its capabilities and assets, while not undermining the natural resource base" (Scoones 1998). "Sustainable agricultural production incorporates the idea that natural resources should be used to generate increased output and incomes without depleting the natural resource base" (Gruhn et al. 2000). Figure 2-2 presents the main elements of the sustainable livelihoods fr-amework as presented by Scoones (1998). This framework can be used as an approach to understanding livelihoods of individuals, households, villages, cities, or whole countries. The household livelihood system is the main focus of this study.
A stress is "a small, regular, predictable disturbance with cumulative effects (Scoones 1998)", such as seasonal shifts in prices, employment, or food availability (DFID 1999). A shock is "a large infrequent, unpredictable disturbance with immediate impact (Scoones 1998), for example, a war, flood, or drought.

POLICY Livelihood
E 1. Poverty
History Natural Capital V Agricultural 2. Well-being
Intensification/ and
Politics Economic/Financial Extensification capabilities
Macro-economic Capital Institutions improved
ConditionsHuman Capital and Livelihood Sustainabilitv
Terms of Trade Organizations Diversification 3. Livelihood
Climate vulnerability
Social Capital and resilience
Agro-ecology enhanced
Physical Capital
Demography Migration 4. Natural
resources base
Social sustainability
Differentiation ensured
Figure 2-2: Sustainable Rural Livelihoods Framework for Analysis
Scoones 1998

In Eastern Zambia small, rural farmers2 depend on the land to grow crops to feed themselves and their family. They rely on good health in order to have the ability to work on the farm or off-farm for an outside employer. Credit services provide avenues for farmers to obtain needed inputs for their crops, and educational resources, such as agriculture extension, enable farmers to improve their farming techniques. Farmers also depend on friends and extended family in times of need. In the sustainable livelihood framework these resources are defined as capital. Capital, in the economic sense, is defined as physical assets, such as equipment, buildings, and inventories, which can be used (along with labor and raw materials) to produce output (Pindyck and Rubinfeld 1997). In the SL framework, capital has five categories that form the "building blocks" of a livelihood. They are human, social, natural, physical, and financial (Scoones 1998; DFID 1999; Hoon et al. 1997). A combination of each of these asset categories is needed for a successful livelihood outcome. Financial capital alone is not sufficient to maintain a farm if a farmer is in poor health or if the land farmed is infertile. Therefore, in order to insure survival, households have to find ways of combining the assets available to them in productive ways (DFID 1999). The ways a household reacts to certain conditions reflects its livelihood strategies.
Human capital, according to the DFID Guidance Sheets, is the health, knowledge, skills, and physical capabilities of the members of a household (DFID 1999; Scoones 1998). Health, education, and ability to labor are all necessary in order to utilize the other types of assets and achieve a positive livelihood outcome (DFID 1999). A study by Deininger and Olinto (2000) indicates that family labor is an important determinant of 2 A small farmer in Zambia is considered to be one who farms on five hectares or less. The term 'rural farms' is used because many people farm home-gardens in non-rural areas (Hildebrand personal communication).

output; that is, larger households tend to be more productive than smaller households. Furthermore, the study indicates that farmers with more education have the ability to overcome market imperfections, as more educated farmers tend to be more integrated into output markets and demand higher amounts of fertilizer and credit per hectare. A poverty profile conducted by Alwang et al. (1996) indicated that households whose head has no formal education are more likely to remain in poverty than those whose head has some education. This is particularly true for female-headed households (FHH-) who typically have little to no education. In Zamrbia, FHHs averaged 1.8 years of education versus an average of 4.2 years for households where a male and female are present (Due 1991).
Social capital is composed of the social resources, such as clubs, associations, networks, and relationships of trust, that people use in pursuit of their livelihood objectives (DFID 1999; Scoones 1998; Hoon et al. 1997). For example, there is a threemonth hunger period from January to March, where some of the poorer households in Eastern Zambia, particularly female-headed households, suffer from food shortages. In order to compensate for this lack of food, members of the household will trade a day's worth of labor in the future in exchange for approximately 5 to 10 kg of maize now. In another case female members of the Lufu village have a well-organized women's club where they pull together food supplies in order to provide food to member households that are suffering during the hunger months. These examples show how social capital provides a kind of buffer to help households cope with shocks. Social networks also "facilitate innovation, the development of knowledge and sharing of that knowledge" (DFID 1999). In the case of the improved fallows, early adopters of the technology

provide an example of the benefits of improved fallow to other farmers who are reluctant to adopt because of other constraints, such as labor. Farmers communicate their knowledge of new technologies with one another may replace or enhance a lack of agricultural extension in the area. The sharing of labor within clubs helps smaller households increase productivity due to a lack of available household labor.
Examples of natural capital include soil fertility, land, forests, trees, water,
nutrients, and air. In essence, natural capital includes the flows and services provided by natural resource stocks that are useful for supporting a livelihood (DFID 1999; Scoones 1998). Failure of crops due to natural effects has a direct link to the health and food security of households. A big obstacle to increased productivity of small farmers in Eastern Zambia is poor soil fertility, an "important form of renewable natural capital" (Sanchez et al. 1997). Other constraints to food production in recent years have been droughts and floods that have devastated the livelihoods of the poor in rural areas. Understanding the flows and services of ecosystems is important in maintaining asustainable system.
Physical capital includes the infr-astructure and producer goods needed to support livelihoods and to produce output (DFID 1999). They are the services, such as transport, water supply, sanitation, and access to information, that help people meet their essential requirements (DFID 1999). Lack of access to necessary producer goods directly affects farmer productivity and their capacity to provide for the household because more time and effort is spent on meeting basic needs (DFID 1999). Inadequate roads and remoteness impede a farmer's ability to take harvests to market. At the same time, a poor road system makes it difficult for producers to bring needed inputs to the farms.

Increased costs of transport affect the prices of needed goods putting both consumers and producers at a disadvantage. The absence of farm extension or educational facilities could cause rural people to miss out on beneficial opportunities and technologies.
Economic or financial capital, which includes cash, credit/debt, and savings, is a resource required for achieving any livelihood strategy (DFID 1999; Scoones 1998; Pindyck and Rubinfeld 1997). Deininger and Olinto (2000) state that farmers who are unable to afford fertilizer "forgo significant productivity and welfare benefits." Their estimates suggest access to fertilizer would increase their income by about US$85 and access to credit has the potential to increase farmer output by 2.6 to 4%. Financial capital helps to pay for health and education services and to purchase food to reduce food insecurity. However, financial capital is the least available asset to the poor, making all other forms of capital even more important (DFID 1999).
The assets mentioned above are all interconnected (as shown in Figure 2-2). People combine all forms of capital in pursuit of their livelihood strategy. Financial capital provides the means to purchase food, education and health care services, which in turn, improve the human capital of a household. Financial capital also enables a farmer to purchase needed inputs contributing to the productivity of the farming system. Ownership of livestock may generate social capital (prestige) while providing physical capital (animal traction) and natural capital (manure) (DFID 1999). Each of these small farming systems is a home first and a business second making the human capital of the fanning family very important. Understanding aspects of each capital is a point that should be considered when researching ways to alleviate poverty and food insecurity.

Food Security and Nutrition
Food security, as defined by Foster (1992), is "access by all people at all times to enough food for an active, healthy life." Thomson and Metz (1997) define it as the "physical and economic access to adequate food for all household members without undue risk of losing such access." If households suffer from food insecurity "it is. because they lack entitlement to food" because they cannot produce or buy enough food to feed themselves, "what economists would call effective demand." They have no way of expressing their full need for food in the marketplace (Thomson and Metz 1997). People that are food insecure are more vulnerable to shocks and stresses, less productive, environmentally more destructive, and less caring for the welfare of fellow human beings which in turn makes the achievement of food security more difficult (Seshamani 1998). Not only is human capital affected by food insecurity but physical and social capital are disrupted as well. The focus on increasing agricultural productivity of farmers is but one aspect of finding ways to alleviate hunger. Access to food is difficult when a household cannot afford to buy the food that it cannot produce. Poor infr-astructure, i.e. roads and transportation to get to market, inadequate market structure, lack of education and technological training, and weak health and nutrition all restrict the household from obtaining the sufficient amount of food. A farm that is producing enough to sustain the members of the household and has a surplus to sell is better prepared to deal with potential shocks and stresses that may occur due to infr-astructure limitations, ineffective economic policies, and poor climatic conditions.
Changes in food security can be identified over time by rising prices and
imbalances between food demand and food supply. Rising prices affect the poor first

because approximately 60 percent or more of their income is spent on food and, at times, that may not be enough to fulfill the household nutritional requirements (PinstrupAndersen, et al. 1999). Household nutritional requirements will depend on the number of people in a household, their age, and sex. Children have higher nutritional requirements than adults and lactating and pregnant women have greater nutritional needs than women who are not pregnant or lactating. Better health reduces the need for food, thus reducing the household food requirement (Foster 1992).
Female-headed households (FHHs) have to hire labor (ganyu) in November and December for land preparation because of the extra labor required to remove trees and stumps. November and December are peak labor periods in Eastern Zambia and some FHHs may find it hard to hire help to have sufficient labor to prepare the soil for planting. Consequently, these households may withdraw their children from school to help so that crops will not be planted late and yields will not be negatively affected.
"The term gender refers to the social, economic, and cultural roles and relations between men and women, which includes their different responsibilities in a given" culture or location (IFAD 2001). Sex is biologically determined, while gender roles of men and women are determined by social norms that can change over time (IFAD 2001; Quisumbing 1996). In Africa, women provide about 70% of the labor input for food production (Ezumah and Ezumah 1996; Pinstrup-Andersen et al. 1999; Lele 1991). Although most statistics indicate that women are less productive in agriculture than men, this is likely due to the greater constraints woman face (Gladwin et al. 2001; Gladwin 1991; Deininger and Olinto 2000; Lele 1991; Due 1991; Quisumbing 1996). Women, for

example, have less access to credit, and in male-headed households women must work in their husband's fields in addition to their own. Female-headed households plant less land because of less labor in the household and limited training and education. Studies conducted in the Northern, Southern, and Cental Provinces of Zambia by Due et al. in 1982 and by Sikapande in the Southern Province of Zambia in 1988 (in Due 1991) show that FHHs plant an average of 3.0 acres in crops versus 6.8 acres for joint-headed households. Labor constraints are one of the main causes of "weak agricultural growth" in Afica (Ezumah and Ezumah 1996).
Households headed by women are more likely to be poor than households headed by men (International Labour Organization; Alwang et al. 1996). In Zambia, women head 47% of the households and because of their lower productivity in agriculture, they consume the majority, if not all, of what is produced in the fields (Due 1991). What this means is that FHHs have very little to sell in the market and have to find other ways of generating cash for their household needs. Ganyu, working in other farmers' fields,is one of the ways that Zambian women earn money to buy food if their own food supplies are low. This detracts from the time they can spend in their own fields, further limiting their agricultural productivity.
Women in male-headed households (MHH) are faced with similar problems since men in MHHs control their labor. In Eastern Zambia, for example, women are typically responsible for hoeing, planting, weeding, and harvesting of subsistence crops, food preparation, fuel wood gathering, water transport, and childcare. In MHHs, women not only work in their own fields but also have to contribute labor in their husband's fields as well as in other household activities that are required. All of their livelihood activities

detract from any personal time the woman may need. Men in Kefa Village in Eastern Zambia believe that women are "born with necks for carrying" water and harvested crops (Skjonsberg 1995) and that they are "bags to carry their babies" (own observation). Women in MHHs are also limited to their own means of generating income to provide for the household, as the man frequently does not share the income he makes from his cash crops. This further inhibits women's ability to intensify production on their own farms (Lele 1991).
Soil Fertility and Improved Fallows
"Soil fertility is an important form of renewable natural capital" (Sanchez et al.
1997). The capacity of the soils to be productive depends on several factors (Gruhn et al. 2000; Brady and Weil 1999); the crop roots need air, water, nutrients, and adequate space to develop. "Soil attributes, such as the capacity to store water, acidity, depth, and density determine how well roots develop. Changes in any of these attributes affect the health of the plant" (Gruhn et al. 2000). Agriculture is essentially a nutrient mining operation. The nutrients that are removed from the soil by cropping are going to feed a household. Since plant nutrients are the least resilient components of sustainability, the nutrient cycle rate is essential for the maintenance of a healthy agricultural system (Sanchez et al. 1997). The nutrient cycle is the continuous recycling of nutrients into and out of the soil (Gruhn et al. 2000; Brady and Weil 1999; Sanchez et al. 1997): nutrients become available to plants through the weathering of rocks, the parent materials of soil, and decaying organic matter. Extensive and continuous cropping removes nutrients faster than the nutrient cycle can replace them and nutrient-depleted soils need inputs from outside the field (Sanchez et al. 1997; Brady and Weil 1999). Traditionally, after the soil quality begins to degrade and

yields start to decrease, farmers leave previously cultivated land unused for a period of twenty to thirty years, which allows natural vegetation to grow and nutrients to cycle. The bushes and trees that grow in the fallow take up nutrients from deeper soil horizons and return them to the surface as leaf litter and through root decomposition. In addition to replenishing the nutrients in the soil, the leaf litter improves soil structure and nutrient and water-holding capacity (Brady and Weil 1999). After the fallow period, the vegetation is cleared and burned and the newly regenerated nutrients become available to subsequent crops. Cropping continues until yields begin to decline to an unacceptable level, due to nutrients no longer being available in the root zone due to crop uptake, leaching, or immobilization (Norman et al. 1995).
Studies of nutrient input/output on farmlands in sub-Saharan Africa have revealed that the average nitrogen, phosphorous, and potassium balances are negative which means that nutrients are being removed from the soil at a greater rate then they are being replaced. This suggests that the overall quality of soil in the region is declining (Brady and Weil 1999). In the soils of Eastern Zambia, nitrogen is the most limiting nutrient (Kwesiga and Beniest 1998; Sanchez et al. 1997) and it is required for the growth and health of crops. Fertilizer would seem to be the best alternative to long-term fallow periods to counterbalance nutrient depletion in the soil. In Zambia, however, the high prices of imported fertilizer, the transport costs to deliver the fertilizer, and the removal of fertilizer subsidies in recent years has made fertilizer an unaffordable option for rural small farmers. Organic fertilizers, such as animal manures, crop residues, and compost, improve soil structure and supply the necessary nutrients for crop production. Yet, most farmers, particularly women, do not have the strength or means to produce and transport the

required 10-20 tons per hectare to fertilize their fields (Kwesiga and Beniest 1998; Gladwin et al. 1997).
Scientists at the International Center for Research in Agroforestry (ICRAF) have been working to find ways of enhancing the traditional fallow practices by planting fastgrowing, nitrogen-fixing trees and legumes in systems called improved fallows technologies (IFTs) into small-scale farming systems. Improved fallow tree species with deep roots have the ability to retrieve nutrients that have leached down to the subsoil below the rooting depth of annual crops (Kwesiga and Beniest 1998; Buresh and Tian 1998; Brady and Weil 1999; Sanchez et al. 1997; Raussen 1997). Nutrients that IFTs take up from the subsoil are transferred to the topsoil and made available to following crops in the form of leaf litter, decomposing roots and branches (Buresh and Tian 1998; Kwesiga and Beniest 1998; Brady and Weil 1999). The decomposing biomass increases soil organic matter and improves soil physical characteristics, including water infiltration and storage. The improved soil structure reduces the loss of nutrients by erosion and leaching, reduces soil acidity and encourages soil biological activity (Buresh and Tian 1998; Brady and Weil 1999; Sanchez et al. 1997). Because IFTs are fast growing, they are able to cycle nutrients and improve the soil in a shorter amount of time than traditional fallows. A period of two years is usually ample time for the IFTs to improve the quality of the soil and provide adequate amounts of nutrients to subsequent crops.
One important consideration in the sustainability of the farming system is that IFTs only fix nitrogen and improve the soil physical characteristics. Some studies reported a decrease in extractable inorganic phosphorous under improved fallow species, probably due to sequestration of phosphorous in tree biomass (Haggar 1991 in Buresh and Tian

1998). However, while phosphorous deficiency is not currently a problem in Eastern Zambia, it may be in the future. To prevent a problem, other technologies, such as manure or fertilizer, can be used in conjunction with improved fallows to provide phosphorous. Of the several species of nitrogen-fixing trees currently researched, Sesbania sesban, Tephrosia vogelii, Gliricidia sepium, and Cajanus cajun are the focus of this study. A description of each species is included in Appendix B.

Chapter Overview
This chapter presents a description of the area studied in the Eastern Province of
Zambia. It provides information about the climate, soils, agricultural practices, land tenure, diet, household livelihood strategies, and other socio-economic characteristics, in order to familiarize the reader with the area.
The studies that were aimed at examining how improved fallows may or may not benefit farmers. Farmers who tested the improved fallow technologies (IFTs) but did not expand the area, farmers who tested and expanded IFTs, or those who never tried IFTs were interviewed in order to compare livelihoods. From May 16 to July 18, farmers were surveyed, with the help of an interpreter, Martha Mwale, about their farming techniques, activities the household does to generate cash, and their opinions about using improved fallows in place of fertilizer (see questionnaire in Appendix C). World Vision predetermined the village sites where the household surveys were conducted. In Chipata North, 15 farmers in Kepha, Chindola, Mayunda, Nkhala, Nyanthani, and Penyesani villages were interviewed. Fourteen farmers in Lufu, Mapato, and Kapombua villages in the Chipata South District and 7 farmers in Chundamira, Mazombwe, and Sekani villages in Katete District were interviewed. Figure 3-1 provides a map of Zambia.

Kawamb4 Js Kom isk
** *
.MIuan as
uilst*io en Deo sgTHE GaphA
The astrn rovncehaseigt aminstrtiv d ititts:Cadi, Chma >, LuansyATN PA RK
two Chipat ALAW1
Figure 1 a b
Chipata, Katete, Lundazi* a, Mfuwe, and Petauke. A Dasistrict Agricultural
Coordinator organizes agricultural activities in each district. Chipata District, where most of this study took place and covers 11,986 kmn2, includes the provincial capital, Chipata town, and is located is in the central-eastern part of the Eastern Province (see
MonguFigure 3-2). Chipata District is the area where farmers market their produce and is
divided into two agricultural districts: Chipata North and Chipata South (Petersen Mazabuka*
NAMIBIA Livng no
0 12S 25V n
0 ,Enxctoqpa-do Ontanica, tr 0 1 250 ktrn
Figure 3-1 Map of Zambia Administration and Demographics The Eastern Province has eight administrative districts: Chadiza, Chama, Chipata, Katete, Lundazi, Nyimba, Mfuwe, and Petauke. A District Agricultural Coordinator organizes agricultural activities in each district. Chipata District, where most of this study took place and covers 11,986 knx2, includes the provincial capital, Chipata town, and is located is in the central-eastern part of the Eastern Province (see Figure 3-2). Chipata District is the area where farmers market their produce and is divided into two agricultural districts: Chinata North and Chinata South (Petersen

~Cbbtnhe Fij itt
Legend O Town
,'\/Administrative Boundary C;jdz
0 ater Bodies
Fresh Water Marsh '
V- Game Reserve
rvNational Park41UNDAZ 1
Forest Reserve~~''Ag
Dryland Crop and Pasture a L"in
hi nzew
AP KE umbiE M
nemwrica.comnkn aps Scale 1:2,50,UU0
Figure 3-2: Map of Eastern Province, Zambia

1999a). The main research facility in the province is the Msekera Agricultural Research Station where groundnut and grain legume studies take place. Katete District, the other area of focus, is south of Chipata District and covers an area of 3,989 km2. It also has a market where farmers are able to sell their produce. Table 3-1, taken from Petersen 1999, provides characteristics of the agricultural districts visited.
Table 3-1: Characteristics of Eastern Province and Agricultural Districts Studied* Annual
Location Area Population Population Growth Number of No. of No. of
(km2) (1999) Density Rate Households Camps Camp 1980-90 (1996) Officers
Rural Urban
Eastern Province 69,106 1,225,000 14.0 4.0 177,100 75,900 129
Chipata N 11,986 373,000 24.3 3.6 10,830 8,170 25 16
Chipata S In CN* In CN In CN In CN 18,810 14,190 37 18
Katete 3,989 138,469 34.7 3.9 20,009 10,862 30 22
-Petersen 1999a
* Disaggregated statistic for Chipata North and Chipata South was not made available until 1997, as a result, some data for each district are aggregated in data for Chipata North.
Ethnicity and Marital Status
The main ethnic groups in the Eastern Province are the Chewa, who are matrilineal, and the Ngoni, who are patrilineal. However, there are other ethnic groups, including: Tumbuka, Nsenga, Lala, Yao, and Kunda (Petersen 1999a). Of the farmers interviewed, twenty were Chewa and fourteen were Ngoni. The other two farmers were migrants from Zimbabwe. Table 3-2, below, presents characteristics of households in each district. The table distinguishes whether the household is young, established, or mature because it is an indication of the household's ability to mobilize labor, and an important determinant of output, i.e. larger households tend to be more productive than smaller households

(Deininger and Olinto 2000). According to Petersen (1999b) young households are newly formed families with children less than five years of age. Mature households include children or adults over the age of eighteen and established households include "everyone in
between" (Petersen 1999b).
Table 3-2: Marital Status and Household Type Marital Status Household Type
Avg. Single Married Divorced/ Widow Polygamous Young Established Mature District HH (% of (% of Separated (% of (% of Total) (% of (% of Total) (% of
Size Total) Total) (% of Total) Total) Total) Total)
1 8 1 1 4 1 11 3
Chipata N 7.9 (6.7%) (53.3%) (6.7%) (6.7%) (26.7%) (6.7%) (73.3%) (20.0%) 0 10 3 0 1 2 7 5
Chipata S 6.9 (0.0%) (71.4%) (21.4%) (0.0%) (7.1%) (14.3%) (50.0%) (35.7%)
0 5 1 1 0 0 7 0
Katete 8 (0.0%) (71.4%) (14.3%) (14.3%) (0.0%) (0.0%) (100.0%) (0.0%)
Total All 1 23 5 2 5 3 25 8
Districts 7.5 (2.8%) (63.9%) (13.9%) (5.6%) (13.9%) (8.3%) (69.4%) (22.2%)
According to Celis et al. in Petersen (1999a), the populace of Eastern Province of
Zambia has one of the lowest educational levels in Zambia. In 1998 16% of males and 63% of females in the Province had no schooling. In 1998, figures for the entire country indicate that 16% males and 31% females were illiterate (World Bank 2000). Five years of education was the average for farmers interviewed in this study. Education figures and
years of residence in each district village are presented in Table 3-3, below.
Peterson (1 999b) sites some reasons for the lack of school enrollment in the
Province. They include:
* Distance to school may require some students to travel a great distance or board
* Poor funding and inadequate facilities discourage students and teachers alike
* Children may be needed to help around the home or farm
* Early marriages or pregnancies

& Parental preferences or misconceptions of the benefits of education, especially
for girls
o Lack ofjob opportunities for graduates
* Poverty at the household level or the inability to pay school fees.
Education levels are an important consideration when determining what medium of
training can be used when introducing new technologies and providing training to farmers.
Written leaflets, charts, or complex analyses would not be an effective way of training
illiterate farmers.
Table 3-3: Number Interviewed, Years of Education and Years in Village
No. Interviewed Avg. Years of Education Avg. Yrs in Village Total All Total Total All
ChipataN 8 3 4 15 5.6 6.3 2.8 5 21.7 25.7 21.5 22.5
ChipataS 4 7 3 14 7.6 5.0 1.3 5.5 23.8 19.7 33.7 25.3
Katete 4 1 2 7 4.5 5 0 3.8 28.3 41 8 27
Total All Districts 19 8 9 36 6.1 5.5 1.9 5.0 23.9 25.3 24.4 24.3
MHH: Male-headed household; FHH: Female-headed household
One of the major determining factors of small-farm productivity and incomeearning ability is the health condition of the rural population. Malaria and HIV/AIDS are the leading causes of death in Zambia: among people aged 15 to 49, 15-19% are infected with AIDS in the Eastern Province (WHO 1999; FEWS Web Version 2000). Such a high rate among the most productive age group definitely has a negative impact on farm productivity. It also endangers the food security of dependents, children and the elderly alike. In addition, malnutrition and mortality among children is a big problem in Zambia.
Twenty-four percent of children under the age of 5 were malnourished in 1997 and the
child mortality rate was 98.4 per 1000 live births (World Bank 2000; Seshamani 1998).

Maize (Zea mays) is consumed daily in a variety of forms: nsima, popcorn, boiled ear, sweet porridge, or beer. Nsima is the most common way to eat maize. Maize is pounded into coarse flour then boiled to a chick "porridge", and is commonly eaten with a relish. Relish is a sauce or stew cooked with a green vegetable, such as rape (Brassica rapa), pumpkin leaves (Cucurbita spp.), or mpilu (Brassicajuncea), or a meat, chicken, kapenta (dried fish), or pork, when it is available. Groundnuts (Arachis hypogaeaare), another staple in the Zambian diet, are pounded into a paste of peanut butter consistency, mixed with vegetables, and eaten raw or roasted as a snack. Other vegetables that are consumed include tomato (Lycopersicon esculentum), onion (Allium cepa), squash (Cucurbita spp.), pumpkin (Cucurbita spp.), beans (Phaseolus vulgaris), okra (Abelmoschus esculentus) and sweet potato leaves (Ipomaea batatas). Bananas (Musa Xparadisia) and mango (Mangifera indica) are the fruits that are eaten the most.
Topography and Climate
The Eastern Province of Zambia is located south of the equator between 10 to 150 south latitude and 30 to 330 longitude. It is bordered by Malawi to the east and Mozambique to the south. Moderate slopes and rock outcroppings interspersed with hills and ridges characterize the terrain. There are two distinct agro-ecological regions in the area: the high Eastern Plateau, 800 to 1500 m above sea level, and the Luangwa River Valley, at 300 to 600 m above sea level. Most agriculture in the Province takes place on the plateau where large tracts of land are covered with dry deciduous forests, known as miombo. Small leguminous trees of the Brachystegia and Julbernardia genera dominate these forests and have considerable grassy undergrowth (Encyclopedia Britannica). In

the hotter, drier areas of the valley, mopane woodlands dominate where Colophospermum mopane and baobab trees, Adansonia digitata, are found.
The climate is sub-tropical to tropical with a unimodal rainfall distribution. Changes in temperature and precipitation mark the shifts between the country's three seasons: the warm, wet season from November to April, the cool, dry season from April to July, and the hot, dry season from August to October. Temperatures during the hot season range from 18 to 310 C and vary from 9 to 230 C in the cool dry season. During the rainy season rainfall exceeds evapotransporation and averages 800 to 1000 mm per year with about 85% of this amount falling from December to March. The length of the growing season is typically about 139 to 155 days depending upon the amount of rain.
In Chipata South soil types are predominately loamy sands to "sandy Alfisols" (Petersen 1999b; FAO 1999, Franzel 1999). Loamy sands and Alfisols are coarse to medium textured soils that have a low to medium fertility and, while suitable for rain-fed agriculture, tend to have poor physical characteristics that deteriorate rapidly upon use. Clays and "loam Luvisols" (Petersen 1999b) occur in Chipata North and Katete. Luvisols tend to have high pH because of their high calcium content and have poor drainage. There are also swampy areas, known as dambos, where fertile gray-brown, loamy sands, non-expanding 2:1 illite clays (Petersen 1999b), and hydromorphic gleysols are found. These soils are relatively easy to hoe and remain moist during the dry season, which allows for a dry season crop, providing a valuable asset to farmers (Raussen 1997).

Farming and Livelihood Systems
Livelihood systems in Eastern Zambia are generally based upon agriculture.
Figure 3-3 facilitates an understanding of the resource flows from crops, livestock, and
the forest to the household. It also illustrates the linkages between the market and
physical environment to the household and farming system.
MARKETS & POLICIES Inputs, Food, Off-farm Work, Income, Technology, Credit, Information. Distribution. Infrastructure
HOUSEHOLD Preferences
Land, Labor, Capital, Social Patterns
H- Composition Customs
CROPS Food, Cash Labor, Food, Incomne,
Maize, Sunflower Leather, Ceremonies
Sweet potatoes
Groundnuts LIVESTOCK
Soybeans Fodder Manure, Labor Cattle, Pigs, Goats,
Cowpeas Sheep, Chickens
Pumpkins 4 -Food,
Paprika, Squash
Fruits, Mushrooms
Thatch Fodder
Fiber, Timber, Poles,
Hunting. Medicine
PHYSICAL ENVIRONMENT Soil Fertilitv. Weather. Pests. Diseases
Figure 3-3: Village Resource Flows, Eastern Province Zambia

Land Tenure
There are three categories of land in Zambia: state land, reserve land, and trust land (Amankwah and Mvunga 1986; Petersen 1999b). Before independence, state land was controlled by the British Colonial government and was set aside for European settlement and economic development. Presently, the Zambian government controls state land and arrange leases and titles for its use. It is usually the most fertile land and mainly used by townships or for resettlement schemes (Petersen 1999b). Most trust land and reserve, or traditional, land is either inherited from parents or relatives or given to the person by the village headman (Petersen 1999b). These two categories of land are generally not registered and are, therefore, neither bought nor sold (Amankwah and Mvunga 1986; Petersen 1999b). Rights are generally given to the household that first cleared the land, however, the Chief or headman has the power to allocate land to an outsider and to settle land disputes that may arise. Women usually receive land from a male relative or their husband. Table 3-4, below, shows average area cultivated or held in fallow for each district, desegregated into male- and female- headed households (MHH and FHH respectively). The area of land that has been uncultivated is land that has never been cultivated and is held in reserve for family or emergency.
Table 3-4: Total All HHs Land Area Cultivated Avg Area Avg Dimba Area Avg. Fallow Area Avg Area
District Cultivated (ha) (ha) (#Farmers) (ha) Uncultivated
(#Farmers) (#Farmers) Land (ha)
ChipataN 3.69(11) 1.39(4) 0.42(3) 0.00 2.36 (7) 1.01 (2) 4.35 (7) 0.88 (3)
Chipata S 1.90(11) 1.29(3) 0.27(9) 0.25 (1) 0.85 (5) 0.81 (1) 2.22 (4) 0.81 (1)
Katete 2.92 (5) 1.69 (2) 0.76 (4) 0.00 0.55 (3) 0.81 (1) 0.70 (2),0.40 (1)
Total All FIHs 2.82 (27) 1.42 (9) 0.42 (16) 0.25 (1) 1.50 (15) 0.91 (4) 3.14 (13) 0.77 (5)

Crop Production Systems
Total arable land in the Eastern Province is about 3.8 million hectares with
approximately 35% of this being utilized for agriculture (Peterson 1999b; Jha and Hojjati 1993). Approximately 75% of the population is involved in agriculture (Petersen 1999a). Maize (Zea mays) is the principal staple food crop and accounts for 80% of the area cultivated (Franzel 1999). Other important subsistence crops include groundnuts (Arachis hypogaeaare) and cassava (Manihot esculenta). Squash (Cucurbita spp.), pumpkin (Cucurbita spp.), beans (Phaseolus vulgaris), and cowpeas (Vigna unguiculata) are typically intercropped with maize or groundnuts. In addition, cotton (Gossypium spp.), soybeans (Glycine mar), sunflower (Helianthus annuus), and tobacco (Nicotiana tobacum) are grown as cash crops. Table 3-5 provides a summary of the average area cultivated by the farmers interviewed for the most common crops planted.
Dimbas are small gardens (<1 ha) located in wet swampy areas, locally known as dambos, and along riversides or springs and are traditionally cultivated by men. However, both male and female-headed households (MHH & FHH) will cultivate dimba plots when available. Men generally grow vegetables for sale in the market while women grow vegetables that are eaten by the family. Rape (Brassica rapa), mpilu (Brassica juncea), tomatoes (Lycopersicon esculentum), sugar cane (Saccharum officinarum), onion (Al lium cepa), okra (Al lium cepa), cabbage (Brassica), bananas (Musa X paradisia), potato (Solanum tuberosum), sweet potato (Ipomaea batatas), green maize (Zea mays), and beans (Phaseolus vulgaris) are the primary crops grown in dimbas. For information regarding area of specific crops planted by household type, see tables in Appendix D.

Table 3-5. Common Crops by District
I Average Area Average Yields # Farmers
District (ha.) (kg/ha) % of Total Area Using
________ (# Farmers) (#Farmners)* Sown Fertilizer
__________1998-99 1999-00 19-9 1999-00 1998-99 11999-00 ____Chipata North ______ __Cotton 0.83 0.88 967 494 14.4 13.3 3
____________ (8) (7) (7) (4)* _____ _____Groundnuts 0.58 0.58 821 407 18.5 18.5
____________ (15) (15) (13)l (9) _____ _____Maize, Hybrid 2.22 2.22 1619 3018 14.4 14.4 3
___________(3) (3) (3) (2) _____ ____ _____Maize, Local 1.19 1.20 1947 1545 38.7 39.1 9
____________ (15) (15) (15), (11) 1_____ ____ _____Chip ata South ________ ___ ________Cotton 0.51 0.61 736 1724 8.2 4.9
____________ (4) (2) (4) (1)* ____________Groundnuts 0.37 0.39 694 315 19.6 17.6
__________ (13) (11) (11) (12)_________Maize, Hybrid 1.08 1.25 2171 2425 17.6 15.1 4
___________(4) (3) (2) (2) ___ ____ ______Maize, Local 0.87 1.01 2050 2096 45.7 53.1
___________ (13) (13) (12) (13) ______________Katete_____ ______ _____ __________ _Cotton 0.75 0.64 1555 494 25.2 10.7
____________(6) (3) (6) (1) ____________Groundnuts 0.79 0.80 755 225 26.3 26.6
___________ (6) (6) (5) (5) _____ _____Maize, Hybrid 0.53 0.53 1668 3852 5.9 5.9 2
____ ___ ___(2) (2) (1) (2) _ _ _ _
Maize, Local 0.83 1.09 1742 1316 32.6 42.5 1
___ __ ___ __(7) (7) (7) (7) _ _ ____ _____Total All Households _________ _____ ___Cotton 0.73 0.77 1117 728 14.8 10.4 3
____________ (18) (12) (17) (5) _____ _____Groundnuts 0.53 0.55 761 334 20.2 20.4
___________ (34) (33) (29) (26) _________Maize, Hybrid 1.34 1.43 1811 3098 13.6 12.9 9
___________(9) (8) (6) (6) ___ ____ _____Maize, Local 1.00 1.11 1949 1725 39.4 43.7 1
__________ (35) (35) (34) (31) ____ _______*Yields for the 1999-00 season are incomplete since some farmers were still harvesting at the time of data collection.

The planting season begins in September just before the rains when households begin to clear and bum weeds and then loosen the earth by hoeing. Most farmers are hand cultivators, although, some farmers do have access to oxen either through ownership' or rental. Women do most of the hoeing while men are involved primarily with clearing the land of debris and incorporating crop residues into the soil (see Table 36 for a complete breakdown of activities by gender). Local maize and groundnuts are the first crops planted with the start of the rains. Hand-hoe cultivators start planting before the rains in October and farmers using oxen begin late October or early November. Minor crops, such as beans, pumpkins, and squash are intercropped with maize or groundnuts. Cotton and hybrid maize are planted in December. Village residents, principally women, devote part of November, and all of December and January to weeding. Fertilizer, when used, is applied by men in December and January. Dimbas are generally planted in April or May.
Labor, which comes directly from the household, is one of the most limiting factors to increased crop production for it determines the size of the field planted. Farmers that own oxen are capable of cultivating a larger field than hand cultivators in the same amount of time. For activities when extra labor is required, such as weeding and harvesting, labor is hired or bartered for maize when household maize stocks are depleted. Clubs are another valuable source of labor: In Lufu village there is a very well organized woman's club where members take turns weeding each other's fields.
I Approximately 30% of the farmers in the Eastern Province own oxen (Petersen 1999a).

Table 3-6: Fanning Activities Calendar
Build Grain Storage__ _ _ _ __ __ _ __
Build Houses
Bum Grass
Collect Firewood
Cultivate Dimba IIIIIIIII // /-/
Cut Thatch //// // "IIIII
Dry Leaves, Maize z /,, /,j/ //j. /,///
Dry Tobacco
Fertilize/Manure MU IIII11111_IIIII_..
Garden Nursery : ii : i /
Harvest "///IM~ IIIIIIIl
Help in Fields _Hunting ___-7_1
Land Clearing IIIIIIIII
Land Prep ___._ __ __Make pots, tools,
and baskets. _______Market ____ 1 __Planting I II I _/_////,Ridging ,__Shell Maize/GNs
Spray Crops _______Stalk pile Maize ______II
Store Grains //'Ji
Water Garden //J/
Watch Animals -771
Farmers in the Eastern Province typically keep chickens, goats, pigs, and/or cattle.
Occasionally dogs, ducks, sheep, pigeons, and guinea fowl are also kept. Table 3-7
shows the average number of animals kept by household type. In addition to providing
milk and labor, cattle are a source of wealth and status. The number of cattle an average
farm owns ranges from 1.5 to 3 (Petersen 1999b). Typically, young boys are responsible
for cattle: during the rainy season, the main cropping period, boys are sometimes kept out

of school to herd cattle away from the fields and confine them to the village where they are fed food scraps. Cattle then forage freely during the dry season frequently grazing along stream edges, in woodlands, and in dambos. This is of concern to farmers who are interested in IFTs since animal grazing could destroy the improved fallow plants.
Young boys also care for goats and sheep. They are slaughtered on special
occasions, sold for cash or bartered for household needs. Goats and sheep commonly graze in the same areas as cattle during the dry season and are confined to the village during the rainy, cropping season and fed cornhusks and food scraps. All animals are restricted to the village at night. Dogs are kept by men and used for hunting, while pigeons and guinea fowl are often just a status of wealth.
Most households have chickens that are kept by women and fed kitchen scraps. Chickens provide eggs and are slaughtered for visitors. Pigs are also kept by women and fed kitchen scraps, maize cobs and husks. Pigs and chickens are more often sold for household needs than goats or cattle. Women, especially elder women, trade animals for labor and fertilizer.
Table 3-7: Average Number of Animals per District
District Chickens Cows/Oxen Goats Pigs Other
ChipataN 18.6 5.0 5.6 0 10.3 4.0 3.8 0 8.5* 3.0'*
Chipata S 8.1 8.0 5.8 0 2.0 4.0 7.3 3.0 0
Katete 9.8 5.0 4.0 0 0 0 7.7 4.0
TOTAL ALL HHs 12.3 5.9 5.5 0 6.1 4.0 5.8 3.7 8.5* 3.0*
-Guinea fowl and sheep; *Ducks
Forest Use
Forest cover around the villages is limited. Products from the forests are
primarily used for firewood, construction materials, fruits, and medicine. Women are

usually responsible for collecting wood, although, in areas where distance from the village has made collection and transport difficult, men have taken over the responsibility. Forests are also used for hunting and for keeping beehives for honey. Fallow Practices
Low soil fertility is a major constraint to agricultural production in Eastern Zambia. Before the introduction of fertilizers, farmers in the Eastern Province would cultivate their land continuously for several years until there was a marked decline in yields. Once the soil had been depleted the land would be left fallow under its natural vegetation for a period of about 20 years. In this manner, farmers were able to restore declining fertility through the decomposition of leaf litter and roots. While farmers appear to have sufficient access to land to practice traditional slash and burn (Petersen
1 999b), current fallow practices are limited. Presently, the only land that is left fallow is land belonging to households with insufficient labor to plant it (Petersen 1 999b). Short grass fallows are sometimes used to replenish soil nutrients or decrease weeds (Petersen 1999b). Crop rotations with groundnuts, chemical fertilizers, and manures are the most common methods used to enhance soil fertility. Table 3-4, above, shows the average fallow area of the households interviewed. A breakdown of fallow area by household type can be found in Appendix D.
Improved fallow technologies (IFTs) are based on the traditional fallow practice of leaving the land uncultivated for several years to allow the nutrients to regenerate in the soil. However, IFT's involve planting fast-growing, nitrogen-fixing tree species to hasten the natural process of soil nutrient regeneration. Additional benefits of the IFTs include firewood, fodder, pesticide, weed suppression, erosion control, and shade.

Most farmers learned about the benefits of improved fallows from their
agricultural camp officer or lead farmer. There are two farmers in each village, one male and one female, who are elected by the village through World Vision, to attend all meetings given by World Vision about improved fallows and bettering their farming practices. A World Vision representative and agricultural camp officer lead each village meeting. Farmers who attended the meetings seemed enthusiastic about the improved fallows and the potential to increase their maize yields. The meetings tend to be very effective in generating farmer interest. Of the farmers that were interviewed, 27 have tested, or tested and adopted, the improved fallow technologies. Table 3-8 shows the number of female-headed and male-headed households with an improved fallow, the type of species planted, and the area of the plot. Some farmers often grew more than one species.
One of the biggest complaints about the improved fallows is the amount of work involved. Women were especially concerned about the amount of labor required for the trees. They just do not have the time to devote to the improved fallow field with all that is required from their household and other fields. That is presumably why most women chose to plant Tephrosia vogelii. Despite the fact that maize yields in a Tephrosia sp. field are less than in a Sesbania sp. field, Tephrosia sp. can be planted directly without nursery establishment and when cut will regenerate. Other benefits of Tephrosia sp. include drought tolerance and its uses as a pesticide for garden and field crops. A species background is included in Appendix B.
Farmers, for the most part, are very interested in finding alternatives to improving their soil fertility. Most believe that fertilizer is bad because the soil will become

dependent on the fertilizer and will require more each year. They like the idea of the improved fallows and have seen the benefit of increased maize yields in neighboring villages. Their biggest question was, "Where can we get more seeds?" The farmers that did not try the IF trees did not feel that their soils were "tired" or they could afford to buy fertilizer and were not interested in other soil fertility options.
Table 3-8: Improved Fallows
No. HHs
District With IF Avg Area of IF (ha) (# Farmers)
Cajanus cajun Gliricidia sepium Sesbania sesban vogelii MHH FHH MHH FHH MHH FHH MHH FHH
12 0.06 0.16 0.13 0.11
Chipata N (80.0) (1) (4) (5) (3)
11 0.18 0.14 0.40 0.28
Chipata S (78.6) (5) (2) (6) (3)
4 0.17 0.11 0.29 0.24
Katete (57.1) (3) (3) (4) (4)
27 0.14 0.15 0.21 0.27 0.19
Total All HIHs (75.0) (4) (8) (10) (15) (6)
Other Income Generating Activities
The livelihood systems of farmers in the Eastern Province include not only agriculture but also a mix of other cash-generating activities. Table 3-9, below, lists several of the activities that farmers do to generate income. Brewing beer or performing ganyu are women's activity, while selling cash sell crops or vegetables is primarily a male activity. Not only are these activities used as a source of cash, but goods are also sometimes bartered for maize, fertilizer, or labor. Women, in particular, rely on small farm animals and other goods to acquire labor for land clearing and house repairs. The biggest income generating activities appear to be selling dimba fruits and vegetables,

beer, and working in other farmers' fields. Selling beer is a female activity and is one of
the primary sources of income to pay for children's school fees. Other activities include
selling honey, milk, charcoal, paraffin, sugar, and cooking oil, brick laying, tailoring, and
traditional healing.
Table 3-9: Income Generating Activities SActivity Women Men
Baking and selling baked goods X
Bicycle repair X
Brew and sell beer X
Buy supplies and sell in village X X
Credit X X
Harvest honey X
Home construction X X
Make and sell bricks X X
Make and sell mats and baskets X
Make and sell pots X
Make and sell farm tools X
Off-farm employment X X
Outside remittances X
Pension X X
Performing ganyu (piece work) X X
Press and sell sunflower oil X X
Rent oxen for plowing and transport X
Sell animals X X
Sell crops and garden vegetables X X
Sell firewood & charcoal X X
Sell grass for thatching X
Sell kapenta (dried fish) X X
Sell mealie meal X
Sell milk X
Sewing and knitting X X
Traditional healer X X

This concludes a summary of the farming systems of Eastern Zambia. It is important to be familiar with the different activities and practices that small farmers employ in order to understand why certain activities are practiced and how they alter when bio-physical or socio-economic conditions change.

Research for this Soil CRSP project was conducted in Eastern Zambia from May to August of 2000. Data were gathered from local farmers in order to understand the farming systems of the area and to be able to create a representative mathematical linear programming (LP) model of the farming system. With this model it is possible to see how improved fallows and other activities affect the sustainability of the household over a five-year period. The question asked of the model is what combination of activities is best for the sustainability of the system.
The LP model was created to simulate two different types of households: femaleheaded (FHH) and male-headed (MHH). The goal of the LP model is to meet certain consumption and cash goals and to maintain a livelihood.
This chapter presents the methods of data collection and analysis. Also estimated in the chapter are the possible benefits of improved fallow technologies on maize production to village households.
Methods of Data Collection
To gather information about the farming techniques and livelihood strategies of small-scale farmers in Eastern Zambia direct observation as well as informal and formal interviews with a structured questionnaire were conducted. Direct observation was used

in the local markets and villages in order to familiarize the researcher with the area and to ease develop relationships with the people. Regular visits to the market allowed the researcher to observe market interactions and the behavior of prices. While in the villages, the researcher conducted the interview in the fanner's field and was, therefore, able to see the improved fallow trees, crops being harvested at the time, and dimba plots that may have been planted.
Formal and informal interviews with farmers, agricultural extension officers, and non-governmental organizations were the main methods of gathering data. Interviews with farmers regarding local fanning techniques, activities done to generate cash, and the improved fallow technologies, were conducted while living with them in their village. Staying in the village enabled the people to get to know the researcher and her objectives and allowed the researcher to learn about village life from the farmers. Other sources of data included a literature review and an Internet search for data, such as maps and rainfall patterns.
A total of thirty-six farmers, male and female, testers, non-testers, and testerexpanders of the improved fallow technologies (IFT), were interviewed. Thirty-five households were analyzed using the LP model. One household was not used due to incomplete data. A composition of study contacts is presented in Table 4- 1. A more complete breakdown by household type of testers and non-testers of the IFT's is provided in Appendix E.

Table 4-1: Composition of Farmer Interviewees
Number of
Type Respondents
Female Non-testers in FHH 3
Female Non-testers in MHH 2
Male Non-testers 5
Female Testers in FHH 6
Female Testers in MHH 5
Male Testers 7
Female Tester-expander in FHH 0
Female Tester-expander in I
Male Tester-expander 7
Total: 36
Linear Programming
An LP model attempts to describe the interactions between the various components of a system with the goal of maximizing or minimizing a particular objective. The goal is to select the best available alternatives given the resource restraints. One of the first linear programs used in agriculture was to balance a diet.Since then it has been used as an analysis tool for all types of farming operations, from large commercial operations to small rural household farms in less developed countries. The system considered here is a small rural fanning household.
There are three components to a linear programming model. The objective
function is the first component, which is to be optimized (i.e. maximized or minimized). In economics this is usually stated in monetary terms the maximization of revenues or the minimization of costs. It is assumed here that the primary goal of the household is to produce enough staple food to meet its consumption requirements after which the household maximizes end year cash available for discretionary spending from crop

production and other income generating activities. The notation of the maximizing objective function is represented by: max Z= (PXj -Cj Zj), where
Pj is the price received per unit of activity X, which is crop production or other income generating activity, and C is the cash cost per unit of activity Z, which is the inputs needed per activity. The number of activities, or columns, is n.
The constraints, which are the cash and consumption requirements, on the
objective function and the available resources, land and labor, are the second and third components of the LP. They are presented below:
I a, Xi bi i = 1, 2...., m
XJ 01, 2,.. ., n, where
ay equals the quantity of resource i used per unit of activity; bi equals the set of available resources i for the X activities; and Xj > 0 is a nonnegative constraint to ensure all activities are in positive quantities and make practical sense. Linear Programming Model Development
The LP matrix is a five-year monthly model built in a Microsoft Excel
spreadsheet and run in Excel's optimization program, Solver. It has 633 activities (columns) and 7,226 rows. The columns are the activities, or livelihood strategies, in which the household participates. The rows are the labor available and the different cash and consumption requirements of each activity and the yields produced. The solution is feasible if the model is able to satisfy all of the constraints present in the household. Below is a list of the specific activities and constraints that are used in the LP:

LP Activities
Crop Production:
* Crops grown
o Fertilized local maize intercropped with beans and squash
o Unfertilized local maize intercropped with beans and squash
o Fertilized hybrid maize sole crop
o Cassava
o Cotton
o Groundnuts
o Sweet potato
* Dimba (small gardens)
o Tomato
o Onion
o Leafy greens
o Bananas
* Improved fallows
o Sesbania sesban o Gliricidia sepium o Tephrosia vogelii
o Cajanus cajun
o Local maize in improved fallow plots
o Hybrid maize in improved fallow plots
* Crops
o Local maize
o Hybrid maize
o Cotton
o Groundnuts
* Dimba
o Tomato
o Onion
o Leafy greens
o Bananas
* Improved fallow seeds
o Sesbania sesban o Gliricidia sepium o Tephrosia vogelii
o Cajanus Cajun
Other Income Generating Activities:
* Sell ganyu (labor)
* Brew and sell beer
Other activities:

" Buy maize
" Hire ganyu
* Monthly maize (hybrid and local) supply
* Monthly cash
Resources and Constraints
Resource Constraint
Household members Depends upon household composition
Ganyu (hired labor) Limited to 30 days
Crop production Amount available
Dimba If available
Consumption Depends upon household composition
Maize 225 kg/person/yr
Groundnuts 72 kg/person/yr
Cassava 108 kg/person/yr
Sweet potato 70 kg/yr
Nondiscretionary Cash Needed
Crop inputs Seed, Fertilizer, Pesticide
Household expenses
School and health fees Food
Other Income Generating Activities
Beer brewing Maize, Cash
Ganyu sell Labor days limited 5/mo.
After validation, two scenarios were posed to the model:
1. Cultivation without an improved fallow option and
2. With an improved fallow and seed selling activity

Two additional scenarios were posed to a selected sample of households from the MHH and FHH group. The first scenario involved reducing the price to sell improved fallow seed by half. The second scenario did not offer an option to sell improved fallow seed on the market. A description of each household interviewed is provided in Table 4-2.

Table 4-2: Characteristics of Modeled Households Total
# Female Male Available Dimba Cash (ZK)
People Female Labor Male Labor Land Area needed/mo HH# District Type Gender in HH Producers (days) Producers (days) (ha) (ha) ( 1000)
Male-headed Households
1 Chipata N Tester F 11 5 486 3 270 5.87 0.25 51.73
2 ChipataN Tester M 4 1 108 1 108 2.62 12.08
4 ChipataN Tester M 21 6 972 9 648 14.32 0.81 64.33
6 ChipataN Non-tester F 3 1 108 2 217 2.02 0.61 21.00
7 Chipata N Tester-expander M 10 2 216 2 162 22.94 28.00
8 Chipata N Tester M 6 3 324 1 108 2.76 65.00
9 ChipataN Non-tester M 7 1 108 3 270 2.25 54.50
12 Chipata N Tester M 9 3 270 3 324 9.43 65.00
13 Chipata S Non-tester M 5 2 162 2 216 1.62 0.10 14.05
14 Chipata S Non-tester M 5 1 108 2 216 7.41 0.20 31.08
18 Chipata S Tester M 4 1 81 2 162 2.40 0.40 64.67
19 Chipata S Tester F 5 1 108 2 162 2.77 0.06 17.17
20 Chipata S Tester F 2 1 108 1 108 3.31 0.25 41.00
21 Chipata S Tester M 9 3 324 2 217 7.94 0.50 161.20
23 Chipata S Tester F 15 5 540 6 595 2.86 0.61 42.50
24 Chipata S Non-tester M 8 2 216 4 378 2.77 0.02 54.54
25 Katete Tester-expander M 10 2 162 4 378 7.03 33.60
26 Katete Non-tester F 6 3 324 2 217 4.80 0.61 61.38
29 Katete Tester M 5 1 108 2 163 4.10 0.61 65.00
30 Katete Non-tester M 8 3 324 3 271 2.25 1.00 65.83
31 Katete Tester-expander M 8 3 324 4 433 1.21 0.81 5.83
32 Chipata N Tester-expander M 9 3 270 3 324 22.64 76.41
33 ChipataN Tester-expander F 5 1 108 2 216 3.34 47.13
34 Chipata S Tester-expander M 15 4 378 5 486 3.79 0.40 241.25
35 ChipataN Tester-expander M 6 2 216 2 216 1.72 0.40 206.00
Female-headed Households
3 Chipata N Tester F 4 2 162 1 108 1.51 22.10
S ChipataN Tester F 3 1 81 2 162 2.30 21.20
10 Chipata N Tester F 9 3 351 2 189 2.68 35.25
11 Chipata N Non-tester F 11 2 189 6 594 4.05 48.33
16 Chipata S Tester F 8 4 378 2 162 2.69 39.13
17 Chipata S Tester F 4 2 189 0 0 2.27 0.25 100.00
22 Chipata S Tester F 8 3 270 4 432 1.31 35.00
27 Katete Non-tester F 12 4 432 2 216 3.15 21.50
28 Katete Non-tester F 7 3 324 2 216 1.46 40.07

Validation of the Simulation Model
Prior to using the model it must be validated to determine how well the model
compares to against actual data collected. In this case area planted in maize and total
area planted in crops without an improved fallow was chosen for the validation process.
Figure 4-1 and 4-2 presents a scatter diagram of the LP model data and the actual
data for the total area cultivated and the area of maize planted, respectively. The diagram
shows how close the LP model simulation data is to the real world data.
Figure 4-1 Scattergram for Total Area Cultivated 10.00.
7.00 .
m 5.00
4.00 rs- -a---2.00-
.... --
0.00,, EActual Data 0 10 20 30 40
Model Data Farmer
Figure 4-2: Scattergram for Maize Area Planted
1 .00---
*Actual Data 0 10 20 30 40
< Model Data Farmer #

Before performing further validation of the LP model and to determine the
appropriate statistical tests to be applied, a Shapiro-Wilk Normality Test was formulated to see if the data are normally distributed. The null hypothesis is that the data are normally distributed.
HO: Data values have a normal distribution.
HI: Data values do not have a normal distribution.
The probability, or the p-value, associated with the test statistic is what helps to determine whether or not to reject the null hypothesis. If the p-value is less than the 95% significance level, cc = 0.05, then the null hypothesis is rejected, the data values are not normally distributed (SAS Sources 2002). The p-values for each data sample are shown in Table 4-3. In every case the p-value is less than 0.05, hence the null hypothesis is rejected. The data are not normally distributed.
Table 4-3: Shapiro-Wilk Normality Test
ShapiroWilk 's W Probabilt
Total area of all crops planted:
Actual Data 0.7618 0.0000
Model Data 0.8922 0.0027
Maize area planted:
Actual Data 0.7228 0.0000
Model Data 0.8578 0.0003
Since the basic assumption of normal distribution is not satisfied, a Chi-Square
Test was computed in StatMost for Windows to determine whether the actual data set and the LP model data set is comparable. The null hypothesis is that the two data sets are drawn from the same distribution.

HO: The distribution from each data set is the same.
H1: The distribution from each data set is different.
Based on the results, shown in Table 4-4, the null hypothesis that each data set has the same distribution is not rejected. That is, the actual data and the data from the LP model have similar distributions and are, therefore, comparable.
Table 4-4: Chi-Square Test Results Actual Data LP Data
Total Area Planted
Minimum Size 34 34
Degree of Freedom 33 33
Chi-Square 9.0479
Probability 0.9999
Maize Area Planted
Minimum Size 34 34
Degree of Freedom 33 33
Chi-Square 18.9696
Probability 0.9758
A Mann-Whitney Test was also computed in StatMost for Windows to determine if the difference between the means of the actual data and the LP model data is significant. With a = 0.05 level of significance, the null hypothesis is that the mean from each data set is the same.
H0: The mean from each data set is the same.
HI: The mean from each data set is different.
The critical values for the Mann-Whitney Test when ac = 0.05 are zo = 1.96 (DataMost Corp 1995). "The z-score is compared with a normal distribution with an infinite number of degrees of freedom (DataMost Corp 1995)." If the computed value is less than the critical value then the null hypothesis cannot be rejected.

The results, presented in Table 4-5, show that for the total crop area cultivated, z is less than 1.96. Thus if the null hypothesis is rejected then there would be a 33% chance of making a Type I error. Therefore, it is concluded that the difference between the mean from the actual data and the mean from the model data for total crop area planted is not significant, indicating that the LP model is an adequate representation of a small farm in Eastern Zambia.
For maize area planted z is greater than 1.96 indicating that the null hypothesis
should be rejected. The mean from each data set is different. This would suggest that the maize yields in the model could be too high. That farmers may have had trouble recalling the actual yields realized the previous season. Or if labor was the constraint, the labor contribution of children in the household may be underestimated in the model.
Table 4-5: Mann-Whiitney Test Analysis Results Actual Data LP Data
Total Area Planted
Sample Size 34 34
Total Sum 85.6929 66.5795
Mean 2.5204 1.9582
Standard Deviation 8 1.5291
z-score 0.9751
Two-tailed P value 0.3295
Maize Area Planted
Sample Size 34 34
Total Sum 46.8168 15.4828
Mean 1.3770 0.4554
Standard Deviation 81.5291
z-score 5.8875
Two-tailed P value 0.0000

In the first year of the model, every household is given a certain amount of cash (an amount equal to what the farmer indicated that he or she needed) the first month of the production cycle, September, and a supply of maize to sustain the members until the time of harvest. No households were feasible at 100% of what it was estimated each household needed. In order to find at what point it would be feasible given the data set, the cash requirements were first decreased by 25%. If the model was still infeasible then the consumption requirements were decreased by 25%. Table 4-7 shows the percentage decrease that cash and consumption requirements had to be reduced in order to reach a feasible solution. The fact that the requirements of the household had to be reduced means that either the data used in the model are incomplete, there are activities not included that would significantly affect what and how much the household produces, or the households are not actually meeting their consumption and cash needs each year.

Table 4-7: Percentage of Consumption and Cash Requirements Achievable in the LP Model
Male Additional
% Male % Female % Beginning Female
HH # Cash Cash % Maize % GN Cassava Cash Beginning Cash
Male-headed Households
1 25 25 25 100 100 10
2 50 50 100 100 100 10
4 75 50 50 50 50 20
6 50 50 100 100 100 10
7* 50 50 50 25 75 10 20
7** 25 25 25 25 75 10 20
8 25 25 75 75 70 10 30
9 25 25 100 50 25 40
12 25 25 50 50 100 10
13 50 50 100 100 50
14 25 100 75 100 10
18 20 20 100 25 100 10 30
19 50 50 75 75 100 10
20 25 25 75 100 50 20
21 20 5 50 50 100 10 20
23 50 25 50 100 25 10
24 50 25 100 25 25 20
25 25 25 50 25 100 20
26 25 25 100 100 100 10 20
29 25 15 75 100 100 .30
30 25 25 75 50 50 10 20
31 100 100 75 100 10 10
32 25 20 50 75 100 10 30
33 25 25 75 75 100 20 30
34 20 10 50 75 25 10 40
35 20 10 75 100 50 20 40
Female-headed Households
% HH % Beginning
HH# Cash % Maize % GN Cassava Cash
3 75 100 100 100 10
5 75 100 100 100 10
10 75 75 100 50 30
11 25 75 100 75 30
16 50 75 100 75 20
17 2 50 100 100 40
22 100 75 100 25
27 50 75 100 50
28 50 100 100 50
*With IF; **Witout IF

Hypothesis one stated that IFTs would improve farmer livelihood systems. The model did indeed indicate that planting an improved fallow plot is beneficial to the household, as indicated by the increase in year-end discretionary cash, see Tables E-5 and 6 in Appendix E. For each scenario with an improved fallow option the model picked an improved fallow planting activity. Appendix E, Tables E-3 and 4 show the final results for each scenario: planting an improved fallow at full seed price, planting an improved fallow with the price decreased by half, and planting an improved fallow without the option to sell seed. Regardless if there was an option to sell improved fallow seed, the model chose an improved fallow planting activity in every case except one.
Hypothesis two stated that female-headed households would be more likely to pick an improved fallow species that required the least amount of labor. In every case, the model chose Tephrosia vogelii or Cajanus cajan as the species to plant for femaleheaded households. Gliricidia sepium was picked only when the IFT seed selling option was omitted. The reason for this is probably due to the fact that Gliricidia sepium does not yield as much seed as the other species and it does not yield seed until the third year, whereas the other species yield seed in the second year. Sesbania sesban was most likely not chosen for female-headed households because the seedlings must be planted in a nursery before they can be planted in the field. Nurseries are usually located in the dimba area and, as previously stated, most women do not have dimba land. The seedlings also require greater care than the seedlings of the other IFT species.

There are some limitations of the LP model that should be considered. Since the LP model is a simplification of the real world, some production constraints and characteristics of each household may be absent. Field interviews may not have captured every piece of information that should be defined in the model. For instance, in the villages of Eastern Zambia, farmers rely not only on their immediate family members for help in the fields and for food, but also share their labor and resources with other members of the community. Additionally, what may typically be a male activity in most households may be a female activity in other households. Also, the production labor contributed by children may not have been fully represented. Furthermore, soil in one farm may be more or less fertile then land in another area. This is not captured in the model.
Another consideration of the model is that most elements (labor, yields, and costs) are fixed for every household. For example, the labor required to plant certain crops is the same for every household. Only the available labor is variable according to household composition. The labor used in the model is only representative of the households interviewed. As such, the sample of thirty-five households does not sufficiently represent the entire study area in Eastern Zambia. Even though the improved fallow technologies were adopted in the model, the actual effects of adoption to each individual household would involve specific calculations particular to that household. With this in mind, the LP model should be used only as a guideline for moving forward and targeting the proper technologies to different households.

CONCLUSION AND RECOMMENDATIONS Maize is the most important crop to households in Eastern Zambia. Because of poor soil fertility and high fertilizer prices, small, rural farmers have experienced low maize yields and, as a consequence, hunger in the months prior to harvest. In response to this problem, researchers at the International Center for Research in Agroforestry (ICRAF) have found that planting fast-growing, nitrogen-fixing trees on land with infertile soils have the potential to replace fertilizer as an input for maize cultivation. The aim of this study was to determine the benefits of the adoption of improved fallow technologies as a livelihood strategy to small farmers in Eastern Zambia. This chapter presents some conclusions of this research concerning using improved fallows as a livelihood strategy and provides some recommendations for further study.
Improved fallows build on the traditional practice of leaving land with infertile soils uncultivated for a period often to twenty years allowing the natural vegetation to grow and to restore lost soil nutrients. The leguminous tree species, Sesbania sesban, Gliricidia sepium, Tephrosia vogelii, and Cajanus cajun, have been introduced to small farmers in Eastern Zambia as a viable alternative to fertilizer use. Each species has its own planting and maintenance requirements and soil improving qualities. While Sesbania sp. provides the greatest nitrogen-fixing characteristics and subsequent maize yields, it also requires a nursery for seedlings and extra care. Thus, Sesbania sp. may not

be appropriate for some households. This study has sought to determine how the adoption of improved fallow technologies (IFTs) as a livelihood strategy may enhance maize cultivation of small farmers.
Linear Programming
A linear programming (LP) model was built to represent small farms in Eastern Zambia. Using data gathered in farmer interviews conducted over a three-month period, the model was designed to simulate their current situation and predict what might happen to farmer livelihoods if IFTs were adopted. Given each household characteristic, LP models were run with and without an improved fallow option to see 1) if IFTs were likely to be adopted and by what types of households and 2) how planting IFTs would affect year-end discretionary cash. No households were feasible at 100% of what it was estimated each household needed. In order to fmd at what point it would be feasible given the data set, the cash requirements were first decreased by 25%. If the model was still infeasible then the consumption requirements were decreased by 25%.
The model also included an IFT seed selling activity. The LP model was run on randomly selected households to see how a decrease in IFT seed prices would affect the decision to plant a particular species. In most cases, at full and half seed price, Tephrosia sp., Cajanus sp., or Sesbania sp. was chosen as the best activity for year-end cash optimization. However, when the IFT seed selling activity was removed Gliricida sp. was then chosen as the best species to plant. This is most likely due to the fact that Gliricida sp. does not yield as much seed as the other species and does not produce until the second year as opposed to the first year for the other species. In every case, an

improved fallow activity was chosen meaning that IFTs can be a practical alternative to fertilizer use for small farmers in Eastern Zambia.
The model also compared female-headed households (FHH) to male-headed households (MHH). Females and males often face different cash, labor, and access to inputs so it was important to see what species may be the most appropriate for particular household types. In most cases, Tephrosia vogelii or Cajanus cajun was chosen as the best species for FHHs. When the seed selling option was removed, then Gliricida sepium was chosen as the species to plant for cash optimization. Sesbania sesban was not chosen mostly because females often do not have access to dimba land needed for the initial stages of seedling growth. Sesbania sp. also requires greater care, which some households may not be able to provide.
Recommendations for Further Study
Since the country gained independence, the Zambian government has played a large role in small-farm production through maize and fertilizer subsidies. Fertilizer subsidies, especially, have caused farmers to become dependent on fertilizer inputs allowing farmers to farm one piece of land continuously for several years. In recent years agricultural subsidies have been removed and, as a consequence, farmers have been unable to afford fertilizer and have realized lower maize yields because of infertile soils. While IFTs have been presented as a feasible alternative to fertilizer, if fertilizer subsidies are reinstated farmers may abandon the use of IFTs and return to using fertilizer. The LP model could help determine how a fertilizer subsidy may affect small farm maize production.

There is great diversity among households in Eastern Zambia. As a consequence, the LP did not capture every aspect that may affect a household's livelihood. For instance, males and females in MI-l-s keep their income separate. Females have ways to generate cash that are different than males which enables them to pay their children's school fees and to buy food when food supplies are low. A study of other income generating activities that families adopt to compensate for low yields and how they may affect the adoption of IFTs may be needed. Further, IFTs require additional labor that may not be available in certain households. Therefore, other ways of generating income to buy food may be a better strategy for some households in place of planting an improved fallow and should be considered when targeting technologies to small households.
Based on this study, improved fallows are a practical alternative to using fertilizer to supplement infertile soils and to realize greater yields. However, the technology is fairly new to many villages and a study of households that have adopted the IFTs over two cycles would be valuable. Additionally, a ten-year LP model would be helpful to see if the technology is viable for the household in the long run. This would involve participatory field research over an entire harvest cycle so that the researcher may capture every aspect affecting household farm production and livelihood.

October 1964 Zambia gained independence and the Republic of Zambia was
1964 to 1991 Kenneth Kaunda Presidency United Independence Party
Equity Goal: Increase food production to ensure self-sufficiency for the growing population, to maintain the provision of cheap food, and to decrease past dependence upon imports. Policies: Uniform pricing for maize Price and interest rate controls Import restrictions
* State controlled marketing board Subsidies for fertilizer and food staples Overvalued exchange rate State farms
1969 National Agricultural Marketing Board (NAMBoard) established
1969 to 1980 Small and medium-scale farmers grew from 23% to 36% of the
rural population. Subsistence households declined from 75% to 62%.
1979 to 1983 The Third National Development Plan was implemented.
1975 0 Copper price collapsed
1980 to 1990 "Operation Food Production"
1981 0 Subsidy on maize consumption reduced fifty percent.
1 Jansen and Rukovo 1992; Valdes and Muir-Leresche 1993; Geisler 1992; Copestake 1997; Wood et al 1990; Wichern et al. 1999.

*Management control over marketing decentralized reducing the role of NAMBoard (National Agricultural Marketing Board) and giving more control to the cooperatives.
*Income tax for farmers reduced from 50 to 15 percent.
*Copper reaches lowest real value in 50 years.
*Producer prices increased by three to fifteen percent.
*Retail price decontrolled for major products except wheat, maize, and candles.
1983 to 1985 Agreements with the IMF
*Reduction of public expenditures
*Kwacha devalued 20 percent and permitted to float against basket of currencies of major trading partners.
*Fertilizer subsidies reduced and price of fertilizer allowed to increase by 60 percent.
*Subsidy to NAMBoard reduced in 1983 but increased in 1985.
*Wheat price controls eliminated.
*Producer prices increased.
*Floor prices established for all controlled commodities except maize.
*Consumer maize subsidies reduced and prices allowed to increase 22 percent in 1984 and reduced further (approximately 40 to 50 percent) in 1985.
October 1985 to IMF induced Economic Restructuring 1987 Flexible exchange rate (market determined) system
implemented through weekly auctions Interest rates controlled through Treasury Bill auction
* Price deregulation of all crops except maize, maize meal, and fertilizer.
*Producer prices for maize increased 95 percent.
*Price controls on import substitutes
November 1986 *Decontrol of consumer maize prices
*Fertilizer subsidies decreased and price increase almost 200 percent.
*NAMBoard monopoly on maize and fertilizer eliminated and cooperatives given more control over marketing.
December 1986 *Copperbelt food riots maize price increases rescinded.
*IMF economic adjustment program cancelled
January 1987 *Foreign exchange auction suspended
March 1987 *Kwacha revalued 87% higher
to May 1987 *Foreign exchange auction restarted and ended

May 1987 New Economic Recovery Programme
Theme: "Growth from our own resources" Policies implemented:
*Debt servicing limited to 10% net export earnings
*Fixed interest rates and exchange rate stabilization
*Reintroduction of price controls
*Centralized control of foreign exchange
*Reduction in rate of money supply from 60% in 1987 to '-40% in 1988.
1988 0 Inflation up 64%
* NAMBoard closed
* Agricultural Marketing Act liberalized marketing except for maize and fertilizer.
November 1988 Kwacha devalued 25% and pegged to special drawing rights
* Minimum reserve requirements raised 5%
January 1989 0 Price Maize meal prices increased to reduce budgetary cost of
maize subsidies.
* Coupon scheme for maize foodstuffs introduced
June 1989 Price controls for all crops except maize lifted.
August 1989 Phase I of Medium to Long term Structural Adjustment Policies
Policy Framework Paper (PFP) Marketing of almost all products liberalized
1990 Import tarriffs
*15% import tariff rate on wheat introduced
*Coffee import tariffs reduced 100% to 50%
*Tobacco import tariffs reduced 100% to 30%
199 1/92 Drought
Reform Program
Goals: Bring down inflation and create stable economic climate for growth and diversification through:
* The reduction of government spending by ending maize meal subsidies
* A balanced budget within two years
*The reduce of civil service
*Non-traditional exports
*Decontrolling prices
*Market based exchange and interest rates

*The promotion of the private sector
*The privatization of state run enterprises
What occurred:
*Maize imports, consumer prices for maize, and prices for transportation liberalized
*Fertilizer subsidies removed
*Fuel prices and interest rates increased
*Kwacha devalued
*Competition and Fair Trading Act prohibits anti-competitive trade practices
*Investment Act tax incentives for investors
*Cash budget system
September 1991 The World Bank and the IMF structural adjustment program
October 1991 Frederick Chiluba elected Movement for Multi-Party Democracy
November 1992 Agricultural Sector Investment Program (ASIP)
Objectives (Wichemn et al 1999):
*Achievement of food security at all levels
*Enhancement of income and development
*Conservation and improvement of current agricultural resource base
*Contribution to sustainable development
*Intensification of agricultural exports
*Liberalization of agricultural markets
1992/93 What occurred:
*Liberalization of financial markets
*Interest rates increased
*Import tariffs for all agricultural products increased
*Kwacha devalued 25% and pegged to special drawing rights
*Minimum reserve requirements raised 5%
1994 Drought
1995 *Customs and Excise Act
*Food Reserve Agency (FRA) established
*Milling industry privatized

The following appendix begins with a brief description of the history of maize, how the plant develops, and describes the climatic, soil, and nutritional requirements for optimal growth. Particular emphasis is placed on nitrogen since it is the most important nutrient for maize growth and is currently being exhausted from Southern African soils. The second part of the paper discusses four improved fallow tree species that are the focus of this study: Sesbania sesban, Tephrosia vogelii, Gliricida sepium, and Cajanus cajan. It compares their benefits and costs to maize production and the farmers in Eastern Zambia.
Maize is grown in practically every part of the world, particularly in the tropics. It is of tropical Mexican origin and is a member of the grass family Gramineae, subfamily Panicoideae. Linneaus classified maize Zea mays in 1737 in his work "Genera Plantarum". Zea is a Greek word used to describe cereals and mays is based on the original Indian words, "mahis" meaning the "source of life" or "marisi". The word maize is also based on the words "mahis" or "marisi". Christopher Columbus named the plant after his crewmembers found a sample in Cuba. It is thought that Columbus introduced the species after he brought it to Europe when he returned from his trip to the New World in the 1490s (Berger, 1962). Production of maize spread throughout Europe and into North Africa and was introduced to West Africa by the Portuguese in the early

l6fl" century (Norman, Pearson, Searle, 1995 and Berger, 1962). It was established as a primary crop in the Congo around 1930 (Norman, Pearson, Searle, 1995 and Berger, 1962). Now maize is produced in nearly every part of the world. It is the third most important cereal crop in the world after rice and wheat. The biggest producers of maize are the United States, China, and Brazil.
Maize Development
A maize plant grows 2 to 3.5 m tall and has a growing season anywhere from 90 to 120 days, depending upon latitude, climate and environmental conditions. It has two main development stages, the vegetative, or initiation stage, and the reproductive stage. These stages can be further divided into the following five phases (Berger 1962):
1. Planting to emergence.
2. Emergence to tasseling and silking.
3. Pollination and fertilization.
4. Grain production from fertilization to maximum dry weight of the grain.
5. Maturation or drying of grain and stalk.
The vegetative phase begins when the planted seed absorbs water, the rootsdevelop, leaves are produced, and the last branch of the tassel is visible, about 55 days after emergence (Ritchie and Benson 1993; Berger 1962). The silks become visible outside the stalk at about 60 days, when the pollen is shed and anthesis (flowering) begins. This is the most critical stage of development for maize. "Stress during pollen shed and silking can cause more yield loss than almost any other period in the crop's development (Nielson 2001)." After pollination and fertilization occurs the kernels start to grow. The process to maturity takes approximately 115 days depending upon the variety of maize being grown.

Climatic Requirements
The growth rate of maize is directly related to temperature. It is very important in determining the time of tasseling (Berger 1962). Cooler temperatures will delay growth. Higher temperatures, with high relative humidity during the day, speed the growth rate in the early stages and also determine the number of grains formed during the grain forming period (Norman, Pearson, Searle 1995). More grains develop with higher temperatures and adequate moisture. However, high temperatures, in conjunction with drought stress, can scorch exposed silks and damage pollen grains thereby affecting yield (Nielson 2001). Optimal temperatures for growth range from 16'C to 36C (Norman, Pearson, Searle 1995; Berger 1962; Saskatchewan Interactive).
Maize is very susceptible to drought at flowering and pollination. Water stress can delay silking and seriously affect grain yield (Norman, Pearson, Searle 1995; Nielson 2001). Symptoms of water stress include leaf curling and darkened coloring. Equally, maize will not tolerate flooded conditions. Annual rainfall should be about 460 to 800 mm for optimum growth (Berger 1962; The Maize Page, Iowa State). Soil Requirements
Maize adapts well to a variety of soils but grows best on well structured, adequately drained, and aerated soils that contain an abundance of nutrients and organic matter. Most maize is grown on loams, silt loams, and clay loams, which exhibit these characteristics. In the tropics Oxisols, Ultisols, Alfisols, and Inceptisols are best suited for maize production (Norman, Pearson, Searle 1995).

Nutrient Requirements
The most important nutrients for maize growth are nitrogen, phosphorous, and potassium, which depend upon the fertility of the soil. Adequate uptake per hectare for maize is about 167 to 241 kg N, 50 to 90 kg P205, and 101 to 196 kg K20 (Berger 1962). Plants take up nutrients throughout the growing season but most nutrients are extracted about 10 days before tasseling until approximately 25 to 3 0 days after tasseling (Norman, Pearson, Searle 1995; Berger 1962). Nutrient deficiencies prolong the interval from emergence to tasseling or silking.
Nitrogen deficiency appears first in older parts of a plant as nitrogen is
translocated from older leaves to the younger leaves. Eventually the whole plant will start to yellow, a process called chlorosis. Other symptoms include stunted growth, reduced flowering, and lower protein content.
Soil Fertility and Improved Fallows
Maize is a very important food source for people living in Southern Africa. With the occurrence of drought and infertile soils, farmers are having difficulties maintaining adequate maize harvests and, hence, having to battle food insecurity. Soils are low in nitrogen primarily due to previous cropping and leaching below the root zone to the lower soil profiles making it harder for maize to access it. In soils where leached nitrogen is retained in the subsurface horizons, deep-rooted trees used in rotation with annual maize crops are capable of extracting this nitrogen from the deep subsoil and using it to enrich the surface soil. Maize yields in fields where improved fallow species were grown have been seen to be equivalent and sometimes greater than with the use of

fertilizer. Such agroforestry practices have the potential to make a significant contribution to agricultural production and food security in Southern Africa.
Of the several species of nitrogen-fixing trees currently researched, Sesbania
sesban, Tephrosia vogelii, Gliricidia sepium, and Cajanus cajun are the focus of this study. A brief comparison of each species is offered below. Sesbania sesban
Sesbania sesban is a small, deep-rooting tree that usually grows one to seven
meters in height (ICRAF Online). It is widely cultivated throughout Africa and Asia but Africa is its center of diversity (ICRAF Online). It will grow well in any soil (UC SAREP Online), particularly sandy and clay loams (Kwesiga and Beniest 1998). It is not unusual to find the tree in seasonally flooded areas that are common along streams or swamp banks, as it ideally suited to flooded areas (ICRAF Online; UC SAREP Online). In addition to increased maize yields, advantages of Sesbania sesban include (Kwesiga and Beniest 1998; Opio 2001; Raussen 1997; Buresh and Tian 1998; ICRAF Online; UC SAREP Online):
1. Weed suppression 2. Drought resistance
3. Good seed production
4. Unpalatable to livestock
5. Used as firewood or stakes
6. Used for making ropes and fishnet
7. Flowers are edible
8. High quality forage source
9. Leaves, roots, and seed can be used medicinally
After a three year fallow, Sesbania sesban provides 23 mg kg'' of inorganic nitrogen, which is immediately available to plants; .51 mg kg'1 day-' of aerobic nitrogen that becomes available over time, and 28 mg kg-1 soil nitrogen in light fraction soil organic matter (SOM), the most active part that will break down during the mineralization

process and become available to the plant. The main disadvantage to planting Sesbania sp. in a fallow field is the amount of labor required; seedlings need a great amount of detail, care, and labor since they must be planted in a nursery before being transplanted to a field. The raised beds in the nursery must have a thin layer of soil from a healthy stand of Sesbania sp., which serves as an innoculant for the root nodules, and a layer of riverbed soil where the seeds are drilled (Kwesiga and Beniest 1998; Raussen 1997; farmer interview 2000). The seed has to be prepared with boiling water the night before planting and once planted in the nursery, must be shaded in extreme cases of heat and drought, watered, and weeded twice daily for two weeks (Kwesiga and Beniest 1998; Raussen 1997; farmer interview 2000). While it has been observed that some farmers have direct seeded the plants with some success (Donald Phiri, World Vision personal communication) farmers with no former experience with nurseries require training and, for female farmers, nurseries increase their already heavy workload and may compete with other farm activities.
In addition to increased labor requirements in the early stages, Sesbania sesban will not do well in areas infested with nematodes (Kwesiga and Beniest 1998). If one area of the field is infested then the entire tree stand will become infected and any susceptible crops planted in the same area will also be infected (Kwesiga and Beniest 1998). Insect pests may also be a problem. Leaf-eating and sucking insects, such as beetles, caterpillars, aphids, weevils, and thrips, can destroy flowers, buds, fruits, or seeds and defoliate trees. Ants and termites can also damage the roots and stem (Kwesiga and Beniest 1998; Raussen 1997).

Gliricidia sepium
Gliricidia sepium is a tree that grows to a height of 2 to 15 meters (ICRAF
Online). It occurs naturally in coastal sand dunes, riverbanks, floodplains and fallow land, and it establishes well on steep slopes (ICRAF Online). Some of the benefits of Gliricidia sepium are as follows (ICRAF Online, New Crop, FactNet) :
1. Will grow well in a variety of soils from pure sand to deep alluvial lake-bed
2. Improves soil aeration
3. Reduces soil temperature
4. Drought resistant and water conserving
5. Tolerates browsing
6. Controls erosion
7. Provides shade for shade loving food crops
8. Suppresses weed growth
9. Resistant to termites
10. Seeds can be stored for up to 12 months
11. Wood can be used for firewood and charcoal production or for household
12. Attracts honeybees, hence, can be an important species for honey production
13. Leaves or seeds can be used as a pesticide
14. High quality food supplement for large and small ruminants
In addition, the seeds can be directly sown provided that the land has been properly prepared. The area planted does require regular weeding and it is recommended that the trees be pruned and pollarded as part of a regular maintenance regime. After two years of fallow when the trees are cut, Gliricidia sepium will regenerate, whereas Sesbania sesban has to be replanted. The disadvantage of Gliricidia sepium is that maize yields are less than yields of maize planted in a Sesbania sesban improved fallow plot, 3600-4050 kg ha"1 versus 4950-5400 kg ha respectively (personal communication World Vision). This means that Sesbania sesban is able to retrieve more nitrogen than Gliricidia sepium. After a fallow of Gliricidia sepium 18 mg kg-' of inorganic nitrogen, .30 mg kg-1 day-' of

aerobic nitrogen, and 20 mg kg"1 soil nitrogen in light fraction SOM is made available (all figures from Kwesiga and Beniest 1998). Tephrosia vogeli
Tephrosia vogelii is a small tree that reaches heights of 0.5 to 4 meters. It is native to Africa and can be found in a wide range of habitat, including savannah-like vegetation, grassland, shrub land, and wasteland. It grows well on acid soils, such as andisols, and well-drained loams, and is tolerant to soils with low pH, although low pH affects the rate of growth and makes the tree prone to diseases (ICRAF Online). Termites can be a severe problem and the trees are susceptible to livestock damage (Raussen 1997; ICRAF Online). Some of the benefits of Tephrosia vogelii include (ICRAF Online; McGregor 1976):
1. Can be direct seeded
2. Tolerant to drought, wind and grazing
3. Holds up under repeated pruning
4. Suitable as a hedge plant because of its dense growth 5. Seeds can be stored for up to 3 years in open storage
6. Unaffected by burning
7. Can be used as an insecticide or for fish poison
8. Honey bees have a role in pollination, which could have a positive role in
honey harvesting
While maize yields after Tephrosia sp. are 3.18 t ha71 compared to that of S. seban, 5.36 t ha-1 (Kwesiga and Beniest 1998), Tephrosia sp. requires less labor. Because of the deep root system of the trees they are more apt to re-sprout, meaning that no seeds are needed for a new improved fallow field and, therefore, less labor is required. Cajanus cajun
Cajanus cajan, or pigeon pea, is a short-lived shrub (1-5 years) that will grow from 0.5 to 4 meters high with thin roots that can penetrate up to two meters of soil

(ICRAF Online). It originated in India, eventually spread to Southeast Asia, and reached Africa around 2000 BC. It is now found all over the tropics, especially East Africa and the Indian subcontinent (ICRAF Online). Its natural habitat is uncertain because it has not been found in the wild. While it prefers grassy areas in warm, tropical climates with 600 to 1000 mm annual rainfall, it will grow in areas with rainfall as high as 2500 mm annually and as low as 400 mm annually. Cajanus sp. will grow from 0 to 2000 meters above sea level and will tolerate a wide range of soils with different physical and chemical characteristics. The most ideal soils are alluvial soils with a ph range of 5 to 7 (ICRAF Online). It is sensitive to high salinity soils and to water logging. In addition to soil improvement Cajanus sp. has other advantages and uses. They are as follows (ICRAF Online):
1. Erosion control 2. Shade or shelter
3. Ideal for intercropping
4. Seeds are edible and nutritious
5. Can be used for fodder
6. Honeybees collect nectar, therefore can be an important honey source
7. Can be used as fuel wood
8. May be suitable for making paper 9. Can be used for light construction
10. Has medicinal uses
Termites can be a problem for Cajanus sp., therefore it should not be planted in areas where termite infestation is known to be a problem (Raussen 1997). Flower- and pod-sucking insects can also hinder growth and health of the plant and during the dry season it is susceptible to damage from grazing livestock.

armer #
Type of farmer NT T/NE TE
District\Village\Camp Gender M F
1. Name
2. Head of household? FHH or MHH?
3. Marital Status El Single El Widow L Married E Divorced E3 Polyg Spouses live together 0 Polyg Spouses live separate
4. Household composition: Young with children<5 yrs; Established; Older with
children>18 yrs
5. Number of People in Household Ages, Gender (number of people sharing food,
income, costs, etc.) 6. Where were you born?
7. How long did you stay in school?
8. How long have you lived in this village?
9. Whose village do you live in (mom, dad, spouse)? 10. Ethnic group
11. Wealth status (ICRAF scaling method)
El Very poor E3 Poor L Fairly well off EWell off FARM DESCRIPTION
12. How many fields do you have? 13. How long have you been cultivating this field? yrs 14. What crops do you grow? Hybrid or local maize? 15. Area planted for each crop.

16. When is each crop planted? 17. Who plants each crop? 18. How long does it take to plant? 18a. How many hrs per day? 19. What is the amt of seed planted for each crop? 20. Did you buy seed for any of the crops? What crops? 21. Who buys it? 22. How much is paid for each amt? 23. When do you buy the seed? 24. What were your yields for each crop? 25. What are your expected/avg yields for each year? A normal year, not an
exceptionally good or bad year. 26. When is each crop harvested? 27. Who harvested it? 28. How long did it take to harvest? 29. Did you sell any crops last year? What ant of crops did you sell last year? 30. When did you sell the crops? 31. How long does it take to sell? 32. Who sold the crops? 33. What amount did you eat? 34. Did you cultivate with oxen? 35. Did you hire/rent the oxen? How much did you pay? 36. When is each crop weeded? 37. How many times is each crop weeded? 38. Who weeds it? 39. How long does it take to weed? 40. Are any crops intercropped (planted together)? If yes, which ones?
41. Do you have a garden (dimba) plot? 42. Did you cultivate the garden this year?

43. What did you cultivate in your dimba? 44. What is the area of your diniba? 45. When is it planted? 46. Who plants it?
47. How long does it take each person to plant? 48. When is the garden weeded? 49. How many times is it weeded? 50. Who weeds it?
51. How long does it take to weed? 52. Did you buy seeds for any of your dimba crops? If so, which ones? 53. When did you buy seeds? 54. How much did you pay for the seeds? 55. When is the crop harvested? 56. Who harvests it? 57. What were your dimba yields last year? 58. What are your expected dimba yields in an average year? 59. Did you sell any of your dimba crops? If yes, which ones? 60. How much did you sell your dimba crops for? 61. When did you sell them? 62. How much maize does your HH need per month? 63. What do you do with the husks? 64. How long does it take to pound maize? 65. Who does it?
66. How long do you wait in line at the grinding mill? 67. How much time is spent collecting firewood? 68. Do you ever buy firewood? If so, how much is spent?
FARM EXPENSES 69. Did the maize you harvested last year last until you harvested this year? 70. What month did the maize you harvested last year finish?

71. How did you get the rest of the maize you needed to last all year? 72. How much did you have to buy for a week (or month)? 73. When did you buy the maize? If do ganyu to get maize, what month? 74. How much did you pay for it? 75. Who paid for it?
76. How much maize do you need to last all year? 77. Do you buy food other than maize? If so, what? 78. How much money do you need to buy the food per month? 79. Who pays for the other food? 80. Do you hire labor?
81. When do you hire the labor? 82. How much do you pay for the labor? 83. Who pays for the labor?
84. Do you use fertilizer, manure, or pesticides on your crops or dimba? 85. If no, why not? (if yes, skip to question #92) 86. Are there any other inputs (other than fertilizer) that are used? 87. If so, what are they? 88. How much cash is spent on these inputs? 89. Who pays for these other inputs? 90. When is fertilizer/pesticides/manure applied? 91. Who does it?
92. How long does it take? 93. Is the fertilizer (or manure) purchased, receive credit for, or do you barter for? (if
barter go to #99)
94. How much is purchased and how much is paid for the fertilizer (manure)? 95. Who purchases the fertilizers? 96. When is the fertilizer (or manure) purchased? 97. If you barter do you exchange work or crops for fertilizer or manure?

98. With whom/how? 99. In the past few years, have you decreased your fertilizer use? 100. From how much to how much? 10 1. Why did you decrease the amount of fertilizer you use? 102. Have you changed the crops that you plant or the way that you plant your crops
because you use less fertilizer? How?
103. Have you noticed changes in your crop yields with less or no fertilizer? 104. If yes, can you give me an example of how your yields have changed (quantify)? 105. Do you ever buy small (<50 kg) bags of fertilizer? If Yes, when? 106. If yes, why do you buy small bags? 107. Do you buy a buy a small bag of fertilizer for your dimba because you need less
than 50 kgs?
108. Do you buy a small bag for your maize field because you have some money but not
enough for a 50 kg bag? FARM ANINULS
109. Do you have any animals? 110. What kind? Goats Cattle Chickens Pigs Oxen
111. How many do you have? 112. What is done to take care of them (feeding, housing, etc)? 113. Who does this? 114. How long does it take each day to care for them? 115. How much food do they get? 116. If the food is bought, where do you get the feed? 117. How much do you pay? 118. If grown, where is it grown? 119. How much labor does it take to grow it? 120. Do you use the animals for meat? 121. If so, how often do you use them for their meat? 122. Are they ever sold? 123. How much money would you get for one?

124. How often do you sell them? 125. Do you ever sell chicken eggs? 126. How much would you sell them for? 127. How often?
128. Who in the family would get the money?
129. How much food is needed for the HH per month (or day/week)? 130. What items in your HH do you need cash for on a monthly basis? 131. How much money do you need per month to pay for these items? 132. Who usually pays for these? 133. Are your children in school? 134. If yes, what things do you have to buy for your children's school? 135. How much do they cost? 136. When and how often do you have to buy these? 137. Who usually pays for this? 138. Who makes HH decisions regarding: Land and its uses; Money; Family FALLOW LAND QUESTIONS: Ask about last planting season or just before planting in Oct-Nov:
139. Kodi muli ndi chisala? (Last November, did you have a fallow field not IF? If No,
skip to #147)
140. Chisala cako, cili ndi zaka zingati? How old was the fallow? (how many years did
it have?)
141. Chisala cako, cili cacikulu bwanji? How big was the fallow? 142. Did you have strength to open it last planting season? 143. Do you rotate crops into the fallow? If Yes, what crops? 144. What is the difference in yield in maize grown in the fallow land and the maize not
grown in the fallow land?
145. Kodi muli ndi 'tengo kapena sanga? Last November, did you have uncleared land
(if No, skip to #153)?

146. Tengo/sanga ili yaikulu bwanji? How big was uncleared land? 147. Kodi muli nayo mphanje? (Last October-November, did you have land ready to be
cleared (come out of fallow) and cultivated?)
148. Mida zonse ndi cisala ili zikulu bwanji? How big are all your fields plus your
149. Munda mulima lomba inali munda ya ndani? How did you get the land you
cultivate now? (Father, mother, headman, etc.) 150. Do you ever work with the camp officer here? 151. How often does the CO work w/ you? 152. Which activities?
a. Mbeu yoyamba -- improved crop varieties
b. kubiala mitengo na chisala -- improved fallows
c. kubwelestsa nthaka -- soil conservation
d. agricultural groups or cooperatives
e. zoweta -- animal husbandry
f. other:
IMPROVED FALLOW QUESTIONS DECISION to {Plant IF Last Season; Don't} 153. In Nov -Dec, were you aware of the program to plant trees to improve your sofils?
Had you heard of it?
154. How did you learn about the trees and how to plant them? (FTC, village meeting,
field day, who)
155. In Nov-Dec-Jan (this year's planting season), did you plant trees in an improved
fallow? (if no, skip to #184)
156. Who decided to plant the improved fallow (you or spouse, father, mo-in-law)? 157. Which trees did you plant? (Species) 158. How big is area of trees? 159. How old is the IF? When did you plant the trees? 160. How long did it take to plant? 161. Where did you find seeds? Did you buy the seeds? 162. How much did the tree seeds costs?

163. Who paid for the seeds? 164. Did you have any problems cultivating the trees? If so, what?
Termites beetles drought animals fire death in nursery other 165. Did you have a method of decreasing the problem? If so, what? 166. How much time does that take? 167. Did you plant any crops in with the IF? 168. Have you cut and plowed the trees under? 169. When?
170. How long did it take? 171. Who plowed the trees under? 172. After you cut the trees, what did you plant in the field? 173. What is the difference in yields from the IF field and your other fields? 174. Do you have a nursery? If yes, Where is the nursery and what type of nursery is it? 175. Were the trees pruned? 176. If so, when are they pruned? 177. How long does it take to prune them? 178. Do the trees have maintenance costs (fertilizer, etc)? 179. If so, what are they and how much do they cost? 180. Who pays for them? 181. Have you planted trees in many fields? How many? 182. If you had not planted an improved fallow here, what would you have done with the
Ask everyone, whether or not they planted trees: 183. Did you have tired soils (nthaka yosira) on the land you cultivated the previous
184. In November-December-January, was fertilizer expensive (yodula) or beyond your
185. (NT & NE) Because you had (mphanje) land ready to come out of fallow, did you
think you did not need to plant trees to improve the soils? YES, satisfied, no
need for trees; NO, not satisfied, needed trees

186. (NT & NE) Because you were able to buy fertilizer, or barter for it, or get credit for
it, were you satisfied with soil fertility, so did not need to plant trees?
YES, satisfied, no need for trees; NO, not satisfied, needed trees
187. (NT & NE) Because you used manure in the past on your fields, did you think you
did not need to plant trees to improve the soils in your fields?
YES, satisfied, no need for trees; NO, not satisfied, needed trees
188. (NT & NE) Because you rotate crops in your fields, in Nov-Dec-Jan did you think
you did not need to plant trees to improve the soils?
YES, satisfied, no need for trees; NO, not satisfied, needed trees
189. At planting time, had you already seen for yourself enough benefits (pindu) to
encourage you to plant trees?
190. What benefits did you see? Where did you see them?
191. (NT & NE) At planting time, were you willing to plant trees and wait 2 years to see
the benefits (pindu)?
192. (NT & NE) Were your soils so bad that it was worthwhile for you to spend time
planting trees during the busy planting season (end November to 2nd week
193. At planting time, were you aware you could direct-seed Tephrosia or Cajanus cajan
(pigeon pea) in fields so they don't need a nursery?
194. Did you have help from your family, or could hire piece-work (ganyu) to plant
trees? OFamily help; U ganyu; U no help
195. (NT & NE) At planting time, did you have relatives, friends, or a camp officer with
seeds/seedlings to give you, or sell you (and you had the money)? did you have
the seeds/seedlings necessary to plant trees?
196. (NT & NE) At planting time, did you have enough land to try a small part to plant
197. (NT & NE) At that time, were you renting or borrowing the land where you
cultivated so that you did not want to plant trees there?
198. (NT & NE) Did you have the power (mphavu) or authority (malamuno) to plant
trees in a plot?

199. (NT & NE) Why not?
200. (NT & NE) At planting time, were you worried about cows or other animals getting
into your improved fallow plot? If yes, did you do anything to decrease the
20 1. (NT & NE) At planting time (when you planted this year's crops), were you
worried about the loss of the trees due to beetles or termites? If yes, did you do
anything to decrease the problem?
202. (NT & NE) At planting time (when you planted this year's crops), were you
worried about the loss of the trees due to fire or drought? If yes, did you do
anything to decrease the problem?
203. (NT & NE) At planting time, did you think the risk of losing the trees in the plot
was less than the risk of depending just on fertilizer (and/or manure, crop
rotations) for soil fertility?
El Y Trees less risky; El N Trees more risky than fertilizer or manure
204. At planting time, did you think the trees were more reliable than fertilizer?
E3 Y Trees more reliable cause fertilizer is often unavailable;
U N Fertilizer more reliable
205. At planting time, did you think the trees were cheaper than fertilizer?
206. (NT & NE) At planting time, did you think the trees were too much work to be
worthwhile to you?
U3 Y Trees too much work; U N Trees are easy
Take out bags of legume seeds to show the informant.
207. At that time, did you know about any of these legumes (Groundnuts, banibara. nuts,
pigeon pea, cowpeas, sunhemnp, velvet beans (mucuna), soy beans)?
208. Of those you know, did you also know (at that planting time) if any of these
legumes improve the soil?
209. At that planting time, were you also willing to plow in _____before the velvet
beans came, and bury the green leaves, just to improve your soil?

2 10. Are you a member of a club? 211. What kind of club? Describe the club activities. 212. Does the club encourage planting of improved fallows? INCOME STREAM QUESTIONS 213. Does anyone in your HH have access to credit? 214. Does your HH use credit? 215. If so, what do you use it for? 216. How do you pay back the borrowed amount? 217. How much credit do you use? 218. How much do you have to repay? 219. Are there other ways you get money?
Activity: Ganyu Milk
Bricks Buns
Beer Traditional medicine
Tuck shop Mats
Pots Sewing
Selling firewood/Charcoal Children in town
Salary/allowances Other:
All The Time or Seasonally:
220. Are you paid in cash for this work? 22 1. Do you receive any food for this work? 222. If yes, how much? 223. What does your spouse do to get money? Ask for every family member
Activity: Ganyu Milk
Bricks Buns
Beer Traditional medicine
Tuck shop Mats

Pots Sewing
Selling firewood/Charcoal Children in town Salary/allowances Other:
All The Time or Seasonally:
224. Do they receive cash for this work? 225. Do they receive food for this work? 226. If yes, how much? 227. Do you have any questions for me?