The Impact of Selected Small-Scale Irrigation Schemes on Household Income and the Likelihood of Poverty in the Lake Tana Basin of Ethiopia

THE IMPACT OF SELECTED SMALL-SCALE IRRIGATION SCHEMES ON HOUSEHOLD INCOME AND THE LIKELIHOOD OF POVERTY IN THE LAKE TANA BASIN OF ETHIOPIA

A Project Paper Presented to the Faculty of the Graduate School of Cornell University in Partial Fulfillment of the Requirements for the Degree of Master of Professional Studies

By Getaneh Kebede Ayele July 2011

ABSTRACT

Poverty reduction has been largely a result of economic growth. The economic growth and extent of poverty in Ethiopia are determined primarily by the growth of agriculture because agriculture is the largest component of the economy. One of the major factors behind the weak performance of Ethiopian agriculture is lack of adequate rainfall, combined with variability in the onset and duration of rainfall. Irrigation development is one approach to address this problem, and it has been given significant attention in economic development programs in the country. This study examines the impact of selected small-scale irrigation schemes on crops grown, total income, and the likelihood of poverty at household level for a particular region. A survey of 180 randomly-selected household heads, semi-structured interviews and focus group discussions were undertaken in Fogera District, Tana basin, Ethiopia to assess irrigation impacts. Descriptive statistics and econometric modeling were used to assess the impacts of irrigation on household income and the likelihood of a household being classified as poor. The research reported herein had five major objectives. The first objective was to identify the major field crops and vegetables grown using small-scale irrigation in the study area. The main field crops grown using small-scale irrigation schemes in the study region are maize, oat, rice and vetch and the dominant vegetables are onion, tomato, potato and pepper. Onion production was the most important source of income from crops grown with irrigation. The second objective was to compare the relative advantages of four types of small-scale irrigation system, with emphasis on household gross income. Farmers using concrete canal river/spring diversion had higher mean cropping income per household on average than other irrigation types. Statistically significant differences were found between the household concrete canal river/spring diversion and

traditional river diversion and pedal pump irrigation systems, but no significant difference exists between concrete river/spring diversion and motor pump, nor between traditional river diversion and pedal pump. Households using any of the four irrigation systems had statistically significantly higher mean gross household income than households not using irrigation. A third objective was to estimate the marginal impact of small-scale irrigation on gross household income controlling for other important factors that affect income.

A censored regression model developed for this objective indicated that access to small–scale irrigation increased mean annual household income significantly (about ETB 3,353 per year, or a 27 % increase over non-irrigating households). The fourth objective of this research was to assess the impact of irrigation access on the likelihood of poverty. Descriptive analysis suggested that irrigating households had a lower probability of being poor than non-irrigating households: of households in the lowest quartile of income, only 12% were irrigating households and the remaining 88 % did not irrigate. A Logit regression model developed to assess the impact of irrigation on the likelihood of poverty controlling for other factors indicated that access to irrigation significantly reduced the odds that a household would be in the lowest quartile of household income, the key poverty threshold used in this study. A final objective was to examine the major problems encountered in the use of the small-scale irrigation systems. These were identified by farmers and development agents as: lack of access to surface water, loss of water through seepage, problem of irrigation water distribution, lack of spare parts for water pumps, high cost of fuel for water pumps, lack of market transparency and marketing facilities, crop disease, and the perceived high cost of inputs.

BIOGRAPHICAL SKETCH

Getaneh Kebede Ayele was born in the Quarit District in Western Gojjam Administrative Zone, Amhara Region, Ethiopia to his father Kebede Ayele, and his mother, Workie Gessesse Ayele, on January 21, 1980 GC. He attended his primary and junior secondary schools at Quarit District in Gebeze Mariam School, and his secondary education at Damot Senior Secondary School in Fenoteselam town. He graduated from Alemaya University with a BSC degree in Agriculture majoring in Animal Science on 3 July 2002 and a BA degree from Bahir Dar University in Economics on 12 July 2008. He was employed by the Ministry of Agriculture and worked at the Werota Agricultural College until starting this MPS program conducted by Cornell University at Bahir Dar University. His opportunity of sharing knowledge and skill from enormously experienced Cornell University professors inspired him to continue his study further to PhD.

iii ACKNOWLEDGEMENTS

First and for most, I would like to extend my unshared thanks to the almighty God for providing me the opportunity for what I have achieved. I am highly indebted to my research advisors Professor Tammo S. Steenhuis and Professor Chuck F. Nicholson for their generous devotion in encouragement, insight, guidance, and professional expertise from the early design of the research proposal to the final write-up of the thesis. No words can suffice to express my honored thank and gratitude to Dr. Amy S. Collick, Seifu Tilahun and Essayas Kaba for their generous assistance and helpful encouragement during my study with all their kindness through sharing the ups and downs. Great appreciation and special thanks to Dr. Angela Neilan, Dr. Bowman and all other professors who taught me for their unreserved help. I am also grateful to Cornell University for funding this research. A special word of thanks goes for all staff members of the Fogera Woreda Office of Agriculture and Rural Development who provide me technical assistance and transport service. My special and particular thanks go to my friend Hailesysus Ambaw, who lived in USA, for his materials support and encouragement throughout my study. My wholeheartedly thanks should go to Habitamu Addis, Meseret Belachew and all my classmates. I would like to forward my warm appreciation and great thanks to my friend Zemenu Yayeh for his support and encouragement throughout my study. Finally, I am extremely grateful to my father Kebede Ayele for his dedicated partnership in the success of my life.

I dedicate this thesis manuscript to all participants in Cornell-Bahir Dar Universities MPS program, especially for Professor Tammo S. Steenhuis and Dr. Amy S. Collick

v TABLE OF CONTENTS

BIOGRAPHICAL SKETCH ...... iii ACKNOWLEDGEMENTS ...... iv TABLE OF CONTENTS ...... vi LIST OF FIGURES ...... ix LIST OF TABLES ...... x ABBREVIATIONS AND ACRONYMS ...... xi CHAPTER ONE ...... 1 1 INTRODUCTION ...... 1 1.1 Background and justification ...... 3 1.2 Statement of the problem ...... 5 1.3 The Goal of the research ...... 7 1.4 The specific objectives ...... 7 1.5 Hypotheses ...... 8 CHAPTER TWO ...... 9 2 REVIEW OF RELATED LITRATURE ...... 9 2.1 Poverty ...... 9 2.2 Water and agriculture ...... 10 2.3 Irrigation development ...... 11 2.4 Irrigation methods ...... 13 2.4.1 Surface irrigation ...... 13 2.4.1.1 Basin irrigation ...... 14 2.4.1.2 Furrow irrigation ...... 14 2.4.2 Flood irrigation ...... 14 2.4.3 Border irrigation ...... 14 2.5 Sprinkler irrigation ...... 15 2.6 Drip irrigation...... 15 2.7 Irrigation-poverty linkages ...... 16 CHAPTER THREE ...... 18 3 THE STUDY AREA AND SMALL-SCALE IRRIGATION TYPES ...... 18 3.1 Description of the study area...... 18 3.1.1 The characteristics of the sample PAs ...... 20 3.1.1.1 Kuhir Michael ...... 21 3.1.1.2 Shina ...... 21 3.1.1.3 Abana Kokit ...... 22 3.1.1.4 Bebekis ...... 22 3.1.1.5 Werota Zuria ...... 23

vi 3.2 Small-scale irrigation types ...... 23 3.2.1 Concrete canal river diversion ...... 23 3.2.2 Spring development small-scale irrigation scheme ...... 28 3.2.3 Motorized pump ...... 29 3.2.4 Pedal pump ...... 29 3.2.5 Traditional river diversion ...... 34 CHAPTER FOUR ...... 35 4 MATERIALS AND METHODS ...... 35 4.1 Research methods...... 35 4.1.1 Approach for data collection, entry and checking ...... 35 4.1.2 Data analysis ...... 39 4.1.2.1 Socio-economic and demographic characteristics of the sample households ...... 39 4.1.2.2 Income evaluation ...... 40 4.1.2.3 Poverty level evaluation ...... 48 4.1.2.4 Poverty Line ...... 49 4.1.2.5 Poverty level comparison ...... 50 4.1.2.6 Econometric model specification ...... 51 CHAPTER FIVE ...... 54 5 RESULTS AND DISCUSSION ...... 54 5.1 Household Socio-economic characteristics...... 54 5.1.1 Family size ...... 54 5.1.2 Family labor ...... 55 5.1.3 Dependency ratio ...... 55 5.1.4 Sex and education of the household head ...... 56 5.1.5 Age of household head ...... 57 5.2 Productive resource ...... 57 5.2.1 Land holding ...... 58 5.2.2 Effect of irrigation on land rent value ...... 60 5.2.3 Production assets ...... 60 5.2.4 Type of houses ...... 61 5.3 Major crops grown using small-scale irrigation...... 61 5.4 Household income evaluation ...... 63 5.4.1 Cropping incomes ...... 64 5.4.1.1 Rainfed cropping income ...... 65 5.4.1.2 Irrigated crop income in PAs ...... 66 5.4.1.3 Total cropping income ...... 68 5.4.2 Livestock income ...... 68 5.4.3 Off-farm and other incomes ...... 71 5.4.4 Summary of income sources at household level ...... 72 5.4.5 Econometric model for income analysis ...... 73 5.4.6 Comparison of sample small-scale irrigation types at household level . 79 5.4.6.1 Sample small-scale irrigation types and irrigated crop income...... 79

vii 5.4.6.2 The small-scale irrigation types and total income of household ..... 82 5.5 Poverty analysis ...... 83 5.5.1 Poverty level in the study area ...... 83 5.5.2 Multivariate Logit regression ...... 86 5.6 Problems encountered in small-scale irrigation development ...... 89 CHAPTER SIX ...... 97 6 CONCLUSIONS AND RECOMMENDATIONS ...... 97 6.1 Conclusions ...... 97 6.2 Policy implications ...... 99 6.3 Limitations and questions for future studies ...... 103 REFERENCES ...... 105 APPENDIX- A: TABLES OF CROP VALUES AND CONVERSION FACTORS. 115 APPENDIX- B: SURVEY QUESTIONNAIRE ...... 117

viii LIST OF FIGURES

Figure 1: Location of Fogera District ...... 19 Figure 2: Guanta river dam ...... 24 Figure 3: Irrigation and livestock ...... 25 Figure 4: Motor pump irrigation from river diversion canal ...... 26 Figure 5: A women and Children fetch water from river diversion for household consumption...... 27 Figure 6: The river diversion dam on Eriza River in Werota Zuria ...... 27 Figure 7: Tanqua Gabriel spring development ...... 28 Figure 8: Motor pump irrigation using Eriza River near Werota town ...... 29 Figure 9: Pedal pump ...... 30 Figure 10: Sample wells constructed for only one cropping season ...... 31 Figure 11: Sample of Vertisol ...... 31 Figure 12: Well constructed from tire materials ...... 32 Figure 13: Irrigation by fetching water from wells ...... 33 Figure 14: Sample well constructed by cement cylinder ...... 33 Figure 15: Water loss through seepage from river diversion canal ...... 90 Figure 16: Water loss from motor pump ...... 92 Figure 17: Parts of pedal pump demonstrate loss of tightness ...... 93 Figure 18: Non-functional shallow well ...... 95

ix LIST OF TABLES

Table 1: Summary of sample size by PA and irrigation types ...... 37 Table 2: Summary of dependent and independent variables codes, definitions and expected sign of effect on household income ...... 46 Table 3: Summary of the dependent and independent variables, codes, definitions and expected signs...... 52 Table 4: Family size, family labor and dependency ratio for irrigating and non- irrigating households ...... 54 Table 5: Household member, gender, and education and age characterization ...... 57 Table 6: Average landing size (ha) at household‟s level ...... 59 Table 7: Average land rental rate ...... 60 Table 8: Mean value of agriculture production assets at household‟s level ...... 61 Table 9 : Housing types in samples households ...... 61 Table 10: The major field crops and vegetables grown using small-scale irrigation ... 62 Table 11: Reason for selecting the major field crops and vegetables for irrigation ..... 63 Table 12: Major crop types and their mean annual production values ...... 65 Table 13: The reasons for non-irrigating households for not irrigating ...... 66 Table 14: Rainfed income for irrigating and non-irrigating households in ETB ...... 66 Table 15: Income from irrigated crop production in ETB ...... 67 Table 16: Total mean annual cropping income at household level in ETB ...... 68 Table 17: Number of livestock (TLU) ...... 69 Table 18: Average annual livestock income ...... 70 Table 19: The mean off-farm and other incomes ...... 71 Table 20: Summary of annual household income sources: ...... 72 Table 21: Tobit estimates of the determinants for household total income ...... 75 Table 22: Marginal effects of determinants on household total income ...... 78 Table 23: The sample small-scale irrigation types and irrigated crop income per irrigating household ...... 80 Table 24: The sample small-scale irrigation types and irrigated crop income ...... 81 Table 25: Small scale irrigation types and total income of a household ...... 82 Table 26:Small-scale irrigation types and the mean annual income of a household .... 83 Table 27: Poverty comparison between irrigating and non-irrigating household ...... 84 Table 28: The average income poverty gap of the poor by sample PAs ...... 85 Table 29: The average income poverty gap between irrigating and non-irrigating households ...... 85 Table 30: Parameter estimates of a logit model for determinants of a household poverty ...... 87

x ABBREVIATIONS AND ACRONYMS

ADLI Agricultural Development Led Industrialization ACSI Amhara credit and saving institution AE Adult equivalent BoARD Bureau of Agriculture and rural development CSA Central Statistics Agency ETB Ethiopian Birr

HA Hectare IPMS Improving Productivity and Market Success LDC Less developed countries MOARD Ministry of Agriculture and Rural Development MOFED Ministry of Finance and Economic Development PA Administrative unit in a district PG Poverty Gap

TLU Total livestock unit WAE Water Aid Ethiopia

xi CHAPTER ONE

1 INTRODUCTION

Poverty alleviation1 has been largely a result of economic growth (Roemer and Gugerty 1997). Because Ethiopia is an agrarian country, agriculture is the leading sector as source of income, employment and foreign exchange and national economic growth is determined by the performance of agriculture. Irrigation plays the key role in the performance of agriculture, which increases income growth. Income growth is essential for economic growth (Hussain and Biltonen 2001). Developing countries that ensure sustainable economic growth can be able to reduce their poverty levels, building up their democratic and political stability. They also improve the quality of natural environment and even reduce their incidence of crime and violence (Loayza and Soto 2002). To understand the role of irrigation in income growth and poverty alleviation, it is useful to review the fundamental sources of economic growth. According to (Maddison 1970) there are three major sources of economic growth. The first is an increase in the amounts of inputs used in production. Additional inputs can move a country out on its aggregate production function to a higher isoquant and higher levels of output. The three major inputs in the development process are population growth (which affects labor availability and labor), natural resource availability (which affects the cost of environmental factors such as land with its associated soils, water, and forest), and capital accumulation (which affects the availability of man-made inputs).

1 Poverty alleviation is the ability to produce goods and services above a minimum level of income needed to maintain the basic needs.

1 These sources of growth cause movement along a given multifactor production function. The second source of growth is a change the way in which a country uses its factors of production, increasing the amount of output produced by these inputs. These outputs increase can result from better organization of production or from shifts in the production function. For example, a new technology can shift the total production curve upward so more output is produced per unit of input. Increases in scale or specialization, increases in efficiency, or technological change are examples. In many cases, market conditions (relative prices) can change, in turn stimulating changes in these factors. The third source is increased human capital as embodied in people (e.g., improved education and health) and improvements in social institutions. Human capital can make labor more productive, contributing to technological progress and increase efficiency (especially when technologies and markets are rapidly changing). Agriculture contributes substantially to the economic growth of many low- income countries. It is often the leading sector of the economy as source of income, employment and foreign exchange. Agriculture employs more than 70 percent and contributes 30 to 60 percent of the gross domestic product (GDP). More than half of the less developed countries population gets their food from own-production. Agriculture output also is used as an input for industries so it can stimulate the growth of industrialization. Improving agricultural productivity thus has contributes to income growth (UNDP 2007). Ethiopia ranks 170 out of 177 the poorest countries on the Human

Development Index (UNDP 2006). Its GDP per capita was $ 350 in 2010 compared to $ 809 for Kenya and $ 1,705 for Sudan (IMF 2011). Half of Ethiopia‟s GDP depends on agricultural activity. Thus, the economy of Ethiopia is largely dependent on agriculture, and about 85% of the population is engaged in it. The dependency on rainfed agriculture coupled with the erratic nature of rainfall is the major factors

2 blamed for the poor performance of the agricultural sector and main cause of widespread food insecurity in the country (FAO 1994). Irrigation has served as one key driver behind growth in agricultural productivity, increasing household income and alleviation of rural poverty, which highlights the various ways that irrigation could have an impact on poverty. According to Lipton et al. (2004) cited by Haile (2008), there are four interrelated mechanisms by which irrigated agriculture can reduce poverty, through: (i) increasing production and income, and reduction of food prices, that helps very poor households meet the basic needs and associated with improvements in household overall economic welfare, (ii) protecting against risks of crop loss due to erratic, unreliable or insufficient rainwater supplies, (iii) promoting greater use of yield enhancing farm inputs and (iv) creation of additional employment, which together enables people to move out of the poverty cycle. In the same way, Zhou et al. (2008) mentioned that irrigation contributes to agricultural production in two ways: increasing crop yields, and enabling farmers to increase cropping intensity and switch to high-value crops. Therefore, irrigation can be an indispensable technological intervention to increase household income. This study will examine the impacts of irrigation on incomes at the household level for one region of Ethiopia. 1.1 Background and justification

Irrigation use in Ethiopia dates back several centuries, and continues to be an integral part of Ethiopian agriculture. In Ethiopia, modern irrigation began in the

1950s through private and government owned schemes in the middle Awash valley where big sugar, fruit and cotton state farms are found (FAO 1997). The main purpose of irrigation development in the 1960s was to provide industrial crops to the growing agro-industries in the country. The agro-industries

3 were established by foreign investors and had the objective of increasing export earnings. During the 1960s, irrigation was seen as part of the modernization of the country's agricultural economy. It was considered as an important investment for improving rural income through the increased agricultural production. But, in 1975 the rural land proclamation was introduced in the country. Following the rural land proclamation, the irrigated private farms were nationalized and converted to state farms by the Derg regime.

By early 1985 in Ethiopia, some 7.7 million people were suffering from drought and food shortages. More than 300,000 died in 1984 alone, more than twice the number that died in the drought a decade before. Before the worst was over, 1 million Ethiopians had died from drought and famine in the 1980s. The recurring cycle of drought produce the need for small-scale-irrigation development expansion to other parts of the country to address drought and food shortages, and the need for more food for the internal market.

Agricultural growth is not produced by passive policies. There is no unique policy prescription that fits the diversity of the agricultural sector in the less developed countries. Enhancing productivity is a common essential requirement. The increase in productivity will determine by the appropriate policy mix. The major lesson that emerges from country experiences is that for agricultural growth to occur, a number of factors need to be addressed in the rural sector such as infrastructure, social services, technology, marketing infrastructure, and seasonal credit availability, along with the building of an appropriate institutional environment (UNDP 2007). The current government has undertaken various activities to expand irrigation in the country. The country‟s Agricultural Development Led Industrialization (ADLI) strategy considers irrigation development as a key input for sustainable development.

4 Thus, irrigation development, particularly small-scale irrigation is planned to be accelerated (MOFED 2010). Ethiopia is believed to have the potential of 5.1 million hectares of land that can be developed for irrigation through pump, gravity, pressure, underground water, water harvesting and other mechanisms (MOFED 2010). According to BOARD (2010) and Awulachew et al. (2005) the total irrigated land in the Amhara region was 347,725 hectares. There are 310 modern irrigation schemes developed in this region.

The irrigation schemes developed have covered an irrigated area of 8,469.2 hectares with 17,443 beneficiaries. Out of these total irrigated areas 5,718.68 hectares is from small-scale and 2,750.58 hectares from medium-scale irrigation schemes. The study area, Fogera District, is one of the eight Districts bordering Lake Tana, source of Blue Nile. This District has an estimated 23,354 hectares of water bodies. The District is endowed with beautiful and diverse natural resources, with capacity to grow diverse annual crops. The altitude ranges from 1774 to 2410 masl. The mean annual rainfall is 1215 mm and ranges from 1100 to 1340 mm (MOA 2005). Therefore, the district has a great potential for small-scale irrigation. The objective of this study is to evaluate the impact of selected small-scale irrigation schemes on household gross income and on poverty reduction in Fogera district. 1.2 Statement of the problem

Agricultural production in Ethiopia is primarily rainfed, so it depends on erratic and often insufficient rainfall. As a result, there are frequent failures of agricultural production. Irrigation has the potential to stabilize agricultural production and mitigate the negative impacts of variable or insufficient rainfall. Irrigation development also can help offset some of the negative effects of rapid population growth (2.6% per year in Ethiopia; CSA 2007). Population growth

5 causes agricultural activities expands into marginal land, which leads to forest, land and water degradation. This environmental degradation can reduce agricultural productivity, which in turn worsens food insecurity and poverty. In order to respond to growing food demand, food production should increase. The three methods to increase food production are: increasing agricultural yield, increasing the area of arable land, and increasing cropping intensity (number of crops per year). Irrigation has the potential to increase both yields and cropping intensity in Ethiopia (Awulachew et al.

2010). Irrigation increases agricultural productivity and farm income per ha, according to previous studies (Nhundu et al., 2010; Gebremedhin and Peden 2002; Hussain 2006). It insulates the national agricultural economic sector against weather- related shocks and provides a more stable basis for economic growth and poverty reduction. It supports the process of transforming traditional subsistence agriculture in to market-oriented production of high value crops (Asfaw 2007).

The development of water resources for agricultural purposes (irrigation) is rising rapidly. According to BCEOM (1998) and Tilahun & Paulos (2004) as cited in Awulachew et al. (2010), in 1990 Ethiopia had an estimated a total of 161,000 hectares of irrigated agriculture, of which 64,000 ha were in small-scale schemes, 97,000 ha were in medium-and large-scale schemes and approximately 38,000 ha were under implementation. This had grown to more than 247,000 ha by 2004, with traditional irrigation schemes alone covering more than 138,000 ha. Currently, the

Ethiopian government gives more emphasis to small-scale irrigation as a means of achieving food self-sufficiency (MOFED 2010). Fogera District is an irrigation potential area, with an estimated 23,483 hectares of water bodies (MOA 2005). However, the living standard of the community

6 is subsistence2. Sustainable economic development will be supported by effective agricultural technology intervention. Equal and fair technology distribution within the community is valuable for balanced economic growth (Kobets 2004). This study will assess the impact of small-scale irrigation on the household gross income and poverty reduction at the household level.

1.3 The Goal of the research

The goal of this research is to evaluate the economic impact of selected small- scale irrigation on income and poverty reduction at household level. It compares households with and without access to small-scale irrigation systems. It also compares households who use four different small-scale irrigation schemes in the five villages of the district.

1.4 The specific objectives

The specific objectives of this project are as follows: 1. To identify the major field crops and vegetables grown using small-scale irrigation in the study area; 2. To compare the relative advantages of the various types of small-scale irrigation system; 3. To examine the major constraints encountered in the use of the small-scale irrigation systems;

2 Subsistence means the production of goods and services is not beyond maximum level of income requisite to maintain the basic needs of household consumption. People need aid if one agricultural season fails due to lack of rain or other adverse events.

7 4. To examine the effects of small-scale irrigation on the gross income at household level; 5. To determine the difference in prevalence of poverty between small-scale irrigating and non-irrigating households; 6. To apply the study findings to make recommendation and policy implementation of small-scale irrigation systems. 1.5 Hypotheses

The hypotheses of this research are: 1. Small-scale irrigation has a positive impact on household gross income, cropping income and livestock income but has a negative impact on non-farm incomes 2. Small-scale irrigation has a negative impact on poverty. The probability of being poor is lower among users of small-scale irrigation compared to non- users in the agricultural sector.

3. Irrigation users have more agricultural productive assets and non-agricultural asset holdings.

8 CHAPTER TWO

2 REVIEW OF RELATED LITRATURE 2.1 Poverty

Poverty definitions and measurement have important implications for targeting and policy. The concept of poverty goes back to the 16th and 17th centuries. Before 1750, there were four approaches to poverty: acceptance (resignation), charity, precarious rescue and theft depending on which side of the fence the observer stood. Since the era of mercantilism, the fight against poverty has been marked and consequently, it was with the advent of the mercantile economy and the urbanization and monetarization of society that the poor had been defined in terms of lacking what the rich had, David (1994) cited by Abraham (2006). Poverty is a highly complex problem. It has multiple causes and manifestations. According to Townsend (1993) poverty is defined as absence or inadequacy of diets, amenities, standards and services that allow people to follow the customary behaviors expected of them by virtue of their membership of society. Poverty exists for people whose resources are seriously lacking when compared to the resources available by the average individual or family (Haile 2008). Poor people are those who are excluded from ordinary living patterns, customs and activities. Some defined poverty in general terms as inability to maintain a minimal standard of living. Others also have defined the poor as those who do not have adequate resources to meet their basic needs. The United Nations (UN 1995) defined absolute poverty as “a condition characterized by severe deprivation of basic human needs, including food, safe drinking water, sanitation facilities, health, shelter, education and information. It depends not only on income but also on access to services”.

9 McClelland (2000) indicated that “Poverty is where people have unreasonably low living standards compared with others; cannot afford to buy necessities, such as a refrigerator for example; and experience real deprivation and hardship in everyday life”. Empirical evidence shows that poverty rates vary when different concepts and measures are used. There are two types of poverty, absolute and relative poverty: absolute poverty is defined as the minimum amounts of essential goods and services a household needed to survive. It is estimated based on the income needed to purchase these subsistence amounts. Relative poverty is the households‟ standard of living falls seriously below what is believed normal for the society in which they live (Ravallion 1998) cited by Haile (2008). 2.2 Water and agriculture

Water, soil, air and sunshine are the four main determinants for plant growth. Therefore, water is essential to plant-growth and crop-production (Widtose 2001). All sectors depend on water. Water is important for agriculture, household consumption, industry, hydropower, navigation, fisheries, recreation, and ecosystems. Without water there is no food production. When there is adequate supply of water, crops grow best and produce most. Water is a basic need for human beings and animals. It is essential for their metabolic processes. Livestock water requirements are mainly provided by direct water intake and partly by the moisture content of their forage. Livestock production requires large quantities of forage. The production of forage requires substantial amounts of water. Therefore, water is vital for all agriculture types. According to Dupriez and De Leener (2002), the sources of water for crop production are rainfall and irrigation water. The two types of agriculture seen from the perspective of water management are:

10 Rainfed cultivation is agricultural production of crop depending entirely on the rain. It relies on the rainfall timing and distribution. Rainfed farming is characterized by plateau cultivation and dry land cropping. Rainfed farming is mostly practiced during one growing season, unimodal, but in some areas two growing seasons (bimodal production) are possible. Irrigated cultivation is agricultural production using irrigation water in addition to rainfall. Irrigated crops benefit from man-made watering with the help of water pipes, canals, reservoirs and pumps. The source of irrigation water is surface water or groundwater. Surface water is obtained in ponds, lakes, rivers and seas whereas groundwater is obtained underground in liquid or vapor state (Dupriez and De Leener 2002). 2.3 Irrigation development

Irrigation is generally defined as the application of water to the land for the purpose of supplying moisture essential to plant growth. It is an age-old art. Irrigation was practiced for thousands of years in the Nile Valley. Egypt claims to have the world's oldest dam built about 5000 years ago to supply drinking water and for irrigation. At that time basin irrigation was introduced and still plays a significant role in Egyptian agriculture. According to Zewdie et al. (2007) irrigation has been practiced in Egypt, China, India and other parts of Asia for a long period of time. India and Far East have grown rice using irrigation nearly for 5000 years. The Nile valley in Egypt, the plain of Euphrates and Tigris in Iraq were under irrigation for 4000 years.

Irrigation is the foundation of civilization in numerous regions. Egyptians have depended on Nile‟s flooding for irrigation continuously for a long period of time on a large scale. The land between Euphrates and Tigris, Mesopotamia, was the

11 breadbasket for the Sumerian Empire. The civilization developed from centrally controlled irrigation system (Schilfgaard 1994). Evidence also shows that irrigation in China was begun about 4000 years ago. There were reservoirs in Sri Lanka more than 2000 years old. As far back as 2300 BC, the Babylonian Code of Khammurabi provided that 'If anyone opens his irrigation canals to let in water, but is careless and the water floods the fields of his neighbor, he shall measure out grain to the latter in proportion to the yield of the neighboring field.'

Other indicator for irrigation development is found in the stony-gravel limestone desert of the Negev area in Israel. Remnants of these ancient irrigation systems date back from the Israelite period (about 1000 BC) and from the Nabattean- Roman- Byzantine era (300 BC to 600 AD). In the absence of permanent water sources, the ancient farmers developed 'runoff' farm systems that used sporadic flash floods for irrigating (Shanan 1987). Ethiopia has a long history of traditional irrigation systems. Simple river diversion still is the dominant irrigation system in Ethiopia. According to Gebremedhin and Peden (2002), the country‟s irrigation potential ranges from 1.0 to 3.5 million hectares but the recent studies indicate that the irrigation potential of the country is higher. According to Tilahun and Paulos (2004) as cited by Awulachew et al. (2010), estimates of the irrigation potential of Ethiopia may be as large as 4.3 million hectares. Traditional irrigation schemes cover more than 138,000 hectares whereas modern small-scale irrigation covers about 48,000 hectares. The total current irrigation covers only about 6% of the estimated potential land area. According to the MOA (2005) and Awulachew et al. (2007), Amhara region has 770,000 hectares of irrigation potential. Different development activities have been underway to utilize these resources. Currently, there are 310 irrigation schemes operating in the Amhara region. The irrigation schemes developed cover an irrigated

12 area of 8,469 hectares with 17,443 beneficiaries. Of these total irrigated areas, 5,719 hectares are from small-scale and 2,751 are from medium-scale irrigation schemes. 2.4 Irrigation methods

Irrigation methods are the systems how to obtain water for irrigation purposes from its sources. According to Dupriez and De Leener (2002), irrigation methods depend on several factors such as topography, water resources, the plants cultivated, the land tenure systems, the growing seasons and the rain and water regimes.

2.4.1 Surface irrigation

There are only two general methods of applying irrigation water. The first is surface irrigation. Surface, irrigation means above the ground, and is the method generally adopted in all countries. There is a great variety of methods of surface irrigation, most of which do not merit serious consideration, because they either fail to recognize the natural laws underlying irrigation, or their cost of installation is unaffordable in the current context. The second is sub-surface irrigation, the application of irrigation water from below. Sub-surface irrigation has the advantage that water so applied is not subject to such direct evaporation from the surface as of necessity accompanies surface irrigation. According to Widtose (2001), surface irrigation methods are furrow irrigation, flood irrigation basin irrigation and boarder irrigation. The choice and adoption of these irrigation methods are depending on the nature of the soil, the contour of the land, the head of the water stream, the quantity of water available and the nature of the crop.

13 2.4.1.1 Basin irrigation

A basin is a piece of land, small or large, surrounded by earth bunds in which water is ponded. The water can be impounded within it to irrigate trees, vegetables or crops grown in patches. The field is divided in to compartments or checks wholly surrounded by levees. The water is contained at the upper end and completely fills the compartments until it over flows at the lowest point of the levees.

2.4.1.2 Furrow irrigation

In this method, the water is guided in the furrow or channels that pass through the whole field, but the water covers only part of the soil surface, so it results in less evaporation. The furrows are separated with ridges. At each ridge, water is conveyed into furrows that can be perceived as narrow basins or borders. Furrowing is applied on steep slopes.

2.4.2 Flood irrigation

In flood irrigation, all of the soil is covered by the water applied. It is the least controlled of all surface irrigation techniques. Water is conveyed in a ditch at the upper part of plot and allowed to spread over the land in a manner directed by the natural landscape. Flooding is best applied when the slope is limited.

2.4.3 Border irrigation

The border method of irrigation is an open-field method. Here the land is divided in to elongated plots confined between low earth banks and configured to slope uniformly from the point of supply. The land surface should slope gently in the direction of flow

14 and it is generally leveled laterally, along all cross sections perpendicular to that direction. Water is guided over the land by field ditches. 2.5 Sprinkler irrigation

According to Dupriez and De Leener (2002), Sprinkler irrigation imitates rainfall. It is also called overhead irrigation. The water is broken up in to fine droplets and falls on the ground or the vegetation. It is the application and distribution of water over the field in the form of a spray, or a jet which breaks in to drops or droplets, created by expelling water under pressure from an orifice. In contrast to surface irrigation, sprinkler systems are designed to deliver water to the field without depending on the soil surface for water conveyance or distribution. To prevent pondings and surface runoff, sprinklers are designed and arranged to apply water at a rate that does not exceed the soil‟s infiltration. Water application efficiency under sprinkling irrigation is strongly affected by wind, especially during daytime when the air is warm and dry, and if the droplets are small and the application rate is low. 2.6 Drip irrigation

The principle of drip irrigation is to wet dry ground with small amounts of water just where the plants can absorb it. Drip irrigation is practiced in dry, arid regions where water is scarce and must be used sparingly. Water is delivered to the points via a set of plastic lateral tubes laid along the ground or buried at a depth of 15-

30 cm and supplied from a field main. These tubes are left in place throughout the irrigation season. Drip irrigation can save water by reducing the portion of the soil surface that is wetted thus, decreasing the amount of direct evaporation.

15 2.7 Irrigation-poverty linkages

Some literature argues that irrigation agriculture causes water logging that create favorable condition for the multiplication of disease causing agents such as malaria, Schistosomiasis and the like. The other environmental problems with irrigation are land degradation and salinity. On the other hand, there is much literatures that shows irrigation is a major driving factor of the increase in rural household income through agricultural growth. These studies strongly argue that irrigation expansion the main policy intervention to alleviate rural poverty. According to Lipton et al. (2004) as cited by Haile (2008) the four main inter-related mechanisms to reduce poverty are: 1. Irrigation increase agricultural production and income, for households with access. These outcomes are observed despite the price decrease that can occur as supply increases (other factors held constant). The rice decrease can allow poorer households to more easily meet their basic needs. Household level

economic welfare is improved for the poor. 2. Irrigation protects from the risk of crop loss due to erratic, unreliable or insufficient rainwater supplies. 3. Irrigation enhances the use yield-enhancing farm inputs. The uses of such farm inputs improve the agricultural production and income. 4. Irrigation creates additional employment. Household and/or laborers are engaged in the irrigation farming that helps to increase the labor productivity

during the dry periods, farm off-season. Water is a valuable input for agriculture. Irrigation water appears to provide many pathways for poverty alleviation. The access to consistent irrigation water can enable farmers to adopt irrigation technologies. Irrigation facilitates the intensity of cultivation that leads to an increase in agricultural productivity and greater returns

16 from farming. The expansion of irrigation opens up new employment opportunities in the household that increase the efficiency of labor and land. This improves farm income, livelihood, and the quality of life in rural areas (Hussain and Hanjra 2004).

17 CHAPTER THREE

3 THE STUDY AREA AND SMALL-SCALE IRRIGATION TYPES 3.1 Description of the study area

Ethiopia is situated in the East Africa and lies between 3°30´ and 14°50´ North latitudes and 32°42´ and 48°12´ East longitudes. It has a surface area of about 1.127

Million km2, of which 1,119,683 km2 is and 7,444 km2 is water. The country has a land boundary length of 5,311km. Ethiopia has special features because of its topography, geology and climate (Awulachew et al. 2001). Ethiopia is a landlocked country consisting of nine independent regions and two city councils divided along ethnic lines. It occupies an area of 1.14 million square kilometers. The country shares its international borders with five African countries: Eritrea in the North, Djibouti and Somalia in the East, Kenya in the South and Sudan in the west. Ethiopia is one of the poorest countries in the world with a population of 83 million in 2008 being the second most populous in Africa next to Nigeria. The nation‟s economy is mainly dependent on rainfed agriculture, which accounts for half the GDP, 60% of exports and 80% of employment (WAE 2008). Ethiopia has nine National Regional States and two Special City Administrations: Addis Ababa and Dire Dawa. Amhara Region is one of the regional states of the Federal Republic of

Ethiopia. Amhara region has a geographical area of about 153,000 Km2. Ethiopia‟s largest inland body of water, Lake Tana, as well as the Semien Mountains National Park, which includes the highest point in Ethiopia; Ras Dashan is located in Amhara region.

18 According to the CSA (2007), Amhara National Regional State has a total population of 17,214,056, with 8,636,875 men and 8,577,181 women. Only 12.2 percent of the population lives in urban areas. The region covers a total land area of 159,173.66 square kilometers and the population density is 108.15 people per square kilometer. The region has 3,953,115 households. The average family size in urban and rural areas is 3.3 and 4.5 persons, respectively. The study was conducted in Fogera District, which is one of the 106 Districts of the Amhara Regional State and found in South Gondar Zone. Fogera is one of the eight districts bordering Lake Tana, source of Blue Nile. It is situated at 110 58 N latitude and 370 41 E longitude. Woreta is the capital of the District. It is found 625 km from Addis Ababa and 55 km from the Regional capital, Bahir Dar.

Figure 1: Location of Fogera District

Fogera district is bordered by Libo Kemkem district in the North, Dera district in the South, Lake Tana in the West and Farta district in the East. The district is

19 divided into 28 rural and 5 Urban PAs. The district has a total land area of 117,414 hectares. The land use pattern of the district includes 51,662 hectares of cultivated land, 25,831 hectares of pasture land, and 16,434 ha for other purposes, and the water bodies‟ account for 23,483 hectares (MOA 2005). IPMS (2005) indicate that flat land accounts for 76%, mountain and hills 11% and valley bottom 13%. The high proportion of plain topography creates the opportunity for irrigation. However, water logging is a common phenomenon in the plain areas of the district. The average land holding per household is about 1.4 hectare with a minimum and maximum hectare of 0.5 and 3.0 hectares, respectively. The altitude ranges from 1774 to 2410 masl. The mean annual rainfall is 1215 mm and ranges from 1100 to 1340 mm (MOA 2005). The 28 rural PAs have a total population of 212,204. Werota, the capital of the district, has one rural PA known as Werota Zuria. The town has 40,404 inhabitants. The number of agricultural households in the district is 44,168 (MOA 2010).

3.1.1 The characteristics of the sample PAs

Five PAs in Fogera district were chosen for the purposes of this study, and it is appropriate to discuss some of their basic characteristics in this section. The sample PAs has some common characteristics. The agro-climatic ecology of the five PAs is similar. In each PA, the belg and meher are two cropping seasons. The belg cropping season is a very short rainy period whereas meher season is the long rainy period. Farmers depend on meher season for rainfed crop production. The onset, duration and quantity of the rainfall are variable. Agriculture is the major occupation of the people in the PAs. The agriculture in all PAs is a mixed crop–livestock farming system. Crop production is rainfed during the rainy season, supplemented for some households by small-scale irrigation in the dry season. The dominant crops grown in the study area are rice, teff (Eragrostis), wheat, barley, maize, beans, peas, chickpeas, lentils,

20 fenugreek and noug. Commonly produced vegetables are onion, tomato, potato, pepper and cabbage.

3.1.1.1 Kuhir Michael

The PA is located 17 Kilometer south-east of Werota town and 39 Kilometer North of Bahir Dar. The main Asphalt road pass from Addis Ababa to Gonder passes through this PA. Gumara town, which is known as an onion and tomato market, is found in this PA. According to Fogera district office of Agriculture 2010 report, the number population and households in the PA were estimated 6,068 and 1,411, respectively. The landscape of the PA is characterized with both plain and upland. Vertisol is the main soil type in the area. Kuhir Michael is endowed with perennial rivers and streams. Gumara River, one of the largest tributaries of Lake Tana, passes through this PA. Farmers use the Gumara River for all of their water need. Gumara River is used for drinking purpose. Both people and livestock use it for drinking, washing and other activities. Irrigation is the other important use of the Gumara River. Motorized pumps are widely used to draw the water from the river for irrigation. Farmers who cannot afford to buy pumps sometimes use those of other farmers by renting in daily and (or) hourly basis. The other river that is widely used for irrigation by the farmers in this PA is the Guanta River. The Guanta is diverted by a cement concrete channel for irrigation purposes. Traditional river diversion from rivers and well water using pedal pumps are also practiced in this PA.

3.1.1.2 Shina

Shina is located 13 km south-west of Werota town. It is found towards Lake Tana. According to Fogera district office of Agriculture 2010, the number of

21 population and households in the PA were estimated 10,052 and 2011, respectively. The average farm size is estimated to be 1.3 hectare per household. The landscape of the PA is plain. The soil type is black at around 7 meters depth but the upper layer is silt. The total surface area of the PA is 3,400 hectares. About 2,071 hectares are used for cultivation, 664 hectares are used for communal grazing, 244 for private grazing land, 83 hectares are covered by eucalyptus tree and 334 hectares are used for residential and governmental institutions. Motor and treadle water pumps are the common irrigation systems. Traditional river diversion and shallow wells are also used for irrigation, drinking, washing and other household consumption in the PA. The two major rivers, Gumara and Reb that have great economic importance to the district traverse this PA. These rivers are mainly used in the PA for irrigation during the dry season using water pumps. The small-scale irrigation methods used in the PA are motorized pumps, treadle pumps, traditional river diversion, and wells.

3.1.1.3 Abana Kokit

Abana Kokit is situated thirteen kilometers north of Werota Town. The main asphalt road from Addis Ababa to Gonder crosses the PA. The topography of the PA is characterized as both plain and upland type. The Fogera district office of agriculture (2010) indicates that the total household and population of the PA are 921 and 3,496. The Reb river passes through the PA. The river is used for irrigation purpose using motorized pumps. There are other small rivers that are used for traditional river diversion.

3.1.1.4 Bebekis

Bebeks is one of the rural PAs in the Fogera district. It is located 19 kilometers southeast of Werota town. According to Fogera district office of agriculture (2010) the

22 number population and households in the PA were estimated 8,595 and 1,465, respectively. In the PA, there is a concrete diversion from three perennial springs used for small-scale irrigation. This is locally known as “Timeket Bahir” which means lake used for Epiphany. The spring diversion is geographically located at 11047′32″ N latitude and 37039′40″ E longitude. The scheme is used to irrigate 120 hectares of land, with 496 beneficiary households. The irrigation water is used for cropping, drinking, washing and other household consumption activities. Gumara River passes through this PA. Traditional river diversion and water pumps are the other commonly used irrigation practices in this PA.

3.1.1.5 Werota Zuria

Werota Zuria is the rural part of Werota town. According to the Werota town administrative agricultural office (2010) Werota Zuria has a total population of 6081. There are 1,415 household involved in crop and livestock farming in Werota Zuria

PA. In this PA, there is a river diversion dam on the Eriza river. The dam was constructed in 2003 by IFAD with 104 beneficiary households. The irrigation potential of the dam is 26 hectares. 3.2 Small-scale irrigation types

3.2.1 Concrete canal river diversion

River diversion irrigation systems are practiced in the four Sample PAs such as

Quahir Michael, Shina, Bebeks and Werota Zuria. Guanta River is the main source of water for the modern irrigation3 system in Quahir Michael and Shina PAs.

3 Modern irrigation locally means type of irrigation in which water is diverted from river or spring development through concrete canal.

23 Figure 2: Guanta river dam

The river is diverted by cement concrete canal for irrigation purposes. According to BOWR (2005), it was constructed in 2003 by the Amhara Regional state water works. The diversion site is located at 11045′22″ N latitude 37047′ 17″ E longitude. The canal river diversion water is used for irrigation, for drinking livestock and other household consumption. The command area of the scheme is 46 hectares of land and serves 91 households of which 85 are male and 6 female. Irrigation farming has an advantage over livestock farming. According to Fogera district office of agriculture (2010), 94% households have both crop and livestock. The borders around the irrigation canal provide good grazing for livestock (picture 4).

24 Figure 3: Irrigation and livestock

The Guanta river diversion and Tanqua Gabriel spring development together are used for irrigation by both Kuhir Michael PA and Shina PA. Kuhir Michael is upstream of Shina. Both PAs have the same water use association. Irrigation occurs by both gravity for lands at lower elevation and by a motor pump for land above the canal. The amount of water in the irrigation canal is sufficient for both gravity and motor pump irrigation and usually there is excess of water flowing back to the Gumara

River.

25 Figure 4: Motor pump irrigation from river diversion canal

In addition, the river diversion water is used for household consumption such as drinking, washing and the like. The picture below shows the woman and the two children fetching water for household consumption. The other modern irrigation scheme is Eriza river diversion in Werota Zuria (Figure 6). The river diversion dam was constructed by Co-SIRARE in 2003.The irrigation potential of the scheme is 26 ha.

Figure 5: A women and Children fetch water from river diversion for household consumption.

Figure 6: The river diversion dam on Eriza River in Werota Zuria

The amount of water is poor compared with Guanta and other spring water sources. The amounts of land irrigated are comparatively low. The area is upland and dry.

27 3.2.2 Spring development small-scale irrigation scheme

The two main spring development irrigation areas are Tanqua Gabriel and Timeket Bahir in Quhir Michael and Bebekis PAs, respectively (Figure 7). The Tanqua Gabriel spring development is constructed by ORDA and EMATLFA in 2001.The irrigation potential of the spring is 70 hectares of land for 351 beneficiary households

Figure 7: Tanqua Gabriel spring development

The second spring development irrigation scheme is the “Timeket Bahir” which means water used for Epiphany. The spring is used for religious purpose with

Ethiopian Orthodox Church. The scheme is developed from three perennial springs.

28 3.2.3 Motorized pump

Motorized pumps are widely used irrigation systems in the five PAs (Figure 8). In Shina, for instance, there are 54 total pumps. Most of the farmers bought them as part of a group. Other households gain access to the pumps through renting from the owners. The Gumara, Rib and Eriza rivers are used for irrigation by using motor pumps.

Figure 8: Motor pump irrigation using Eriza River near Werota town

3.2.4 Pedal pump

In the five PA, 295 households received pedal pumps from office of agriculture and rural development on credit basis in 2006, 2007 and 2008. The amount of credit required for each treadle pump was 350 ETB. Although the treadle pumps are highly demanded by the farmers, there is an insufficient supply. These pedal pumps were imported from India (Figure 9). They are lightweight and easy to operate. One key problem is limited access to spare parts for repairs in the case of malfunction. In

29 addition, the suction and delivery hoses are easily damaged. Loss of integrity of the joints around the foot press is another problem. At present, most of these treadle pumps are non-functional.

Figure 9: Pedal pump

The Indian pedal pumps are regarded as superior to those made in Ethiopia due to better pumping power. With Indian pumps, the water can be pumped long distances through its long delivery hose. The major water source for pedal pump is shallow wells. Shallow wells are widely used water sources for irrigation and other household consumption (Figure 10).

There are 2378 shallow wells in Shina PA alone. There are households who have up to eight wells. Most of the household‟s water consumption such as drinking, washing and irrigation are obtained from shallow well. The depth of the wells is estimated in the range of one to seven meters.

30 Figure 10: Sample wells constructed for only one cropping season

Very shallow wells that are one to three meters depth are dry during very dry periods such as February, March and April. Wells that have a depth from four to seven meters are perennial water sources. These shallow wells are extensively used by farmers to irrigate small plots of land, by fetching water using buckets and jars. The shallow wells are used for only one cropping season because during rainy season the wells are damaged because the area is flood prone and dominated by Vertisols.

Figure 11: Sample of Vertisol

31 As a result, a common practice is for farmers to construct new wells each cropping season. There are traditionally used mechanisms to avoid shallow well failure. Some farmers use car tires or cylinders made from cement to construct the well.

Figure 12: Well constructed from tire materials

This saves the wells from frequent failures. These shallow wells are constructed by farmers themselves. The well water is used through jar and bucket irrigation (Figure 12) in addition to the treadle pumps. In general, all family members of the household are involved in jar and bucket fetching irrigation from well water, but females are the main actors. Some households use two or more shallow wells in one irrigated farm plot. The wells were constructed in the boarders of the irrigable lands for ease of watering the irrigated crops (Figure

13).

32 Figure 13: Irrigation by fetching water from wells

Figure 14: Sample well constructed by cement cylinder

The commonly grown crops by fetching water from shallow well are tomato, onion, cabbage, and lettuce.

33 3.2.5 Traditional river diversion

Traditional river diversion is the dominant method used by farmers in all five PAs. This irrigation system is simple for farmers to practice by inheriting the knowledge from grandparents but the amount of water and seasonality of rivers are major problems. Many farmers use traditional irrigation to complement other irrigation systems like modern river diversion and motor pump irrigations.

34 CHAPTER FOUR

4 MATERIALS AND METHODS 4.1 Research methods

4.1.1 Approach for data collection, entry and checking

Household data collection was undertaken in five villages that have irrigation and non-irrigation water users. Data collection methods included a survey, semi- structured interviews and focus group discussions. Data were collected at household and community level with the assistance of development agents. Each PA has three developmental agents who live and work with the farmers. Using development agents as assistance for data collection is important for the reliability of the data because farmers are more likely to report accurate information to development agents, especially on income, land size and other taxable assets. The sample households were selected by utilizing the following three-stage stratified sampling procedure. The first stage involved consultation with Fogera District Agricultural and Rural Development offices, and five PAs were selected purposively on the basis of their similarity in agricultural practices, surface water resource potential for irrigation, and the type of small-scale irrigation they used. In the second stage, household lists in the selected PAs were obtained from village administration and Development agents‟ office. The different types of small-scale irrigating and non-irrigating households were selected from this list. In one of the five

PAs, Shina, all households are irrigation users in one way or another. No sample was taken for non-irrigation user households in this PA. Proportional sampling method was used to develop the sample.

35 In the final stage, households were listed by each small-scale irrigation category then the random sampling technique was used to select sample households from each household type using a random number table. The objective was to carefully examine and compare the income and poverty level of small-scale irrigation users and non-users. Based on this multi-stage sampling process, the total sample households were selected on a random sampling basis from five villages in Fogera District. Pervious literature showed that an appropriate sample size is determined by number of factors. As the number of factors increase, the sample size should also increase to avoid biased results. According to Marks (1966), as cited in Green (1991) a minimum of 200 subjects should be used for any regression analysis, whereas Schmidt (1971) suggested a minimum subject-to-predictor ratio ranging in value from 15-to-1 to 25- to-1. According to Cohen (1988) if the research uses 15 predictors, the minimum sample size is 138 and if the research uses 20 predictors, the minimum sample size is

156. Rao and Richard (2006) suggest that an appropriate sample size depends on the type of problem investigated, required precision and to a certain extent, the resources available. Following these guidelines, a total sample size of 180 households was selected from the five villages with proportional samples size was taken from irrigating and non-irrigating households (Table 1). In addition to the structured survey, information was collected through focus group discussions with the Fogera district agricultural office socio-economic and irrigation experts, development agents and irrigating and non-irrigating farmers. Focus group discussions were first held with model4 farmers who use small-scale irrigation.

4 Model farmer, in the context of this study, means a farmer who adopts modern irrigation technology and earns a higher income from it.

36 A second discussion was with development agents. After thoroughly discussing the problems and opportunities of irrigation development with farmers and development agents, a focus group with Fogera district office of agriculture socio-economic experts and irrigation experts was undertaken.

Table 1: Summary of sample size by PA and irrigation types

Irrigating Non- Sample Concrete Traditional PA Motor Pedal irrigating per canal river river pump pump PA diversion diversion Kuhir Mikael 6 4 4 4 22 40 Shina 8 0 8 8 0 24 Kokit 0 4 4 8 22 38 Bebekis 6 8 4 0 24 42 Werota Zuria 4 6 4 0 22 36 Sample 24 22 24 20 90 180 /irrigation Another source of information for this study was key informants. The key informants were selected by asking farmers randomly “who is your model farmer in the PA with irrigation farming?” then the more frequently proposed farmers were selected for key informants. In addition to this, local leaders and extension workers are used in the selection of key informants. The same procedure was used in all PAs for key informant selection. The survey data were recorded and organized in a Microsoft Excel spreadsheet. SPSS 16 software was utilized for data entry and editing. A print-and- verify method of data entry checking was performed to avoid errors in the recording process. The missing values, zeros and not applicable values were identified for verification. Each variable was examined not only for outliers but also for the general acceptability of the figures compared to national and regional information from other sources. The inconsistent values were also checked with the questionnaire to identify

37 data entry errors. The database in SPSS then was converted to the STATA econometric software format and further data editing was undertaken. The household was used as the basic survey unit for the analysis. A household was defined as a number of people living and eating together in the same dwelling and share the same income. From November 2010 to February 2011, observation surveys and interview of households were conducted. The interview was conducted over 54 days. Household heads were involved in responding the interview because household heads are often make decisions concerning their households issue in the study area. Twenty-eight rural and five urban PAs exist in Fogera District. Five PAs were purposively selected on the basis of their similarity in agricultural practices, surface water potential for irrigation, and the type of small-scale irrigation they used. The different types of small-scale irrigating and non-irrigating households were collected from the respective PAs at development agents‟ office. Irrigating and non-irrigating households were proportional, 90 households from each category then total sample size was 180 households. The respondents were selected randomly using a random number table. To collect the data, the survey questions were carefully translated in to the local language (Amharic). This helped to convey the questions effectively to the rural interviewees. The data collection assistants were selected and trained for one and half days on administering and completing the questionnaire, and a pre-testing fieldwork was also organized for half a day. On the basis of the pre-test feedback, some adjustment was made on the questionnaire and the final questionnaire was organized. The questionnaires (see Appendix B) have six main parts. These are: 1. Household demographic characteristics; 2. Resource endowments of a household; 3. The crop production from rainfed agriculture;

38 4. Crop production with irrigation; 5. Livestock production; 6. Credit, input and extension service supports in production; 7. Non-farm and off-farm income; In order to characterize the selected small-scale irrigation systems, the major problems encountered in relation with irrigation systems, the reasons why non- irrigating households do not irrigate were developed using structured checklists. The survey data initially were recorded and organized in a Microsoft Excel spreadsheet. The data were then transformed to a database in SPSS (Version) for data entry and editing. A print-and-verify method of data entry checking was performed to avoid errors in the recording process. The missing values, zeros and not applicable values were identified for verification. Each variable was examined not only for outliers but also for the general acceptability of the figures compared to national and regional values. The inconsistent values were also checked with the questionnaire to identify data entry errors. The database in SPSS then was converted to STATA format for analysis.

4.1.2 Data analysis

4.1.2.1 Socio-economic and demographic characteristics of the sample households

To estimate the impact of small-scale irrigation on income improvement and poverty reduction, an assessment of selected socioeconomic and demographic characteristics of the sample household is vital. Impact assessment of any development intervention is methodologically difficult and complex task to undertake. Ravallion (2005) and Baker (2000) argued that no single method should dominate the impact evaluation of any development intervention but instead rigorous impact

39 evaluations should be open-minded in the choice of methodology. The most important thing to do in impact evaluations is to derive robust and meaningful close proxies or indicative estimates that are comparable between and within individuals or groups based on the aims of a particular development intervention. The effect of the irrigation on irrigation participating and non-participating households will be evaluated using descriptive statistics and econometric modeling that controls for other contributing factors. Households that did not use irrigation or those who were living adjacent to the irrigation users and the different small-scale irrigation scheme users will be used as the comparison group. To examine the impact of irrigation on the household poverty status, gross income was used as the main outcome of interest. In addition, various other socio-economic indicators were examined. The sample households‟ demographic characteristics indicate the various characteristics of, and differences among, the study groups. For the socio-economic analysis particular importance is given to age, sex, and education of household head, family size, land holding, access and sources of credit, extension and market services. These variables were analyzed using descriptive statistics such as average, percentage, minimum, maximum and frequency distributions.

4.1.2.2 Income evaluation

To evaluate impact of small-scale irrigation on annual gross income of the household, all sources of income such as agricultural (cropping and livestock) incomes and non-agricultural (off-farm and other) incomes should be considered. The examination of off-farm income in household gross income evaluation is valuable because of the working hypothesis that non-irrigating households have larger off-farm income than irrigating households. Non-irrigating households may use off-farm and

40 non-farm activities as compensation of irrigation. Therefore, considering of off-farm and non-farm income avoided the overstated income share from irrigation activities. To compare the annual mean income of the four different irrigation schemes, a single factor analysis of variance (one way ANOVA) was used. When the four irrigation schemes annual mean income comparison was computed, the ANOVA F (sometimes called the overall F or omnibus F) were statistically significant, that means there is a difference of variance (the assumption of equal variances has been violated). Thus, post hoc Multiple Comparison tests, commonly known as follow-up tests, were applied (Morgan et al. 2004). To control for other factors that influence household incomes this study uses an econometric modeling approach. As stated Nicholson et al. (2004) and Zhou et al. (2009), household gross income is a function of many determinants including household characteristics, asset holding, village location characteristics, and the prices of goods and services.

Mathematically, this can be written as: Y = f (X, D, Z, A, P) (1) Where Y is an endogenous variable (household gross income), X is household characteristics (education of household head, household family size, age of household head and gender of household head), D is Irrigation access, Z is village characteristics, A is the Value of household production assets, and P is the prices of agricultural products and inputs.

In this study, the principal dependent variable is annual household income (INCOMEHH), which includes agricultural income (cropping and livestock incomes) and off-farm income. The values of agricultural incomes are computed by multiplying the amount of each agricultural product (sold and consumed) with their annual average price. A principal objective of this research is to examine the impact of irrigation on

41 annual gross income of household. The income data were collected during November 2010 to February 2011. The independent variables were identified from previous studies and the nature of the study area. These variables are expected to result in (and therefore, explain) income variation across households in the study area (Table 2). The independent variables are as follows: • Access to irrigation (IRR): Irrigation supplements moisture, which enables

farmers to maximize agricultural production. It is assumed to have a direct relation with the total income of a household. Nhundu et al. (2010), Hussain and Biltonen, (2001) and Haile (2008) identified a strong positive relationship between access to irrigation and household income. Access to irrigation for household is a dummy variable, 1 if a household has access to irrigation and 0 otherwise. • Cultivated land size (LANDSZ): total cultivated land is the total sum of the

household‟s own and/or rented in/out from/to other households and measured in hectares. This does not include the grazing land and fallowing land. Farmland is the major input for agricultural production in rural households. Total cultivated land should have a positive relationship with income of a household (Kamara et al. 2001). • Family size of a household in adult equivalent (FAMSZADUL): family size in adult equivalent of a household is calculated by using the conversion factor in

(Appendix-Table A2). A household family size in adult equivalent is calculated by multiplying each household member with respective conversion factor and then summing. Its size depends on the age of each family members of a household. In rural households, family labor is the major input used in agricultural production. Households with large family size in adult equivalent

42 have more labor for agricultural production. Family size in adult equivalent is correlated positively is expected to positively affect total income and negatively affect the probability of poverty for a household. • Education (literacy) level of a household head (EDUHHL): Education has paramount impact on income improvement and poverty alleviation. It is likely that educated farmers would more readily adopt irrigation technologies and may be easier to train through extension support. The variable entered in the

model as dummy variable with 1 if a household head can read and write, and otherwise 0. • The number of livestock owned in TLU (LIVESTO): This is a continuous variable measured in terms of Tropical livestock unit (250 kg live weight) .The number of livestock owned by a household in TLU is calculated by conversion factor for Tropical Livestock Unit (TLU) (Appendix-Table A3). A household livestock size in TLU is calculated by multiplying the number of each type of

animal by an appropriate conversion factor and then summing. Households with higher livestock holding will lead to higher probability of getting excess livestock for selling and hence generating additional income, particularly the owner of more oxen lead to an ability of ploughing more land on time, thereby achieving crop yields and earning higher income. • Use of credit (CREDIT): This is a dummy variable with 1 for user and 0 otherwise. Access to credit is hypothesized as having a positive relationship

with income and negative relation with poverty. According to Norton et al. (1970), credit helps farmers purchase inputs such as seeds, fertilizers and chemicals. • Productive asset holding (ASSETHH): This variable includes all assets that are categorized under agricultural activities but different from land and livestock

43 resources. It is the continuous variable measured in ETB. The value of productive assets is estimated by the household head considering the current price and the salvage value of the asset. As indicated by Maddison A (1970), as people accumulate physical capital, this allows the people to expand production by changing the marginal productivity of inputs such as land and labor. Therefore, households with more productive assets can produce more and increase their total income. Household productive assets should influence

household total income positively and poverty level negatively. • Gender of the household head (GENDEHH): This is a dummy variable with 1 for male and 0 otherwise. Male household heads are expected to have higher income compared to female household heads because of better labor inputs used in male-headed households. • Age of a household head (AGEHH): Age is a continuous variable and measured in years. In Ethiopia, household head is the decision maker for farm

activities. Age is one of the factors that determine decision making of a person. Advanced aged household heads are more reluctant to accept new technology and agricultural production styles than younger household heads. Thus, age of household head is hypothesized to have negative contribution to household income. The relationship between household head age and total income of a household is assumed to be a linear function, based in part on estimated equations.

• Dependency ratio of the household (DEPRATIO): The dependency ratio is equal to the number of individuals aged below 15 and/or above 64 divided by the number of individuals aged 15 to 64, expressed as a percentage (John 2002). Dependency ratio is important because it shows the ratio of economically inactive compared to economically active. The dependency ratio

44 of agricultural households provides planners and policy makers with an indication of agricultural labor availability in male and female managed holdings and their abilities to actively participate in agricultural programs and projects. Members of holdings with high dependency ratios might not be able to participate in programs and projects due to time, labor and/or financial constraints (FAO 2010), that is dependency ratio is thought to be negatively related to income of households.

• Use of input (INPUT): the use of inputs influences household income from crop production. The main inputs used in the study area are chemical fertilizers, improved seeds and agricultural chemicals. Households who use one or more of these farm production inputs will usually have higher crop yields and hence higher income. Thus, a binary variable was specified with a value of 1 for households that had used one or more of these inputs during the previous cropping season, and 0 for households that had not used any of these

inputs during that time. • Prices of inputs and outputs are among the exogenous variables that determine the gross income of a household. However, the price of goods and services are almost identical for all households in the study area. Therefore, price effects cannot be determined and price information is not directly incorporated in the model. Following previous studies, the determinants of household gross income were analyzed by multiple regression models. The model is of this form:

Y = α + D + δX+ σZ+ φA + ԑ (2) Where α is an intercept, and δ, σ and φ are parameters to be estimated.

45 Table 2: Summary of dependent and independent variables codes, definitions and expected sign of effect on household income

Expected Variable Variables definition and measurements sign INCOMEHH Annual household gross income in ETB Dependent Dummy variable for irrigation (1 = access to IRR + irrigation, 0 = No access) LANDSZ Total cultivated land, hectares + FAMSIZE Family size of a household in adult equivalent + Education of the household head (1 = read and EDUHH + write, 0 = Does not read and write) AGEHH Age of a household head, years - Use of production inputs 5(1 = use of inputs, 0 INPUT + = No use of inputs) LIVESTO The number of livestock owned, in TLU + DEPRATIO Dependency ratio of the household - Use of credit (1 = use of credit, and 0 = No use CREDIT + of credit) ASSETHH Asset owned by household in ETB + Sex of the household head (1 = male, and 0 = GENDEHH + female) Some households may not derive income from livestock, off-farm and other activities; therefore in this study, the impacts of irrigation on income were estimated using a Tobit model. This approach was developed by Nobel laureate economist James Tobin in 1958 for analyzing situations whenever dependent variable can take zero values. There are many previous studies with similar works (Nicholson et al. 2004; Zhou et al. 2009; Aschalew 2009; Barket et al. 2002). The specific form of the Tobit model is described as follows:

* Yi = βX‟i + ԑi (3)

* We define a new random variable Y transformed from the original one, Y , by Y* = 0, if Y (3a)

5 Use of input means the application of one or more agricultural inputs such as chemical fertilizer, improved seed, chemicals (pesticide, herbicide) in the last one cropping season.

46 Y* = Y, if Y (3b)

And where Yi is the observed dependent variable measuring combined livestock income, off-farm income, cropping income and household total income, Y* is a latent variable, X is a vector of explanatory variables that influence incomes , β is a vector of parameters to be estimated, and ԑ is a random disturbance term with mean 0 and variance σ2. On the basis of the Tobit model specification, the unknown parameters of the explanatory variables can be estimated by maximizing the corresponding likelihood function.

„ Where Yi is the income of a household, Xi is an explanatory variables create influence on household income, � is a coefficient of the independent variables, � is the normal density function, � is the normal distribution function, σ2 is the variance of the error term epsilon in the third equation.

The coefficients of dependent variables in Tobit model are not directly proportional with change of the independent variable. Therefore, to understand the change of household income as a result of a unit change of the coefficient of independent variables, the estimators of the variables should be transformed in to the vector of first derivatives. The marginal effect in Tobit model illustrate that the change of the dependent variables as a result of the changes of respective independent variable (Xi) by a unit. On the basis of the above Tobit model specification. The marginal effects of the independent variables on household income are represented as:

47 The marginal value identifies the direct impact of irrigation on household income. The hypothesis of that irrigating household have higher income than non-irrigating household is tested by the sign of the marginal effect of the irrigation access variable (D, if significant) and the magnitude indicates the size of the impact. That is, it fulfills a main aim of this study to analyze the marginal effect of irrigation on irrigating households income compared with non-irrigating households income being other things constant. This helps policy makers to understand the value of future irrigation.

4.1.2.3 Poverty level evaluation

Poverty is a multidimensional concept and its definition and measurement has been the subject of much debate. The household poverty line often is represented as a very basic living standard. Poverty indicators are often constructed by comparing household income with the mean income or median income (midpoint). Poverty usually is analyzed on the basis of income or consumption indicators. The World Bank uses poverty line of one dollar (PPP6-adjusted) per day, but this has been criticized for being too narrow. According to Nilsson et al. (2010), there is no obvious best way to calculate measures of absolute purchasing power that are comparable both across time and space when relative prices vary both over time and between countries. The World Bank estimates rely heavily on household surveys, resulting in questionable

6 PPP (purchasing power parity) means the application of one price across countries for all goods and services, or representative groups (baskets) of goods and services.

PUS = ( EUS$/ETB$) x (PETB)

PUS = Price of goods in USA

PETB = Price of goods in Ethiopia

EUS$/ETB$ = US dollar/Ethiopian birr exchange rate.

48 comparability and low coverage for certain regions such as Sub-Saharan Africa. Thus, one approach to poverty definition is to use a relative poverty approach, based on the characteristics of households in the sample Following pervious literature, combined sample households (both irrigating and non-irrigating) are ranked according to their current income. This ranking is then used to determine which quartile a household is in based on current income. The households in the lower quartiles are relatively poor whereas those in the upper quartiles are relatively well off. On the basis of previous studies (e.g., Simtowe et al. 2010), a working hypothesis is that more of non-irrigating households will be in the lowest-income quartile. In addition, it is expected that the low-income quartile households have a lower mean share of income from crop production than households in the higher income quartiles, but a larger mean income share coming from livestock compared to households in larger income quartile. The low-income quartiles are also expected to have a larger income share from non-farm activities compared to households in the upper quartile.

4.1.2.4 Poverty Line

Although the relative poverty approach has some advantages, it is also possible to develop more specific absolute poverty, typically defining a somewhat arbitrary “poverty line” on the basis of income or consumption indicators. Previous authors have used different definitions. Ethiopia has not established any official poverty lines, so Schreiner and Chen (2009) used the international poverty lines in dollars at 2005 purchase-power parity, with the lowest of their thresholds at $ 1.00 per person per day. According to Dercon (1997), the threshold for absolute poverty is $ 0.45 per day per

49 adult (or access to 2200 calories per adult per day) and the moderate poverty level is $0.60 per day per adult (2750 calories per adult per day). In this study, the poverty line is determined on the basis of current income because detailed consumption data were not available. Using the dollar amounts and the recent dollar to ETB exchange rate, the absolute and moderate poverty lines are ETB 3,073 and ETB 3,687 per person in adult equivalent per year, respectively. For the purposes of this study, the absolute poverty line is the value of current income at the twenty-fifth percentile for sample households (3,225 ETB per person in adult equivalent per year) and the moderate poverty line is defined as the value of the thirtieth percentile of current incomes (ETB 3,457 per person in adult equivalent per year). These values are reasonable close to those described by Dercon, and are consistent with income and food poverty prevalence of 29.2 and 28.2 %, respectively, in recent years (MOFED, 2010). However, both the absolute and moderate poverty lines defined are below the World Bank‟s poverty line set at one dollar per day per adult, which would be ETB 6,231 per year per adult at current exchange rate ($1 = ETB 17.07). Using the official exchange rate in the National Bank of Ethiopia, the absolute poverty line ETB 3,225 per year per adult is approximately $189/adult/year. The estimated moderate poverty line for the study area is ETB 3,457 per year per adult is approximately $202/adult/year.

4.1.2.5 Poverty level comparison

The poverty level comparison between irrigating and non-irrigating households is valuable to estimate the impact of irrigation on poverty reduction. Poverty level comparison helps to estimate the extent of irrigation‟s impact on rural poverty alleviation. Poverty level comparisons between irrigating and non-irrigating households use the following poverty measures developed by Foster et al. (1984):

Where Pα is the poverty level indicator for a sample of households, m is the number of households below the poverty line, n is the number of households, z is poverty line

th yi is income per adult equivalent of i household, α is the poverty sensitivity parameter that can take on a variety of values. When α = 0, the result is the prevalence of poverty or the head count ratio, that is the proportion of people falling below the poverty line. When α = 1, the equation gives the depth of poverty. It is also called poverty gap index. This shows the amount of income necessary to bring everyone in poverty up to the poverty line, divided by total population. This can be thought of as the amount of income that an average person in the economy would have to contribute for poverty to be eliminated.

4.1.2.6 Econometric model specification

Assessing the impact of irrigation on the likelihood that a household is in poverty is one of the objectives of this study. Thus, poverty is the dependent variable, and is determined by independent variables such as irrigation, household characteristics, asset holdings and access to services (Table 3). In this analysis, the independent variable is binary (1 if the household is classified as poor when its annual income is in the lowest quartile, and 0 if the household is classified as non poor). Under this limited dependent variable model, the probability that the i th household is poor is given by:

Prob (y = 1│X) = F (Xi, β) (7)

Where Zi is a function of explanatory variables (Xki), and expressed as:

Zi = β0 + β1X1i + β2X2i+ β3X3i + β4X4i +… βkXni + μi (7a)

Where μi = error term.

51 th If Pi is the probability of the i household is being poor, then (1-Pi) is the probability of not being poor. Since the dependent variable, poverty, is unobserved and the resulting model is nonlinear, it cannot be estimated by using OLS so maximum likelihood can be used. Green (2002) indicates that either Probit or Logit models are mainly used for the dependent variable that takes dichotomous values (e.g., yes or no) or a choice between two alternatives. Both the Logit and Probit models guarantee that the estimated probabilities lies in the range 0 to 1 and that they are non- linearly related to the explanatory variables. Following Habitamu (2009) and Haile (2008), the dichotomous dependent variable poor/non poor is estimated by Logit model, for the sake of its mathematical convince. The probability of being poor can be expressed in binary choice models or a logistic distribution function as:

(8)

Where: exp (the value of e), the base of natural logarithm.

Table 3: Summary of the dependent and independent variables, codes, definitions and expected signs.

Variable Expected Variables definition and measurements code sign Probability to being poor (1 = poor, and 0 = non- Poverty Dependent poor) Dummy variable for irrigation (1 = access to IRR - irrigation, 0 = No access) LANDSZ Total cultivated land, hectares per capita. - FAMSIZE Family size of a household in adult equivalent - Education of the household head (1 = read and EDUHHL - write, 0 = Does not read and write) AGEHH Age of a household head, years + Use of production inputs (1= use inputs, 0 = No INPUT - use of inputs) OXEN The number of oxen owned, in TLU per capita - DEPRATIO Dependency ratio of the household -

52 Use of credit (1= Use credit, 0= Does not use CREDIT - credit) ASSETHH Asset owned by household in ETB per capita - Sex of the household head (1 = male, and 0 = GENDEHH - female) For the nonlinear dependent variable, the marginal effect of each independent variable is not straight forward to interpret. In the Logit model the marginal effect of each independent variable on poverty should be transformed to Log odds ratio coefficients. Therefore, the regression equation of the odds ratio is:

The Log odds ratio shows change of the probability that a household is being poor/non-poor if the independent variable (Xi) changes by one unit. The statistical t- tests, chi square, minimum, maximum and percentages were done by Statistical Package for Social Science (SPSS) for Windows Release (SPSS, Inc., Chicago, Illinois). The income and poverty analyses were done by STATA/SE 10.0 for

Windows (Stata Corp LP, College Station, Texas USA) software was used for data processing.

53 CHAPTER FIVE

5 RESULTS AND DISCUSSION 5.1 Household Socio-economic characteristics

This section describes the analysis of survey data and its interpretation. In the first section, the sample households‟ demographic characteristics are discussed. Particular reference is given to the factors hypothesized to influence income, such as family size, education level, land holding, asset holding, labor availability, access and source of credit for irrigating and non-irrigating households. These descriptive analyses help to frame the econometric results obtained in the study.

5.1.1 Family size

Family size is useful for formulating various development plans and for monitoring and evaluating their implementation. Average family size at the national level in Ethiopia was 4.7 (CSA 2007). In the study area, the average family size was 5.93 with a minimum 2 and maximum of 11. The t-test shows that there is significant difference in family size between the irrigating and non-irrigating households at a 5% level of significance (Table 4).

Table 4: Family size, family labor and dependency ratio for irrigating and non- irrigating households

Irrigating Non-Irrigating Total t-value for Characteristics households households Households difference (N=90) (N=90) (N=180) Family size, 6.3 5.6 5.9 2.5** persons Family size in AE 4.9 4.3 4.6 3.1*** (family labor) Dependency ratio 1.5 1.4 1.4 0.6 ***, ** indicate significant at the 1% and 5% significance levels, respectively

54 5.1.2 Family labor

In rural Ethiopia, household family is the main source of labor for all income sources. Family size in adult equivalents indicates the sample households‟ average family labor force for agricultural production and other income-generating activities. The average family size in adult equivalents in the study area was 4.62 with a minimum 1.74 and maximum of 8.24. The t-test shows that there is significant difference between irrigating and non-irrigating households at 1 % level of significant

(Table 4). Thus, irrigating households have owned better labor input than non- irrigating households.

5.1.3 Dependency ratio

The dependency ratio shows the ratio of economically inactive compared to economically active. Economically active members of a household, whose age is from

15 to 64, are assumed to be the principal sources of income for the household. Household members under 15 and over 65 are assumed to be economically inactive and dependent on economically active members of a household for education, clothing and health care (John 2002). The dependency ratio of agricultural households provides planners and policy makers with an indication of agricultural labor availability in male- and female-managed holdings and their abilities to actively participate in agricultural programs and projects. Members of holdings with high dependency ratios might not be able to participate in programs and projects due to time, labor and/or financial constraints, that is, dependency ratio is thought to be negatively related to income of households (FAO 2010). In the study area, the average dependency ratio was 140 %, which means every 100 economically active persons had 140 extra persons to feed, cloth, educate and medicate. Economically active members (45

55 percent) were less than non-active household members (55 %). This can have important implications for poverty alleviation efforts. No statistically significant difference was observed between irrigating and non-irrigating households for the dependency ratio (Table 4).

5.1.4 Sex and education of the household head

In the study area, the head of the household generally is responsible for the co- ordination of the household activities. As such it is pertinent to examine attributes such as sex and education of the head as one component of irrigation participation decisions. Of the 180 sampled households, about 88% were male-headed. The percentage of non-irrigating female household heads was more than irrigating (Table 5). There is a significant difference in the sex of the sampled household heads for irrigating and non-irrigating households at a 5 % significance level (Table 5). Economic growth is driven by change in people‟s capabilities or their human capital, as affected particularly by their education. Educated people can more easily contribute to the generation of new technologies and more readily utilize those technologies. It is one of the main factors affecting adoption of irrigation technologies to improve agricultural productivity (Maddison et al. 1970). The education level of household heads is higher for irrigating households than non-irrigating households (Table 5).

56 Table 5: Household member, gender, and education and age characterization Non- Chi- Irrigating Total Irrigating square test Characteristics households households households for (N=90) (N=180) (N=90) difference Percent Percent Percent χ2 Household head gender Male 93 83 88 Female 7 17 12 Total 100 100 100 4.4** Household head education Illiterate 21 65 43 Read and write 48 23 35 Elementary complete 27 11 19 Junior complete 3 1 2 High school and above 1 0 1 Total 100 100 100 24.7*** Age of household head 15-30 years 17 14 16 31-45years 53 47 50 46-64 years 28 34 31 65 and above 2 5 3 Total 100 100 100 1.9 ***, ** indicate significant at the 1% and 5% significance levels, respectively.

5.1.5 Age of household head

The average age of the household heads in the study area was 42 years with a minimum of 24 and maximum of 78 years. The age of the household head influences whether the household benefits from the experience of an older person, or has to base its decisions on the risk-taking attitude of a younger farmer. There is no significant difference in the distribution of household head age of the sampled households between irrigating and non-irrigating household heads (Table 5). 5.2 Productive resource

Agricultural production requires resources such as labor, natural resources, agricultural tools and other capital assets. In the foregoing sections, it has been

57 discussed that household income has a critical link with access to productive resources such as labor, land, oxen and agricultural assets. Therefore, the study looks the access of these resources between irrigating and non-irrigating households. Knowing this helps to judge irrigation‟s impact on household‟s income difference.

5.2.1 Land holding

Land is the major productive asset in agrarian countries like Ethiopia.

Cultivated land appears to be the most important scarce factor of production. In the study area, own land, rented and shared lands were used for cultivation. The average land holding size of the sample households in the study area is 1.1 ha, which is comparable to the national land holding of 1.0 hectares. There is no significant difference between irrigating and non-irrigating households in average land holding size (Table 6). Thus, the overall land holding per household among the study group is similar. However, there is a significant difference in their cultivated land size.

Irrigating households have larger cultivated land area than non-irrigating households. Irrigation may generate income and allow accumulation of other productive assets by irrigating households, which facilitate cultivation of additional land through share in and rent in (Table 6) from non-irrigating households. The average grazing land for irrigating and non-irrigating households was 0.21 and 0.16 hectares, respectively. Irrigating households also use more grazing land than non-irrigating, and the difference between them is statistically significant at the 1 % significance level. Sharing of farmland from any other households is commonly practiced in the study area. Share in of farmland is practiced by 70 % and 40 % of irrigating and non- irrigating households; respectively. Irrigating households share in more farmland compared with non-irrigating households, whereas sharing out of farmland was done

58 by only 12 % and 21% for irrigating and non- irrigating households, respectively. Non-irrigating households share out more their own farmland compared with irrigating households (Table 6). Irrigating household participation was higher for land share in but less for land share out. The converse is true for non-irrigating households, which may be due to the fact that irrigating households have better potential to cultivate additional land than non-irrigating households. The proportions of households, who share in, share out and rent in were 62 %, 16% and 15%, respectively. The share in, share out and rent in land size were 0.28, 0.09 and 0.07 ha, respectively.

Table 6: Average landing size (ha) at household‟s level

Non- Irrigating Total t-value Irrigating Characteristics households households for households (N=90) (N=180) difference (N=90) Land holding 1.10 1.00 1.10 1.5 Cultivated land 1.10 0 .80 1.00 4.4*** Grazing land 0.21 0.16 0.19 2.1** Land share in 0.36 0.21 0.28 2.8*** Land share out 0.05 0.14 0.09 -2.4** Land rent in 0.09 0.04 0.07 0.8 Land rent out 0.00 0.001 0.001 - Fallow and woodlot 0.19 0.14 0.16 1.8 land ***, ** indicate significant at the 1% and 5% significance levels, respectively.

Rent in of farmland is practiced by 21% and 11 % of irrigating and non- irrigating households, respectively. The mean rented in farmland from any other household for both irrigating and non-irrigating households were 0.09 and 0.04 ha, respectively. Non-irrigating household (7%) participates in rent out of their farmland whereas irrigating households did not rent out any farmland. Irrigating household participation was higher for land share in but less for land share out. The other type of land tenure is fallow and woodlot (Eucalyptus tree and other perennial crops) land.

59 The average fallow and woodlot land size of the sample households 0.16 ha, which is 0.19 and 0.14 ha for irrigating and non-irrigating, there is no significant difference between them.

5.2.2 Effect of irrigation on land rent value

In Ethiopia land is a public property. Sale of land is not allowed, but land rental and sharing through agreement between users for one or two cropping seasons is common in the study area. The rental value of the land depends on the quality of the land and the access to irrigation. The average rental values of land accessed with irrigation and land without access to irrigation were ETB 5,816 and ETB 2,867 per ha per one crop season, respectively. This is consistent with the hypothesis that irrigation increases the value of net returns to land. Households who have farm plots with access to irrigation water thus will have higher incomes per ha from land rent (Table 7).

Table 7: Average land rental rate

Land not Land accessed t-test for Characteristics accessed for for irrigation difference irrigation Land rental rate, 5,816 2,867 3.9*** ETB/ha *** indicates significant at the 1% significance level.

5.2.3 Production assets

Agricultural production assets include motor pumps, treadle pumps, plough sets and equipments necessary for agricultural activities. The production assets for irrigating and non-irrigating households are valued by considering the salvage value of each asset. As mentioned in the literatures review section of this paper, irrigation

60 development has several benefits and roles, one of these benefits are increasing wealth of households. Irrigating households have, on average, more agricultural production assets than non-irrigating households (Table 8). This difference is statistically significant at the 1 % significance level.

Table 8: Mean value of agriculture production assets at household‟s level

Non- Irrigating Total irrigating t-value for households households Characteristics households difference (N=90) (N=180) (N=90) Production assets, 2,362.6 771.8 1,567.2 5.1*** ETB *** indicates significant at the 1% significance level.

5.2.4 Type of houses

Types of housing are an indicator of improving the well-being of rural households. In rural Ethiopia most of the houses are grass-roofed, but wealthier households will have a corrugated iron roof. A higher percentage of irrigating households in the sample had corrugated iron roofed houses than non-irrigating households, but statistically there is no significant difference (Table 9).

Table 9 : Housing types in samples households

Non- Chi-square Irrigating Total Housing types irrigating for difference Percent Percent Percent χ2 test Grass roof 23 30 27 Corrugated roof 77 70 73 Total 100 100 100 0.46 5.3 Major crops grown using small-scale irrigation

The crops grown without irrigation (rainfed) in the study region are teff, barely, wheat, maize, finger millet, rice, oat, vetch, check pea, onion, tomato, potato

61 and pepper. In addition to their rainfed cultivation, irrigating households produced cash crops for the second round within a year in dry periods using irrigation water. The main field crops grown using small-scale irrigation schemes in the study region are maize, oat, rice and vetch and the dominant vegetables are onion, tomato, potato and pepper. The first main crop season is from June to November. In this period both irrigating and non-irrigating households produced rainfed crops. The second crop season is practiced in dry seasons from December to April. In this cropping season, only irrigating households can cultivate using water from irrigation. Access to irrigation has been regarded as a powerful factor that provides a greater opportunity for multiple cropping, cropping intensity, and crop diversification (Saleth et al. 2003). Households who have access to small-scale irrigation can cultivate twice a year. Thus, irrigation increases the intensity of cropping. The most common field crops (cereals) produced by small-scale irrigation are maize, oats, and vetch whereas the most commonly produced irrigated vegetables are onion, tomato and potato (Table 10).

Table 10: The major field crops and vegetables grown using small-scale irrigation

Percent of irrigating Crop types Obs. households growing Field crops (Cereals) Maize 19 21 Oats 15 17 Vetch 4 4 Maize and Oat 7 8 Three or more crops 6 7 Vegetables Onion 46 51 Tomato 6 7 Potato 2 2 Onion and Tomato 18 20 Three or more 9 10 In field crop (cereal) cultivation using small-scale irrigation, maize was the dominant. It is grown by 21 percent of irrigating sample households. Oats and vetch are the second and third major field crops, grown by 17 % and 4 % respondents,

62 respectively (Table 10). Vegetables were the more commonly produced crops with small-scale irrigation systems. The most frequently grown crop was onion, grown by 51 % of irrigating sample households. Onion is better than other vegetables in terms of amount of yields produced and demand in the market, but requires frequent watering. Therefore, onion irrigation with scarce surface water access areas like Werota Zuria has lower yields compared with areas with more water resources, such as Quhir Michael and Shina. Tomato and potato were less commonly produced than onions.

Twenty percent of respondents grew both onion and tomato, whereas 10% of households produced three or more vegetable crops. Crops grown using small-scale irrigation were few in number (Table 10), but there are different reasons why they are grown by irrigating households. The major factors for production decision were good production (50 %), better price (24%) and easier to cultivate (24%). There are other reasons such as disease resistant, seed availability; water scarcity and others accounted 16 % of the respondents (Table 11).

Table 11: Reason for selecting the major field crops and vegetables for irrigation Percent of irrigating Reasons Obs. households responding Good production 45 50 Better price 22 24 Easier to cultivate 9 10 Two or more reasons 15 16 5.4 Household income evaluation

Household gross income is derived from agricultural (crop and livestock) sales and value of crops and livestock products retained for household consumption7. The value of retained crop and livestock products was calculated using annual average nominal prices. In the case of irrigating households, individual household cropping

7 Thus, this is a “full (gross) income” value rather than a “cash (gross) income” value.

63 income was computed from both rainfed and irrigated crops but for non-irrigating households, cropping income was derived from only rainfed crops. The off-farm and non-farm incomes were also computed as part of gross household income. The main reason is to examine the hypotheses that irrigating households‟ income was greater than non-irrigating. Non-irrigating households may use off-farm activities to compensate for their lack of irrigation. Therefore, to evaluate the income difference between irrigating and non-irrigating households due to irrigation, the study considers the off-farm and non-farm incomes. In the hypothesis, irrigating households have higher income than non-irrigating households due to access of irrigation. Non- irrigating households may have better income in off-farm and non-farm activities as a compensation of irrigation then considering all income sources are important to evaluate impact of irrigation on household gross income.

5.4.1 Cropping incomes

The most common crops grown in the study area are onion, rice, maize, vetch, teff, finger millet, barley, oat, chickpea, onion, tomato, potato and pepper (Table 12). These crops are grown as staple and cash crops in the study area. The estimation of crop income uses taking the mean annual average price for both the sold and home- consumed crops. The major income source crops for irrigating households were onion (35%), rice (17%) and oats (12%) whereas for non-irrigating households were rice (25%), wheat (13%) and finger millet (12%). Onion and rice were the two main sources of income from crops in the study area. The mean income difference shows that irrigating households were better off in all cropping income than non-irrigating households except wheat and pepper. The largest income was from onion produced using small-scale irrigation. This suggests that small-scale irrigation development

64 increases the incomes of rural household because it directly influences the highest income source, cropping.

Table 12: Major crop types and their mean annual production values

Non-irrigating t-value Ave. Irrigating households households for Major annual difference crop price, Percent Produc- Produc- Percent of in types (ETB of total tion value, tion value, total crop produc- /100kg) crop 000 ETB 000 ETB income tion value income Teff 626 1.5 5 1.5 11 0.7 Barley 317 0.4 1 0.3 2 0.6 Wheat 463 0.5 1 1.7 13 0.5 Maize 340 1.6 5 1.2 9 1.1 Finger 289 1.6 5 1.6 12 1.0 millet Rice 629 5.5 17 3.3 25 2.5** Oat 573 4 12 1.4 11 2.3** Vetch 330 1.2 4 0.8 6 2.4* Check 385 0.6 2 0.6 5 1.5 pea Onion 473 11.4 35 0 0 6.1*** Tomato 237 3.2 10 0 0 2.1*** Potato 222 0.3 1 0.4 3 0.6 Pepper 1663 0.4 2 0.4 3 0.9 Total - 32.3 100 13.4 100 6.6*** ***, **,* indicate significant at the 1%, 5% and 10% significance levels, respectively.

5.4.1.1 Rainfed cropping income

Rainfed crops were cultivated by both irrigating and non-irrigating households. But, unlike irrigating households, non-irrigating households depend only on rainfed cultivation. The major reasons for non-irrigating households not irrigating were lack

65 of surface water access8 in their farm plot (85%), lack of financial capital (8%) and labor (7%) (Table13).

Table 13: The reasons for non-irrigating households for not irrigating

Percent of non- Reasons Obs. irrigating households No surface water access 77 85 Shortage of money 7 8 Shortage of labor 6 7 Total 90 100 Lack of surface water access was the most important limiting factor; however the literature indicates that the groundwater table is high in the study area (Johnston et al. 2010 and Girum 2010). Thus, in addition to surface water, groundwater based irrigation development might be given additional consideration as a means of irrigation development in the study area. The mean annual income from rainfed cropping was ETB 14,189 (Table 14). Statistically, there is no significance difference between irrigating and non-irrigating households in their mean annual rainfed income.

Table 14: Rainfed income for irrigating and non-irrigating households in ETB

Non- Irrigating Total irrigating t-value for Characteristics households households households difference (N=90) (N=180) (N=90) Rainfed income 15,011 13,366 14,189 1.15

5.4.1.2 Irrigated crop income in PAs

The average irrigated land size 0.65 ha per household with a minimum of 0.12 ha and a maximum of 2.75 ha. The major irrigated crops in the study area are onion,

8 Lack of surface water access means the farm plots of a household were topographically inaccessible with irrigation water from river, pond, lake or any surface water.

66 tomato, maize, oats and vetch. The mean annual cropping income from sample irrigating households was ETB 17,271 (Table 15). Table 15: Income from irrigated crop production in ETB Std. Maximu PA Obs. Mean Minimum F-test dev. m Werota Zuria 14 9,730 9,072.2 649 34,967 Bebekis 19 17,213 12,074.7 920 41,037 Kokit 14 9,609 9,168.3 1,060 29,240 Kuhir Michael 19 30,704 26,198.0 1,531 80,380 Shina 24 15,550 12329.9 2,114 48,086 Total 90 17,271 17011.5 6,49 80,380 5.3*** *** indicates significant at the 1% significance level. The PAs with higher mean income from irrigated crop production were Kuhir Michael, Bebekis and Shina, respectively. These PAs use concrete river/spring diversion for irrigation water. Kuhir Michael and Shina were upstream and downstream of Guanta river diversion and Tanqua Gabriel spring development. The amount of water for irrigation was high in these PAs. In addition to Guanta and Tanqua Gabriel diversion water, Kuhir Michael used Gumara river and Shina used both Gumara and Rib river. In addition to this, Kuhir Michael and Shina have water use associations. The association has multiple purposes; some of them were avoid conflicts and irrigation water theft, used as a source of market information, supply farm inputs for member households, protect and amend river diversion canal when damaged. Thus, these PAs have better irrigation water use system and can earn better income from irrigation farming. Werota Zuria has the lowest income from irrigated crop production. The only water source for the PA is Eriza River; the amount of water was low during flowering period of irrigated crops during January and February. Water scarcity was the main problem in the PA. On the upper part of this river, there is concrete canal river diversion but the amount of water became declined at peak irrigation season, which

67 caused a conflict between the upper and lower stream user of the river. Therefore, water scarcity was the main cause for low income from irrigated crop production.

5.4.1.3 Total cropping income

Total cropping income is the amount of mean annual income of a household obtained from both types of cropping systems, rainfed and irrigation. The mean annual income of a household from cropping income in the sample PAs was ETB 22,824

(Table 16).

Table 16: Total mean annual cropping income at household level in ETB

Non- Irrigating irrigating Total t-value for Characteristics households households (N=180) difference (N=90 (N=90) Mean annual cropping 32,282 13,366 22,824 7.7*** income *** indicates significant at the 1% significance level The total mean annual cropping income of irrigating households was substantially higher than that for non-irrigating households. The t-test shows that there is a significant difference between irrigating and non-irrigating households at 1% level of significance (Table 16). This suggests that irrigation markedly increases income, but this will be more appropriately tested using econometric analysis.

5.4.2 Livestock income

The type of agriculture in the study area is settled agriculture with a mixed farming system (i.e., integrated crop and livestock production). Livestock are the most important productive assets in the household. In the study area, livestock are important source of power for ploughing, transportation, and riding. Livestock also consolidate the social organization as they serve in payment for blood compensation and gifts for

68 relatives. They play role in religious and cultural ceremonies and serve as source of prestige. It also considered as a saved asset used during periods of food shortage. The average livestock holding for sample households was 4.02 TLU. Irrigating households possess a larger average number of livestock (4.89) than non-irrigating households (3.15). There is a significant difference between irrigating and non-irrigating households at the 1% significance level (Table 17). Table 17: Number of livestock (TLU) Irrigating Non-irrigating Total t-value for Characteristics households households (N=180) difference (N=90) (N=90) Average number of 4.89 3.15 4.02 5.4*** livestock (TLU) *** is significant at the 1% significance level. Livestock play a significant role as income sources in rural poor Ethiopia. Sale of live animals and their products are main livestock-related income sources in the study area. The livestock income category includes income from the sale of livestock, livestock products (i.e. milk, eggs, honey etc.) and other by-products like hide and skin. The values of sale and own consumption livestock and livestock products were estimated based on the average annual nominal prices. The livestock products were collected on a weekly basis, and converted to estimate annual income (Table 18). The highest mean livestock income among the study PAs are reported in Shina. Livestock farming system in the study areas is free grazing on communal grazing lands. The high livestock income in this PA was due to better communal grazing land. The PA has high water potential. The two major rivers in Tana basin, Gumara and Rib pass on the PA. The borders of these rivers are rich in livestock feed resources. On the other hand, the PA also contains more irrigated areas than other PAs. In addition to high surface water access, the groundwater table is high so all sample households are irrigation users. From focus group discussions and key informant interviews because

69 the people have enough food from irrigated and rainfed cropping, animals are not being sold to get food. Therefore, each household has high livestock numbers and livestock income.

Table 18: Average annual livestock income

Std. PA Obs. Mean Minimum Maximum F-test deviation Werota Zuria 36 3,822 3463.9 0.0 12,388 Bebekis 42 1,105 1277.3 0.0 5,622.4 Kokit 38 2,472 3343.4 0.0 18,992 Shina 24 5,377 4894.1 580.6 20,462.8 Total 180 2,783 3348.5 0.0 20,462.8 8.6*** *** is significant at the 1% significance level. The second highest mean livestock income is reported in Werota Zuria. The average livestock income was ETB 3,822 with a minimum of 0 and a maximum of 12,388. Werota is the capital town of the district. Because there are hotels and restaurants in the town, demand for livestock products like milk and egg was high, which encourages households to produce more. In addition to this, crosses of the

Fogera dairy cow breed with exotic dairy breeds, especially Holstein, were common in the PA. There are households who practice intensive dairy and fattening farming in the area. The lowest mean income from livestock is reported in Bebekis. On average, the livestock income was ETB 1105 with a minimum of 0 and a maximum of 5,622. The PA is an upland area with poor livestock feed resources. The community is also relatively poor. Most of the household members in this PA are employed in other PAs as daily laborers. They sell their livestock frequently to purchase grain, so their livestock holdings and livestock income was low. The mean livestock income for irrigating and non-irrigating household was ETB 3,232 and ETB 2,440, respectively. Irrigating households had larger livestock income than non-irrigating households, but statistically there is no significant

70 difference. The overall mean income of livestock in all sample households is ETB 2,783 with a minimum of 0 and a maximum of 20,463. This indicates that livestock farming is another of the major income sources of the study area. There is a significant difference among sample PAs at the 1% significance level (Table 18).

5.4.3 Off-farm and other incomes

Off- farm and other incomes are important parts of total income in rural households of Ethiopia. They are significant for purchasing power and food security. Petty trading was one source of off-farm income in the study area, for instance, onion and tomato trading in Gumara town. The sale of onions and tomatoes transported to Bahir Dar and Gonder towns was another income source. Werota, the capital town of the district provides off-farm income for the surrounding households, especially in petty trading. Some households have houses in town and rents are another source of income. Thus, average off-farm and non-farm income share were highest in Werota

Zuria compared to other PAs. The other sources of off-farm income in the area were employment on other farms during weeding and harvesting seasons, sale of wood, sale of local drinks (tela), renting of irrigable lands, artisan (blacksmith, weaving and pottery), brokering, sale of wood (charcoal), house rent and remittance.

Table 19: The mean off-farm and other incomes

Irrigating Non-irrigating Total t-value for Characteristics households households (N=180) difference (N=90) (N=90) Off-farm incomes 622 667 645 - 0.3 The average off-farm income for sample households was ETB 645 (table 19). Irrigating households also get off-farm income from the rent of water pump and houses rent in the town. Water pumps were rented on average for ETB 12.50/hour.

71 The difference in off-farm income between irrigating and non-irrigating households is not statistically significant.

5.4.4 Summary of income sources at household level

The total mean annual household income in the study area was ETB 26,251 (Table 20), which is roughly equal to the average per capita income for Ethiopia as a whole. From the total mean annual income of a household, cropping contributes the highest income share (86%) followed by livestock (11%) and off-farm (3%), respectively. Irrigating households earn higher income from cropping than non-irrigating households. However, there is no significant difference between irrigating and non- irrigating households in their livestock and off-farm incomes. The total income significant difference arises from the cropping income difference, which is suggestive of the both the mechanism and the degree to which irrigation access increases household incomes. The next section discusses the results of econometric analysis that assesses the impact of irrigation controlling for other factors that influence income.

Table 20: Summary of annual household income sources:

Non- Irrigating t-value irrigating Total Characteristics households Percent for households (N=180) (N=90) difference (N=90) Crop income 32,282 13,366 22,824 86 7.7*** Livestock 3,132 2,433 2,783 11 1.4 income Off-farm 622 667 645 3 - 0.3 income Total income 36,036 16,466 26,251 100 7.6*** *** indicates significant at the 1% significance level.

72 5.4.5 Econometric model for income analysis

This section provides additional analysis of the impact of irrigation on the performance of gross incomes of the irrigating households. This study uses an econometric model for the analysis of impacts of irrigation on household income. In this analysis, the dependent variable is the household annual total income derived from rainfed and irrigation crops, livestock and their products, off-farm and other incomes reported over the past 12 months (from November 2009-October 2010).

The income analysis was estimated using a Tobit (censored regression) model. The analysis was carried out using Stata software. Multicollinearity was examined using Variance inflation factor (VIF) and correlation coefficients. The values of the VIF for explanatory variables were found to be less than 10 and total of eleven explanatory variables were entered in to the regression analysis. The dependent variable, total income of a household, has non-zero value. According to Holloway et al. (2004), use of a Tobit model at a non-zero dependent variable may increase the magnitude of bias, and to avoid this problem they develop alternative approaches. In the censored regression model, the minimum of the observed values defines the maximum for the censoring value. That is an upper bound on n is the minimum of the set (yi , i c) by the same reasoning, because the observed total income is non-negative. A logical lower bound on total income (п) on n is zero. In other words, logic constrains the feasible choice for п to the closed interval

(10)

This equation provides the censoring points to can be vary within the range of values that lies below the minimum of the observed, positive quantities.

73 On the basis of this alternative, the observed total minimum income at household level is ETB 1,256, that is, a non-zero value. By considering the above revised approaches Tobit regression model was used with 1255 as lower limit. The estimates of coefficients by the Tobit regression model as tool of parameter estimation are depicted below (Table 21). The Tobit analysis suggests that several variables have a statistically significant impact on the total income of the household, many of which are consistent with the hypothesized relationships. The analysis indicates which determinants are more important for the improvement of total household income. Some variables appear to be insignificant; this may be due to the relatively small sample size involved. Education of household head (EDUHH) has statistically significant positive impact on the total income of a household. This seems rational; educated human capital can more easily adopt technologies like irrigation and make more informed production decision. Education can increase the marginal productivity of labor. The increase in productivity of labor is one of the important factors to increase income of a household. Household family size in adult equivalent (FAMSZADUL) and livestock holding in TLU (LIVESTO) are positively associated with household total income, and both of them are statistically significant. Household family size in adult equivalent means a larger amount of labor available to the household. Labor increases productivity per ha of land, and in turn, household total income increases for a given land base. The positive association between labor and household total income seems reasonable. Livestock holding in TLU (LIVESTO) contributes to total household income directly through the sale of livestock and their products, and indirectly through use as a source of draught power for crop production activities.

74 Table 21: Tobit estimates of the determinants for household total income

Variable Coef. Std. Err. t P>|t| AGEHH -16.54 50.73 -0.33 0.74 EDUHH 4915.29 1487.37 3.30 0.00 GENDERHH 98.65 1755.29 0.06 0.96 DEPRATIO -1031.12 692.57 -1.49 0.14 FAMSZADUL 1554.59 505.83 3.07 0.00 LIVESTO 2285.07 374.29 6.11 0.00 IRR 3359.46 1222.01 2.75 0.01 ASSETHH 2.81 .34 8.26 0.00 LANDSZ 10291.91 1607.31 6.40 0.00 INPUT 4688.55 1738.96 2.70 0.01 CREDIT -894.16 1052.57 -0.85 0.39 Constant -10696.22 3561.15 -3.00 0.00 Sigma 6778.27 358.72

Number of obs. 180 LR chi2(11) 386.63 Prob > chi2 0.00 Pseudo R2 0.09 Log likelihood -1834.24

Access to irrigation (IRR) influences the household total income significantly with a positive sign as expected. As Maddison (1970) suggest, access of technology (irrigation) shifts the production function and offsets the diminishing marginal return by doing so increases income and used as a source of economic growth. According to Makombe (2007), the production function analysis of irrigated and non-irrigated farm plots, the result shows that irrigation shifts the agricultural production frontier to a higher level. The marginal productivities of land and labor for the irrigated farms are almost four, and five times more, respectively. Thus, access to irrigation is one among many factors that increase household incomes. Household production asset value (ASSETHH) influences the household total income significantly with a positive sign. This tells us households with high production assets can produce more and increase their total income. This is consistent with the economics of transformation and growth principles (Maddison et al. 1970) as

75 people accumulate physical capital allows the people to expand production by changing the marginal productivity of inputs like land and labor. Land size (LANDSZ) is positively associated with household total income as expected. Land holding is highly significant to the household total income. Land is important fixed input to increase production and income. Use of input (INPUT) influences household income significantly, and as hypothesized has a positive impact. Households who use input have higher household income. As Maddison et al. (1970) suggest, one of the main strategies for agricultural development depends on the availability and financing of new inputs like chemical fertilizers, new seeds, pesticides and the like. Age of household head age (AGEHH), Gender of household head (GENDERHH), Household dependency ratio (DEPRATIO) and Household access to credit (CREDIT) have no statistically significant effect on the total income of a household.

The previous discussion indicated the sign and statistical significance of the coefficients from the Tobit model. However, in that model the coefficients do not directly represent the marginal effect, that is, the impact on household income from a one-unit change in the independent variables. The marginal effects are calculated using equation (5) using the mean values of the independent variables. The marginal effect estimates reveal that the land size (LANDSZ) has the largest impact. That is, a one ha land change has an impact on income for 10,275 ETB per year (Table 22).

Thus, land holding size is very important input in rural poor households to increase their annual income (although it will typically be difficult for a household to markedly increase the size of its landholding). Because agriculture is the main source of income and livelihood for more than 85% of the country‟s population, land access is a critical and sensitive political issue in contemporary history of Ethiopia (Helland 1999). In the

76 study area, land is a scarce resource. Land share in/out and rent in/out is common. Even thought the cost in cash of land is not far from the estimated marginal impact of land, the additional costs such as transaction cost and monitoring cost are high. Therefore, it is not easy to increase a land access for the individual household. Education of household head (EDUHH) is another important factor that influences the annual total income of a household. The analysis shows that education (literacy) significantly increases the household‟s total income by ETB 4,903.3.

Use of inputs (INPUT) influences household income from crop production. The main inputs used in the study area are chemical fertilizers, improved seeds and chemicals (pesticide, herbicide). Households who used one or more of these inputs increase their income significantly. The marginal effect of the Tobit model revealed that households who have access for inputs can increase their income by ETB 4,670 per annum. Although it was not tested formally in this study through the estimation of a production function, other agricultural inputs probably are complementary with irrigation. Thus, consideration should be given not only to irrigation water access but also to other agricultural inputs. Access to irrigation (IRR) has a significant impact on the total income of a household, ETB 3,353 per year (or a 27% increase in the mean income without irrigation). This supports the initial hypothesis that access to irrigation increases households‟ income, controlling for other factors. Households who have access to irrigation can cultivate their irrigated land two or more times a year. Although the econometric analysis cannot indicate directly why the increase in income occurs, irrigation allows the farmers to practice crop intensification 9 and diversification10,

9 Crop intensity means cultivating two or three times per year 7Crop diversification means producing two or more crops per one crop season

77 which increases crop yields and revenues from crop sales. Irrigation likely also increases the marginal land and labor productivity, increases the crop production and then promotes household income.

Table 22: Marginal effects of determinants on household total income

Determinant dy/dx Std. Err. z P>|z| AGEHH -16.51 50.65 -0.33 0.74 EDUHH 4903.33 1481.80 3.31 0.00 GENDERHH 98.48 1752.30 0.06 0.95 DEPRATIO -1029.43 691.43 -1.49 0.14 FAMSZADUL 1552.03 504.99 3.07 0.00 LIVESTO 2281.29 373.67 6.11 0.00 IRR 3353.29 1219.40 2.75 0.01 ASSETHH 2.80 0.34 8.26 0.00 LANDSZ 10274.89 1604.70 6.40 0.00 INPUT 4669.84 1725.20 2.71 0.01 CREDIT -892.63 1050.60 -0.85 0.39 Livestock holding (LIVESTO) also affects annual total income of a household. An increase of household‟s livestock holding by one TLU is estimated to increase the total income of a household by ETB 2,281 per annum. As expected, the value of productive assets owned by the household (ASSETHH) also increases total income of a household. The increase in asset holding of a household by ETB 1,000 is estimated to increase total household income by ETB 2,800. This suggests that if households invested in more productive assets, they would pay for themselves in a relatively short time. (However, it is important to note that the effect is on gross, rather than net income, so the return on productive assets cannot be directly calculated based on these results.)

Household size in adult equivalent (FAMSZADUL) also increases the annual income of a household. A one-unit increase family size in adult equivalent increases the total income of a household by about ETB 1,600.

78 5.4.6 Comparison of sample small-scale irrigation types at household level

In addition to the overall impact that irrigation has on household incomes, it is relevant to consider the income generated by different types of irrigation systems. Each small-scale irrigation type has its own advantages and disadvantages. Sample households use concrete river/spring diversion, motor pump irrigation, treadle pump irrigation and traditional river diversion.

5.4.6.1 Sample small-scale irrigation types and irrigated crop income.

The concrete canal river/spring diversion generates more income per household on average than other irrigation types. The mean annual irrigated cropping income from concrete canal river/spring diversion is ETB 25,610 (Table 23); 40% of total mean annual income from all irrigated crops). This indicates that concrete canal river/spring diversion irrigation produces the highest income share compared with the other sample small-scale irrigation types. The reason is the amount of water supplied is large, especially in Kuhir Michael and Shina PAs. Moreover, water wastage is relatively low due to the water user association. The association establishes a water use timetable for all user households, and then the amount of water supplied to irrigated farm plots is equal and adequate. This limits water theft and conflict among irrigation water users. The concrete canal river/spring diversion water user pays for the water. The payment depends on the amount of irrigated land owned by a household. The average payment was ETB 40 / ha/year. The users believed that the payment is low compared to the benefits provided by the service, which is consistent with the findings of this study with regard to the impact of irrigation water access on household incomes.

79 The small-scale irrigation type that produced the second-largest amount of income per household was the motor pump. Motor pump irrigation users can access water easily from any perennial rivers. Motor pump are widely used in the study area. Rivers such as Gumara, Rib and Eriza are widely exploited by motor pump irrigation users. The mean annual irrigated cropping income from motor pump is ETB 22,422 (Table 23). Households who have a motor pump get additional income by renting the pump for other households. However, an issue is the cost of fuel and durability of the pump.

Table 23: The sample small-scale irrigation types and irrigated crop income per irrigating household

Irrigation Std. Obs. Mean Minimum Maximum Percent types deviation Concrete canal 24 25,610 21,253.4 4,895 80,380 40 river/spring diversion Motor pump 24 22,422 17,886.9 1,176 62,841 34 Pedal pump 20 9,502 7,217.1 1,658 29,240 12 Total 90 17,271 17,011.5 649 80,380 100 Many households use traditional river diversion in the study area. The mean annual irrigated crop income from this irrigation system is ETB 9,615 per household. This irrigation system has no direct cash cost; the river diversion is made by user farmers using their own labor. For this reason, poor rural households are more often interested in and use this irrigation system. The main problem in traditional river diversion was the shortage of water due to lack of rivers accessible to traditional diversion. The more easily traditionally diverted rivers often provide limited water in the dry months before irrigated crops are harvested. Pedal pumps have high demand by irrigation user households in the study area due to their low cost and the easy water access they allow. In the Fogera plain, the

80 water table is high, so many households use bucket and jar irrigation from shallow wells. These households have high demand for treadle pump especially the Indian treadle pump. The mean annual income from pedal pump irrigation is ETB 9,502 per household. It is relevant to compare the income from crops under different types of irrigation systems, even thought this does not control for other factors that may influence cropping income and therefore cannot be used directly to indicate that one irrigation type is more profitable than another. As noted in Chapter 4, with four irrigation types there are six pair wise comparisons, which will be tested for differences with a combined overall significance level using the Games-Howell test. The statistical mean comparison revealed that concrete canal river/spring diversion has a significant difference with traditional river diversion and pedal pump at the 5 % and 1 % significance levels, respectively (Table 24). Motor pump irrigation has also a significance difference with traditional river diversion and pedal pump at a 5 % significance level. However, there is no significant difference between concrete river/spring diversion and motor pump, nor a significant difference between traditional river diversion and pedal pump.

Table 24: The sample small-scale irrigation types and irrigated crop income

Post Hoc multiple comparisons, Games-Howell Mean (I) Sample small- (J) Sample small-scale Std. Difference scale irrigation types irrigation types Error (I-J) Traditional river diversion 15994.8** 4.859.8 Concrete canal Motor pump 3187.6 5.670.3 river/spring diversion Treadle pump 16107.6*** 4.628.8 Traditional river Motor pump -1.2807.2** 4.257.6 diversion Pedal pump 112.8 2.720.4 Motor pump Pedal pump 12920.0** 3.991.9 ***, ** indicate significant at the 1% and 5% significance levels, respectively.

81 5.4.6.2 The small-scale irrigation types and total income of household

Irrigation has significant impact on the total income of a household in addition to the impact on cropping income. Small-scale irrigating households have higher mean value income than non-irrigating households. The amount of annual total income of a household is influenced by the type of small-scale irrigation used.

Table 25: Small scale irrigation types and total income of a household

Std. Irrigation types Obs. Mean Minimum Maximum deviation Concrete canal 24 46,530 26,736 17,775 106,000 river/spring diversion Traditional river 22 25,544 11,613 10,860 50,797 diversion Motor pump 24 44,840 22,521 6,802 92,741 Treadle pump 20 24,419 10,453 6,948 44,174 Non-irrigating 90 16,466 11,232 1,256 44,907 Total 180 26,251 19,930 1,256 106,000 Canal river diversion users have the highest average total household income, with a mean of ETB 46,530. The average annual income for motor pump irrigating households is similar to that of concrete canal river/spring diversion, with ETB 44,840. Traditional river diversion and pedal pump irrigation have mean annual incomes significantly lower than canal or motor pump users, with ETB 25,544 and 24,419, respectively. It is pertinent to compare the total mean annual income under four different types of irrigating and non-irrigating households, even thought this does not control for other factors that may influence mean annual income of household (Table 26). For the four types of irrigation and non-irrigation systems, there are nine pair wise comparisons that were tested for differences with a combined overall significance level using the Games-Howell test. The statistical mean comparisons revealed that the four irrigation systems have a significant difference with non-irrigation system.

82 Table 26:Small-scale irrigation types and the mean annual income of a household

(J) Sample small- (I) Sample small- Mean scale irrigation Std. Error scale irrigation types difference (I-J) types Traditional river 20986.3** 5992.9 diversion Concrete canal Motor pump 1689.5 7135.8 river/spring diversion Pedal pump 22111.1*** 5937.1 Non-irrigating 30063.6*** 5584.6 Motor pump -19296.7*** 5221.5 Traditional river Pedal pump 1124.8 3404.9 diversion Non-irrigating 9077.3** 2744.4 Pedal pump 20421.6*** 5157.2 Motor pump Non-irrigating 28374.0*** 4747.1 Pedal pump Non-irrigating 7952.5** 2620.1 ***, ** indicate significant at the 1% and 5% significance levels, respectively. 5.5 Poverty analysis

5.5.1 Poverty level in the study area

As described in chapter 4, the absolute poverty line (i.e., people unable to attain their minimum nutritional requirements) was defined as the value of current income at the twenty-fifth percentile of sample households and moderate poverty line is the value of current income at the thirtieth percentile for sample households. These poverty threshold values allow computation of the proportion of households in poverty, and the poverty gap. The absolute poverty head count ratios of irrigating and non-irrigating households were 7 % and 43%, respectively (Table 27).The moderate poverty head count ratios of irrigating and non-irrigating households were 10% and 50 %, respectively. In the study area, of the sample population who live below the absolute poverty level, 88% are non-irrigating households and only 12% are irrigating households. This suggests that irrigation may have a significant impact on rural poverty alleviation.

83 Table 27: Poverty comparison between irrigating and non-irrigating household

Absolute poverty line Moderate poverty line Head count Poverty gap Head count Poverty gap ratio (P0) index (P1) ratio (P0) index (P1) Irrigating 0.07 0.29 0.10 0.33 households Non-irrigating 0.43 0.43 0.50 0.47 household The poverty gap index shows the proportionate shortfall of average income from poverty line. Because the definitions of absolute and moderate poverty were similar (a difference of only about 200 ETB per person per year), the calculated poverty gap values for absolute and moderate poverty levels are similar for a given household type (Table 27). The estimated poverty gap index using the absolute poverty line for irrigating and non-irrigating household were 29 %and 43%, respectively, while the poverty gap index using the moderate poverty line for irrigating and non-irrigating households were found to be 33% and 47%, respectively. The average income gap of poor people is ETB 943 and 1399 for irrigating and non- irrigating households, respectively. The estimated average income required to bring the poor people out of poverty (poverty line) for non-irrigating households was higher by ETB 546 than irrigating households. The poverty gap index is much larger for non- irrigating households, which again suggests that irrigation may play a role in poverty reduction. The poverty gap concept can also be expressed in terms of monetary values rather than as a proportion, and this is undertaken for individual PAs (Table 28) and by irrigation use status (Table 31).

84 Table 28: The average income poverty gap of the poor by sample PAs

Mean income per adult Mean of income poverty PAs equivalent of the poor, ETB Gap, ETB Werota Zuria 2,290 935 Bebekis 1,500 1,725 Kokit 2,210 1,015 Kuhir Michael 1,995 1,230 Shina 2,652 573 Overall 1,887 1,338 The overall income gap of poor people was ETB 1,338. The estimated average income gap of poor people differs by PA; the gap is lowest in Shina (ETB 573), and highest in Bebekis (ETB 1,725, nearly three times that in Shina), this is due to that Shina is high irrigation potential area. The two irrigable rivers Gumara and Rib are exploited in this PA. In Shina all households are irrigation users. Thus, the poverty gap is low whereas in Bebekis the main irrigation water source is spring development. The amount of water is limited. The access of irrigation from rivers is limited. Thus, households who are accessed to river and spring irrigation have high income whereas those who cannot get are low in income. Therefore, the poverty gap is high in this PA.

Table 29: The average income poverty gap between irrigating and non-irrigating households

Mean income per adult Mean of income equivalent of the poor, ETB poverty Gap ETB Irrigating 2,282 943 Non-irrigating 1,826 1399 Total 1,887 1338 The average income gap of irrigating households was lower than non-irrigating households. This suggests that access to irrigation reduces the poverty gap (and thus reduces the average extent of poverty). The numbers of households below the moderate poverty line are fifty-four (based on the thirtieth percentile of current income and N=180 total households). Of these 54, 49 (91%) are non-irrigating households. The five households below the moderate poverty line use different irrigation

85 technologies: two use pedal pump, one uses traditional river diversion, one uses motor pump, and one uses concrete canal river/spring diversion. The number of irrigating households below the poverty line is small, which makes it difficult to assess the impact of irrigation types on the likelihood of a household being in poverty.

5.5.2 Multivariate Logit regression

A Logit regression model is used to assess the impact of various factors including irrigation access on the probability that a household is in poverty. For this analysis, the poverty threshold is the absolute poverty line, which is defined as the current income at the twenty-fifth percentile of the sampled households. The overall significance of the regression is good (Table 30) based on the probability of getting a LR test (as indicated by the small p- value from the LR test of < 0.00001). Moreover, many of the coefficients of independent variables in the model are significant and have the expected sign.

The estimated coefficient for dummy variable access to irrigation with the odds of being poor over non-poor was negatively correlated and significant. This suggests that the probability to being poor decreases if one has access to irrigation, other factors being constant. This likely is due to the influence that irrigation on crop intensity and crop diversification. Cropping intensity is higher in irrigated household as compared to non-irrigating households. The estimated coefficient for dummy variable access to irrigation with the odd of being poor over non-poor was negatively correlated and significant. This suggests that the probability to being poor decreases if one has access to irrigation, other things factors being constant. This probably is due to the influence that irrigation has on crop intensity and crop diversification. Cropping intensity is higher in irrigated household as compared to non-irrigating households. Because the definition of the poverty

86 threshold in this study is based on current income, and previous results suggest that access to irrigation increases income (especially from cropping), it is not particularly surprising that the likelihood of poverty is lowered by irrigation use.

Table 30: Parameter estimates of a logit model for determinants of a household poverty

Determinants Coef. St. Error Odds ratio Std. Err. AGEHH 0.02 0.02 1.02 0.02 DEPRATIO 0.08 0.32 1.08 0.34 GENDEHH -1.58** 0.70 0.21 0.14 EDUHH -1.73*** 0.56 0.18 0.09 LANDSZ -1.95 ** 0.51 0.01 0.01 ASSETHH -0.001 0.001 0.99 0.001 IRR -1.95 *** 0.51 0.14 0.07 NBOX -2.40 * 1.29 0.09 0.12 Constant 3.26 1.56 Number of Obs. 180 LR chi2(8) 92.39 Prob > chi2 0.00 Log likelihood - 63.76 Pseudo R2 0.42 ***, ** and * are significant at 1 percent, 5 percent and 10 percent significance level, respectively. However, other factors also influence the likelihood that a household is in poverty. As expected, the coefficient of household education is negatively correlated with poverty and significant. The results suggest that household head who is literate had a lower probability of being poor compared with those who are illiterate. Education is assumed to increase productivity and thereby lead to higher levels of welfare for the household.

Poverty also is more likely for female-headed households. This may be due to the fact that female-headed households are responsible for all household tasks in addition to farm activity. Most of female-headed households are older and less educated compared with counterpart male-headed households. Irrigation and other farm operation need high labor especially for intensive ploughing, but female-headed

87 households are disadvantageous with respect to labor endowments. Irrigating female- headed household (7% of total households) were few compared to non-irrigating female-headed household head (17% of total household). Female- headed households also cultivated small size of land (0.89 ha) compared with male- headed household (1.01 ha). The coefficient of land holding per capita was negatively correlated with the probability of a person being poor and statistically significant. The odds ratio illustrates that a one-ha increase in land holding per capita, the odds of being poor decrease markedly (although this is not surprising given that it would result in a doubling of average farm size). As expected, the number of oxen owned was negatively correlated with the probability of a person being poor and statistically significant (but only at the 10 % level). This shows that oxen are an important means of land cultivation and basic factor of production. Households who own more oxen have better chance to not be in poverty because the possession of oxen allows effective utilization of the land and labor resources of the household. A number of variables had no statistically significant impact on the odds ratio. Asset holding per capita was negatively correlated with the probability of a person being poor, but somewhat surprisingly was not statistically significant. Household head age also had no statistically significant impact on the probability of a person being poor which contrasts with findings of previous studies such as Bigsten et al.

(2002). The dependency ratio often is assumed to be positively correlated with the probability of being poor, because the burden of supporting family members too young or too old for productive work falls on other members of the household. In this study, the dependency ratio is computed by taking only old age members above 64

88 years and children with age less than 15 years. The probability being poor may increase if other sick, disabled, or weak members of the household are considered. In addition Ethiopia has high population growth rate, 2.6 percent (CSA 2007), which increases the dependency ratio and increase the probability being poor. Although the coefficient had a positive value for this sample, it was not statistically significant. Consistent with the initial hypothesis, the Logit regression analysis indicates that access to irrigation markedly reduces the odds that a household will be in poverty, at least based on the poverty definition used in this study. Other variables that reduce the likelihood of poverty are household head education, per capita land holding, ownership of oxen and male headed of household head. 5.6 Problems encountered in small-scale irrigation development

Small-scale irrigation has immense potential to improve the incomes of poor rural households in developing countries like Ethiopia, but it is never free from problems. A field survey with focus group discussion and key informant interviews indicates that small-scale irrigation‟s great benefit is accompanied with multidimensional problems. The problems of small-scale irrigation technology development range from individual household‟s biased attitudes to institutional arrangements. The major problems encountered in small-scale irrigation in the study area are problems related to cost, institutional problems, the policy environment, design issues, cultural factors and environmental problems. Loss of water through seepage is the main problem in small-scale irrigation systems in the study area. The non-durability of the physical structure of irrigation schemes and the Vertisol nature of the study area causes high water seepage from river diversion canals. Seepage from irrigation canals is the main causes for water losses in Kuhir Michael.

89 The area development agent Ato Mekonen says “the water loss through canal seepage is the main problem for the source of water shortage for the downstream PA, Shina.” The canals were constructed in 2003, and the canal‟s service length and the black soil nature of the area may cause the canal to be non-functional in the absence of a strong water use association. The current association mends the canal when there is damage, and protects against any misuse or activities that might damage the canal. Water loss through seepage occurs with motor water pumps also (Figure 16). In

Werota Zuria PA, Ato Amare cultivated oats using a motorized pump drawing water from the Eriza River. The main problem faced in his irrigation activities is the water loss through seepage from the delivery hose.

Figure 15: Water loss through seepage from river diversion canal

90 The lack of spare parts in the local market plus the expense of new ones causes difficulty for his irrigation activities. Therefore, seepage causes water shortage in the study area in addition to evaporation and transpiration. Problems with irrigation water distribution also exist in the study area. In the five PAs studied, water distribution and water use principles are unregulated except in Kuhir Michael, Shina and Bebekis where there are water use associations. This causes many conflicts between upstream and downstream irrigating households. For instance, in Werota Zuria PA there were conflicts on the Eriza River between downstream and upstream irrigating households. The main cause of the problem is the amount of water is very small in Eriza River at the end of February; the upstream community always uses all amount of water through modern concrete canal river diversion. There are many households who use motor pump irrigation in the downstream but receive no water during parts of the year. This creates conflict between upstream and downstream water users. Finally, the district judiciary court and the area political leaders resolved the issue through a water use program that allowed use from Monday to Friday for the upstream irrigation users and from Saturday to Sun day for downstream users. Generally, water distribution is the main issue in any irrigation schemes. The study revealed that there are no standardized programs and plans to irrigate each cultivated crops. Irrigation water use depends only on spatial location of the farm plot; it does not consider the amount of water required for the type of cultivated crop, time interval of water application and the size of each irrigated land sizes.

91 Figure 16: Water loss from motor pump

Lack of spare parts for water pumps and shortage of fuel is an issue. The lack of imported spare parts for motor pumps and treadle pumps are main causes for reduced efficiency in small-scale irrigation in the study area. Since there are many perennial rivers like Gumara and Rib in the study area, motor pump irrigation is used by many households. The main problems in motor pump irrigation are the frequent damage of the pump, lack of awareness of how to operate, cost of fuel and of the pump, and lack of credit. In the focus group discussion, farmers prefer the Robin motor pump, which is made in Japan. Farmers are more interested with this motor pump because of its durability and ease to operate. But, the office of agriculture supplied the Haowmax

92 motor pump on credit, which is made by Chain. The office of agriculture supplied this motor pump rather than Robin due to the higher capacity and the lower priced fuel used (Nafita). In contrast, the Robin used a more costly type of fuel (Benzene). But, farmers strongly complain about the Haowmax motor pump because it is easily damaged. The office of agriculture responds to this problem with discussions with the Chain Company to solve the difficulty. Pedal pumps are the other small-scale irrigation technology in the study area.

The pedal pumps used in the study area were imported from India. In the key informant interview and focus group discussion the main problems with pedal pumps are lack of spare parts and non-functionality due to long service. At present, most of these Indian-made pumps no longer have suction hose and delivery hose and have lost tightness at the joints (Figure 17).

Figure 17: Parts of pedal pump demonstrate loss of tightness

93 The pedal pumps are also highly demanded by farmers but supply in the market is limited. The office of agriculture has substituted an Ethiopian-made pedal pump, but users complain about its weight (and therefore higher labor requirement to operate) and its low water pumping capacity. At present, most of these pedal pumps are non-functional because of lack of spare parts, but farmers still have high demand for the Indian treadle pumps because of their simplicity to operate. Lack of market and marketing facility is another issue. Although not directly related to the functioning of irrigation systems per se, the market is considered one of the main problems in the study area. Cultivated vegetables using small-scale irrigations like onion, tomato, potato and the like are highly perishable and bulky crops, so an efficient marketing channel is necessary. However, the study area marketing system does not always facilitate outcomes desired by farmers. One reason is the similarity of products and marketing patterns; onion and tomato are the dominant crops, often harvested by farmers at the same time, which leads to a high availability and low prices during the main marketing period. Compounding this, because there is no efficient storage system in the study area, products quality deteriorates rapidly, which means that farmers must sell within a very short time, often at what they consider low prices. In some PAs, such as Kuhir Michael and Shina, which have water use associations, market risk is relatively low. The water use association has different teams such as marketing team, input supply team, conflict resolution team and the like. One of the main duties of marketing team is following the market for their products, and timing sales to increase returns to farmers. Farmers also perceive that market intermediaries are not pricing products fairly, which suggests reduced returns and less incentive to invest in the use of irrigation.

94 Shortage of surface water is another problem. There are rivers that have water for only some of the dry months. Their seasonality is unpredictable, varies depending on the climatic conditions each year. This seasonality nature of rivers in the area causes water shortage especially at the flowering period of irrigated crops. The traditional river diversion and treadle pump irrigation users are more seriously affected by this type of water shortage. Traditional river diversion irrigation practiced on such simple rivers that can easily dry up. In the study area, treadle pump irrigation users use shallow well as their water source. However, these shallow wells dry out during dry months of the area January, February and March. As shown in the picture (Figure 18), the shallow well dry at around the flowering and fruiting periods of cultivated crops. The above picture was taken on 29-January-2011.The cultivated crop on the irrigated farm plot was tomato.

Figure 18: Non-functional shallow well Crop diseases are another factor of importance. The study area is intensively cultivated with the same crops for long periods of time. Onion and tomato are repeatedly grown crops. In addition to the loss of productivity and fertility, this

95 cultivation strategy facilitates crop disease like root rot and cut warm. Imported inputs to control these problems, such as herbicides and pesticides, are costly for farmers to purchase. Therefore, diseases and pests can limit the economic benefits of small-scale irrigation activities in the study area. The price of imported inputs such as fertilizer, chemical and fuel has increased over time, in part due to depreciation of the ETB in world currency markets. One result is that the application of fertilizers on their farm plots is below the recommended levels. Chemicals like pesticides and herbicides are also costly to apply.

96 CHAPTER SIX

6 CONCLUSIONS AND RECOMMENDATIONS

The objective of this study was to assess the impact of small-scale irrigation on total income and poverty at the household level. The study was conducted in Fogera district, Upper Blue Nile basin, focusing on four small-scale irrigation types and five sample PAs. The selection of irrigation types and sample PAs are purposively on the basis of the irrigation potential. 6.1 Conclusions

Access to irrigation increases the opportunity for crop intensity and diversification, which increase cropping income. Irrigation is becoming a practice to increase total annual income for many households in the study area. In addition to their normal rainfed cultivation, irrigating households cultivate cash crops using small- scale irrigation. The main irrigated crops were onion, tomato, potato, maize, oat and vetch. Irrigated crops were selected due to good production potential, economic returns and ease of cultivation, respectively. Onion and rice were the major income source crops for irrigating and non-irrigating households, respectively. The main income sources of rural household in the study area were cropping, livestock and off-farm activities. Irrigating households have significantly larger mean annual income than non-irrigating households, but income also differs based on the type of irrigation used. The average annual total income of sample small-scale irrigating households was larger for concrete canal river/spring diversion and motor pump users compared to traditional river diversion and pedal. The findings of this study are consistent with previous ones such as Nhundu et al. (2010) and Hussain and Biltonen (2001).

97 Econometric analyses that control for other factors that influence household income indicate that accesses to small–scale irrigation increases mean household income significantly (about ETB 3,353 per year, or a 27 % increase over non- irrigating households). This is hypothesized to occur primarily through crop intensification and crop diversification, but this was not examined in this study. It is important to note that other factors (such as production input use) also had large effects on household income, and this study did not explore in detail the complementarities between irrigation access and other input use. The other objective of this study is to assess the impact of irrigation on the likelihood that a household was in poverty. The results indicate that irrigation development has a profound impact in alleviating poverty. The poverty analysis indicates that a much higher proportion of those who are poor are non-irrigating rather than irrigating households. Thus, the poverty prevalence in non-irrigating households is by far greater than in irrigating households. This suggests that irrigation has an important influence on rural poverty alleviation. Econometric analyses indicate that use of irrigation reduces the probability of a household being poor, controlling for other factors. Because small-scale irrigation increases mean annual household income, irrigating households have lower probability of being poor than non-irrigating households. From the extremely poor households, only 12% were irrigating households and the remaining 88 % did not irrigate. In the Logit model analysis, the estimated coefficient for dummy variable access to irrigation with odds of being poor over no-poor was negatively correlated and significant. The probability to being poor decreases if one has access to irrigation, other factors being constant. This suggests that irrigation has significant impact on rural poverty alleviation.

98 The study identified many problems in irrigation development through group discussion and key informant interviews. The main problems are lack of access to surface water, loss of water through seepage, problem of irrigation water distribution, lack of spare parts for water pumps, high cost of fuel for water pumps, lack of market transparency and marketing facilities, crop disease, and the perceived high cost of inputs. 6.2 Policy implications

This study has found that irrigation development helps to increase household income and reduces the incidence of poverty at the household level. Based on these findings as well as the outcomes of focus group discussions and key informant interviews, further development and refinement of small-scale irrigation systems appears merited. This, of course, raises the question about this might best be undertaken. Although a formal analysis of strategies for future irrigation development is beyond the scope of this research, following actions are suggested to facilitate future irrigation development.

1. Ensure irrigation water access, especially through groundwater

Access to irrigation has significant impact to promote total income and reduces the probability of households being poor. The main reason for non-irrigating households (85 percent) not to irrigate is lack of access to surface water. Moreover, the availability of surface water for only short periods in some areas causes loss of lower crop yield in dry periods, even for irrigating households. Therefore, in addition to surface water, the use of groundwater for small-scale irrigation is likely to be valuable for future irrigation development. Previous studies indicate that the study area, Lake Tana basin has high groundwater potential (Johnston et al., 2010). According to

99 Abedin et al. and Girum (2010), the Fogera alluvial deposited flood plains have considerable potential for shallow groundwater. Therefore it is appropriate to give attention to exploit both the surface and groundwater potential of area.

2. Renewed and improved the existed concrete canal river/spring diversion

The concrete canal river/spring diversion has great impact on poverty reduction by increasing household income. All these types of sample small-scale irrigation were constructed more than eight years ago in the study areas. They are now cracked and there is a water seepage problem. The irrigated land coverage is also small compared with the potential of the area. The Bureau of Agriculture and rural development is responsible for irrigation development cooperated with governmental and non- governmental organizations. In the interview and focus group discussions, concrete canal river diversion user households feel that the water use payment is very low. They want better service with better payment. Therefore, the Bureau of Agriculture and Rural Development should give more emphasis on the mending of the existing schemes and scaling up of the irrigation schemes for sustainable income growth and poverty reduction at household level.

3. Supply water pumps on the basis of users demand

Water pumps like the motor pump and pedal pump are used in the study area for irrigation. The supply of these irrigation technologies should be more closely aligned with the pump characteristics that farmers demand. For instance, the India pedal pump and Robin motor pump have high demand in the study area, but the access to such water pumps is limited. In particular, farmers do not have access to the India pedal pump in the market. Water pumps are profitable for user households. For

100 example, the annualized cost 11 of a motor pump is ETB 785 based on an initial cost of ETB 6,430 a useful life of 8 years, and a salvage value of ETB 2,000 and an interest rate of 5 %. The marginal return of irrigation access (although not specifically for the motor pump, which serves as one example of the technologies that could be used) is ETB 3,353 per year. The net margin return 12(NMR) of the motor pump is ETB 2,568 per year and it would require two and half years to pay back the cost of the pump. Pedal pumps were provided on credit for ETB 430 by office of agriculture, to be paid back within a year. Thus, concerned governmental and non-governmental organizations should give emphasis on the supply of water pumps in demand driven.

4. Strengthen education and training

Education has paramount impact on income improvement and poverty alleviation over time. The two econometric model analyses indicate that literacy has a large positive impact on household income and also reduces the likelihood that a household will be in poverty. These effects likely occur because illiterate households have difficulty accessing extension services and adoption recommendation.

11 Annualized cost of the motor pump is the amount of cost incurred in a year for the equipment which has useful life longer than one year. Monke and Pearson (1989) suggest:

Where

12 Net margin return above pump cost (NMR) = Additional annual revenue from irrigation – annual cost of the pump.

101 Education and training facilitates the effective communication between farmers and agricultural information providers like extension workers. Although the specific approach to be recommended requires further study, attention should be given to strengthen education and training for sustainable poverty alleviation in the long r

5. Improving the marketing system

Returns to irrigation are affected by the marketing channel, in part because the main irrigated crops (onion and tomato) are harvested at similar times by farmers and are perishable. An effective marketing system will facilitate irrigation adoption. Hence, the concerned bodies like governmental extension services, farmers‟ cooperatives and non-governmental market organizations should support the further development of the efficient marketing systems in the study area. This may include provision of marketing facilities, information provision and monitoring of costs and returns in the supply chain.

6. Ensure access for imported inputs

The important imported inputs are chemical fertilizers, herbicides and pesticides. In the study area, these inputs are used below the recommended level because of their high cost and shortage of supply. Access and proper utilization agricultural inputs are important for sustainable agricultural productivity and improvement. The government, cooperative organizations and private organizations should give attention on the supply of these inputs on time and in adequate amount. Further studies of the marginal returns to these inputs compared to their costs could facilitate development of approaches to increase input use, when appropriate.

102 7. Strengthening water use association

There are water use association in the three sample PAs, Shina, Quhir Michael and Bebekis. In focus group discussion and key informant interview, these associations have multiple purposes, such as equal distribution of water, conflict resolution, input supply, source of market information. Although not formally analyzed in this study, it appears that the water management and marketing functions undertaken by the associations can have a significant impact a on the current and future returns from irrigation use. That is, it is not simply access to irrigation water per se that increases household incomes, but an organizational and institutional structure that maintains adequate water access and provide information for improved management and marketing decisions. The concerned bodies should further study the impacts of, and encourage the establishment of additional water use associations to promote irrigation development. 6.3 Limitations and questions for future studies

This study focuses on the impact of irrigation on gross income and poverty reduction at household level. However, there are limitations that need further in-depth analysis, including the net income analysis of irrigation technologies using cost- benefit analysis. Another issue needing further research is the groundwater potential and the choice of irrigation technology types (small-scale, medium scale or large scale) and their impact on income and poverty. The impact of irrigation on actual livelihood change on the community like nutritional outcomes, and other indicators of household well-being need further study. Another issue to be addressed is that irrigated crops were cultivated and harvested by all farmers at the same time, which causes the perceived problems of marketing and post harvest handling in the study area.

103 Some of the key limitations relate to the ability to generalize from this thesis work. These limitations are: The study looked at one region of Ethiopia (with presumably greater potential for irrigation development) for a relatively short period of time. This can make it difficult to generalize about irrigation‟s impacts elsewhere in Ethiopia and in other developing countries. It is also a challenge to sort out the dynamic impacts of irrigation from a single-period study. As one example, the study treats livestock as exogenous (given in a particular year) but higher incomes from irrigation over time may allow additional livestock accumulation, which could further increase incomes. Another limitation is that the study considered gross income, rather than net income, and did not assess whether higher incomes resulted in improved outcomes such as nutritional status, health status or education. Future research questions that follow not so much from the study direct finings but from the study suggested actions are:

• Which types of irrigation are most cost-effective (cost-benefit analysis comparing net income to costs) under what conditions? • What strategies can address the shortages of parts? • What programs might best make pumps and parts available to farmers? • What educational efforts could improve returns from irrigation? • What are the complementarities between irrigation, and other inputs, including education?

• What organization for and activities of water use associations benefit farm households, and why? • To what extent would improvements in the marketing system increase farmer returns and facilitate irrigation adoption?

104 REFERENCES

Abdinasir I. 2000. Small holder dairy production and dairy technology adoption in the mixed farming system in Arsi highland, Ethiopia [Dissertation]. [Germany (Berlin)]: Humboldt University. Abedin ZM, Enyew A, Recent Achievements and Priorities in Irrigation Water Management Research in Ethiopia with Particular Reference to Amhara Region [Internet].[cited 2011 July 7]. MoWR/MoARD/USAID/IWMI

Workshop, Bahir Dar. Available from http://publications.iwmi.org. Abraham S. 2006. Evaluating the Effects of Irrigation on Poverty and Inequality in Banja Woreda Amhara Region [thesis], Addis Ababa University, Ethiopia. Aschalew D. 2009. Determinants of household participation in water source management: Achefer, Amhara region, Ethiopia [theses], Cornell University, Ithaca, NY, USA. Asfaw D. 2007. Scaling up Agricultural Water Development in Africa, the Case of

Ethiopia: Minister of Water Resources, Federal Republic of Ethiopia. Awulachew SB, Merrey D, Van Koopen B, and Kamara A. 2010. Roles, Constraints and Opportunities of Small-Scale Irrigation and Water Harvesting in Ethiopian agricultural Development: Assessment of Existing Situation. ILRI workshop; 2010 March 14-16; Addis Ababa, Ethiopia: International Water Management Institute (IWMI). Awulachew SB, Yilma AD, Loulseged M, Loiskandl W. 2007. Water Resources and

Irrigation Development in Ethiopia. Colombo, Sri Lanka: International water management institute; [Working paper 123].

105 Awulachew SB, Merrey JD. 2005. Assessment of Small Scale Irrigation and Water Harvesting in Ethiopian Agricultural Development [abstract]. International water management institute: Colombo, Sri Lanka. Baker J. 2000. Evaluating the Impact of Development Projects on Poverty: A Hand book for Practitioners. Washington, DC: The World Bank. Barket A, Khan S, Rahman M, Zeman S. 2002. Economic and social impact evaluation study on the rural electrification program in Bangladesh. NRECA

International Ltd. Partners with the Rural Electrification Board of Bangladesh and USAID for the Rural Power for Poverty Reduction [RPPR] Program [USAID CA 388-A-00-97-00040-00]; Dhaka. Bergh A, Nilsson T. 2010. The globalization and absolute poverty; a panel data study. Version to be presented at Nationell konferens i nationalekonomi, October 1-2, 2010, Lund University. Bigsten A, Kebede B, Shimeles A, Taddesse M. 2002. Growth and Poverty Reduction

in Ethiopia: Evidence from Household Panel Surveys, Department of Economics, Göteborg University; [Working Papers in Economics No 65]. BOWR. 2005. Amhara Regional Bureau of Water Resource report [Internet]. [cited 2011 April 8]. Available from: http://waterwiki.net. CSA (Central Statistics Agency). 2007. Summery and statistical report of 2007 population and housing census of Ethiopia: Population size by age and sex [internet][cited 2011 Jan 18]. Available from: http://www.csa.gov.et.

CSA (Central Statistics Agency). 2004. Welfare Monitoring Survey 2004 Analytical Report [Internet]. [cited 2011 June 24], Addis Ababa; Available from: http://www.csa.gov.et. Cohen J. 1988. Statistical power analysis for the behavioral Sciences. 2nd ed. New York: Academic press.

106 Dauda TO, Asiribo OE, Akinbode SO, Saka JO. 2009. An assessment of the roles of irrigation farming in the millennium development goals. African journal of Agricultural Research; 4 (5): 445- 450. Dearcon S. 2001. Economic Reform, Growth and the Poor: Evidence from Rural Ethiopia. Centre for the Study of African Economies (CSAE), Oxford University; [Working Paper series 2001- 8]. Dearcon S, Krishnan P. 1998. Changes in Poverty in Rural Ethiopia 1989-1995:

Measurement, Robustness Test and Decomposition; WPS/98-7: Center for the study of African Economics, institute of statistics and economics; Oxford University. Deaton A. 1997. The Analysis of Household Surveys: A Micro Econometric Approach to Development Policy. Baltimore: Johns Hopkins University Press for the World Bank [Internet] [cited 2011 July 12]; Available from: http://imagebank.worldbank.org.

Dercon S. 1997. Poverty and deprivation in Ethiopia. Analysis of 1995/96 HBS/WMS Ethiopia Poverty Profile. Center for the study of African economies, department of economics and Jesus College, Oxford University. Dupriez H, De Leener. 2002. Land use and life: Ways of water, Run off, irrigation and drainage; Tropical Handbook. Publisher: CTA and TERRES ET VTE, Netherlands; ISBN: 2-87105-011-2. FAO. 2010. Agricultural populations and households, agri-gender statistics Toolkit:

data items, Accra, Ghana. FAO. 1997. Irrigation technology transfer in support of food security proceeding of a sub regional workshop; 1997 April 14-17; Harare, Zimbabwe: water report 14. FAO.1994. Irrigation, drainage and salinity. An international source book FAO/ UNESCO. No 7:163-171.Sri Lanka.

107 Faurès JM. 2006. Securing food for a growing population: the water scarcity challenge; FAO Land and Water Development Division, Rome, Italy. Foster J, Greer J, and Thorbecke E.1984. A Class of Decomposable Poverty Measures. Econometrica; 52 (3):761-766. Gebremedhin B. and Peden D. 2002. Policies and Institutions to Enhance the Impact of Irrigation Development in Mixed Crop-Livestock [Internet].[cited2011 March 13]. MoWR/EARO/IWMI/ILRI Workshop; Addis Ababa, Ethiopia.

Available from: http://publications.iwmi.org. Girum A. 2010. Ground water and surface water interaction in Lake Tana sub-basin using Isotope and Geochemical approach [theses] [internet]. Addis Ababa University [cited 2011 June 14]. Available from: http://etd.aau.edu.et. Goodswill M, Dawit K, Dejene T. 2007. A comparative analysis of rain fed and irrigated agricultural production in Ethiopia. Irrigation and drainage systems journal; 21(1): 35-44, Springer, Netherlands.

Greene W. 2002. Econometric analysis, 5th Ed., Prentice Hall, Upper Saddle River, New Jersey 07458. Gujarati M. 2004. Basic Econometrics, 4th Ed. Publisher: McGraw−Hill Companies. United State Military academy, West point, New York. Habitamu, T. 2009. Payment for environmental service to enhance resource use efficiency and labor force participation in managing and maintain irrigation infrastructure, the case of Upper Blue Nile basin[thesis], Cornell University,

Ithaca, NY, USA. Hagos F, Holden S. 2003. Rural Household Poverty Dynamics in Northern Ethiopia 1997-2000. Paper for CPRC Conference, Manchester. Haile T. 2008. Impact of irrigation development on poverty reduction in Northern Ethiopia (dissertation), National University of Ireland, Cork.

108 Helland J. 1999. Land alienation in Borena: Some land tenure issues in a pastoral context of Ethiopia, Eastern Africa Social Science Research Review, Chr. Michelsen Institute, Bergen, Norway. Holloway G, Nicholson C, Delgado C, Staal S, Ehui S. 2004. A revised Tobit procedure for mitigating bias in the presence of non-zero censoring with an application to milk-market participation in the Ethiopian highlands. Agricultural Economics; 31(2004) 97-106.

Hussain I, Biltonen E. 2001. Irrigation against Rural Poverty: An Overview of Issues and Pro-Poor Intervention Strategies in Irrigated Agriculture in Asia; Proceedings of National Workshops on Pro-Poor Intervention Strategies in Irrigated Agriculture in Asia Bangladesh, China, India, Indonesia, Pakistan, and Vietnam. 2001 August; International Water Management Institute; Colombo, Sri Lanka. Hussain I, Hanjra MA. 2004. Irrigation and poverty alleviation: Review of the

empirical evidence, IWMI, Irrig. and Drain. 53: 1–15. Hussain M. 2006. Impact of Small Scale Irrigation Schemes on Poverty Alleviation in Marginal Areas of Punjab, Pakistan. Publishers: Euro Journals Publishing, Inc. 2006. International Research Journal of Finance and Economics, ISSN 1450- 2887 Issue 6 (2006). IPMS (Improving Productivity and Market Success). 2005. Fogera Woreda Pilot Learning site Diagnosis and Program Design. Technical report. Publisher:

ILRI: Nairobi, Kenya. IMF (International Monetary Fund). 2011. World Economic Outlook Database [Internet]. International monetary fund [cited 2011 July 23]. Available from: http://www.imf.org.

109 John M. 2002. Dependency ratio, population, standard grade Geography [Internet]. [cited 2011 June 12]. Available from: http://www.scalloway.org.uk/popu. Johnston R. and McCartney M. 2010. Inventory of water storage types in the Blue Nile and Volta river basins. Colombo, Sri Lanka: International Water Management Institute: 48p [IWMI Working Paper 140]. Kakwani N. 1993. Statistical inference in the measurement of poverty. Review of Economics and statistics; 75: 632-640.

Kamara A, Van Koopen B, Magin L. 2001. Economic viability of small-scale irrigation systems in the context of state withdrawal: the Arabie Scheme in the Northern Province of South Africa. 2nd WARSFA/WaterNet Symposium: Integrated Water Resources Management: Theory, Practice, Cases; Cape Town, 30 – 31 October 2001. International Water Management Institute (IWMI), Pretoria, South Africa. Kobets M. 2004. Organic farming in the context of sustainable development.

Agricultural policy for human development project. USAID & UNDP, No.5; 2004 May, Ukraine. Loayza N, Soto R. 2002. Economic Growth: Sources, Trends, and Cycles, Santiago, Central Bank of Chile, Chile. Maddison A .1970. Economic progress and Policy in Developing countries, New York: W.W.Norton and Co., 1970, pp. 34. McClelland A. 2000. No child…: Child poverty in Australia, summarised by Bette

Moore in Child poverty: the facts, Brotherhood of St Laurence, Fitzroy, Vic. [Internet]. [cited 2011 July 2]. Available from: http://catalogue.nla.gov.au/Record/301758.

110 MEDC (Ministry of Economic Development and Cooperation). 1999. Poverty situation in Ethiopia, Welfare Monitoring Unit [Internet]. [cited 2011 June 19]. Available from: http://www.socialwatch.org/node/10741. MOA .2010. Ministry of Agriculture and rural development; Fogera district office of Agriculture. Werota Town, Ethiopia. MOA (Ministry of Agriculture). 2005. Fogera Woreda Pilot Learning Site Diagnosis and Program Design. In: ILRI Fogera PLS planning workshop program; 2004

October 1- 2; Woreta Town. MOFED.2010.The Federal Democratic Republic of Ethiopia, Growth and Transformation Plan (GTP) 2010/11-2014/15, Draft. 2010 September, Addis Ababa. [Internet]. [cited 2011 April 10]. Available from: http://www.ethiopians.com/Ethiopia_GTP_2015.pdf MOFED .2010. Trends and prospects for meeting MDGs by 2015, millennium development goal report. [Internet]. [cited 2011 July 4]. Addis Ababa,

Ethiopia. Available from: http//www.dagethiopia.org. MOFED. 2007. Annual Report on Macroeconomic Developments for 2006/07, Addis Ababa, Ethiopia. MOFED. 2000. Sustainable Development and Poverty Reduction Program. EFA 2000 Assessment - Country Report–Addis Ababa, Ethiopia Morgan A, Leech L, Gloeckner W, Barrett C. 2004. SPSS for introductory statistics use and interpretation. 2nd ed. Publishers: Lawrence Erlbaum Associates,

Mahwah, New Jersey. Nhundu, K, Gwata , C, Mushunje, A.2010. Impacts of Zimbabwe European Union micro-project programme (Zim/Eu MPP) in funding smallholder irrigation projects on food security and income levels:A case study of Mopane irrigation

111 scheme in Zvishavane, Midlands province, Zimbabwe. African Journal of Agricultural Research Vol. 5(14), pp. 1759-1771. Rao S, Richard J. 2006. Introduction to Biostatistics and research methods. 4th Ed. Prentice-Hall of India Private limited, New Delhi. Ravallion, M. 2005. Evaluating anti-poverty programs.Washington DC, World Bank, Policy Research Working Paper WPS3625. Roemer M, Gugerty, KM.1997. Does economic growth reduce poverty? Technical

Paper [Internet]. [cited 2011 Feb 16]; Harvard Institute for International Development, March 1997. Available from: http://pdf.usaid.gov. Saleth, RM, Samad, M, Molden, D and Hussain, I .2003. Water, poverty and gender an overview of issues and policies, Water policy, Vol 5: No 5/6. Schmidt, FL. 1971. The relative efficiency of regression in simple unit predictor weights in applied differential psychology. Michigan state University Educational and Psychological Measurement, Vol. 31(3): 699-714; SAGE

1971 Oct-1. Schilfgaarde VJ. 1994. “Irrigation, a Blessing or a Curse”; Agricultural Water Management. 25: 203-219. Schreiner M, Chen S. 2009. A Simple Poverty Scorecard for Ethiopia [internet].[cited 2011 May 29] CSA, April 2009, Addis Ababa. Available from: http://www.microfinance.com/#Ethiopia. Sen AK. 1976. Poverty: an ordinary approach to measurement. Econometrica, Vol.

44 (2): 219–59. Shanan, L. 1987. The impact of irrigation. In: Wolman MG, Fourner FGA, editors. Land Transformation in Agriculture. Publishers: Wiley, J & Sons Ltd. Simtowe F, Shiferaw B, Solomon A, Abate T, Muricho G. 2010. Socio-economic Assessment of Legume Production, Farmer Technology Choice, Market

112 Linkages, Institutions and Poverty in Rural Ethiopia [Internet]. International Crops Research Institute for the Semi-Arid Tropics (ICRISAT); Nairobi, Kenya. Available from: http://www.icrisat.org. Topcu Y, Uzundumlu SA, Guler OI. 2010. Economic effectiveness analyses of potato farms: The case of Erzurum Province, Turkey, Scientific Research and Essays Journal Vol. 5 (17): 2560-2566. Townsend P. 1993. The International Analysis of Poverty. Publishers: Harvester

Wheatsheaf, Journal of European Social Policy Pioch Vol. 4 (4): 305. UN.1995. The Copenhagen Declaration and Program of Action: Report of the World Summit for Social Development. A/CONF.166/9; 1995 March 6-12; Copenhagen, Denmark. UNDP. 2007. Globalization, Agriculture and the Least Developed Countries [Internet]. Issues Paper, Istanbul. 9-11 July 2007. Available from: http://www.unohrlls.org.

UNDP. 2006. What is poverty? Concepts and measures [internet] Poverty in focus; United Nation Development Programme International poverty center [cited 2011 June 12]. SBS-Ed, BNDES,10 andar 700076-900 Brasilia DF Brazil. Available from: http//www,undp-poyertycentre.org. WAE (Water Aid Ethiopia). 2008. Water aid; Think local, act local II. Impediments to effective financing of sanitation services in Ethiopia. The case of three local governments; WASH report [Internet]. [Cited 2011 June 5]. Available from:

http://www.wateraid.org. Waganew A. 2004. Socio economic and Environmental impact assessment of community based small-scale irrigation in the upper basin [internet].[cited 2011 March 25] . Addis Ababa University, Ethiopia. Available from: http://www.iwmi.cgiar.org.

113 Widtose, JA. 2001. Irrigation Practices. Publisher: Updesh Purhohit for Agrobios; India, Jodhpur. ISBN NO: 81-7754-110-2. Woldehanna T, Alemu T. 2002. Poverty profile of Ethiopia, 1999/2000. A report for Welfare monitoring Unit (WMU), Ministry of Finance and Economic Development (MOFED), Federal Democratic Republic of Ethiopia (FDRE). Economics Department, Addis Ababa, Ethiopia.World Economic Outlook Database-April 2011, International Monetary Fund.

Zewdie M, Moti J,Ascimelis G.2007.Assessment of Wendo Wesha irrigation scheme in Awassa Zuria. Proceddings of research project completion workshop; 2007 Feb 1-2; Addis Abaaba, Ethiopia. Zhou Y, Zhang Y, Abbaspour CK, Yang H, Mosler JH. 2009. Economic impacts on farm households due to water reallocation in China‟s Chaobai Watershed [Internet]. [cited 2011 June 21]. Agricultural Water Management 96 (2009) 883–891. Available from: http://www.syngentafoundation.org.

114 APPENDIX- A: TABLES OF CROP VALUES AND CONVERSION FACTORS.

Table A1: The price of crops and vegetables in 2009/2010 Crop type Nov. Dec. Jan. Feb. Mar. Apri. May Jun Ave. price Teff 694 620 583 606 568 610 525 798 626 Barely 315 316 285 307.5 302 314 298 399 317 Wheat 400 476 477 491 478 485 467 429 463 Finger m. 300 332 320 340 320 360 345 400 340 Maize 300 295 256 307.5 293 259 267 331 289 Sorghum 338 332 322 308 278 275 263 255 296 Rice 675 503 622 610.5 596 652 659 716 629 Beans 397 458 339 473.5 438 440 434 417 425 Peas 415 432 405 433.5 472 482 480 454 447 Chickpea 385 372 343 402.5 390 409 404 375 385 Vetch 250 220 300 300 380 400 390 400 330 Lentil 680 630 659 682 679 738 737 736 693 Nug 471 589 591 559.5 516 581 577 556 555 Oat 400 400 500 550 540 680 710 800 573 Rapeseed 285 270 239 261 223 298 306 373 282 Fennfrek 560 599 810 635 648 678 705 688 666 Pepper 1365 1555 1585 1636 1770 1749 1769 1471 1613 Tomato 421 301 275 202.5 103 111 122 360 237 Onion 558 503 516 500 428 390 327 558 473 Garlic 610 606 602 616.5 618 610 617 545 603 Potato 244 205 229 305 165 197 214 220 222 Kosta 242 208 178 219 206 230 215 205 213 Cabbage 283 265 241 231.5 225 230 231 236 243 Honey 1840 1866 2000 2173 3200 2240 2184 2202 2213 Skin 10.5 10 10 11 10 11.5 11.5 12 11 Milk 356 356 336 332.5 348 350 349 436 358 Butter 3505 3708 3760 4025 4270 4560 5049 4680 4195 Source: Fogera District Agricultural and Rural Development office (2010)

115 Table A2: Conversion factor for Adult equivalent Years of age Men Women 0-1 0.33 0.33 1-2 0.46 0.46 2-3 0.54 0.54 3-5 0.62 0.62 5-7 0.74 0.70 7-10 0.84 0.72 10-12 0.88 0.78 12-14 0.96 0.84 16-18 1.14 0.86 18-30 1.04 0.80 30-60 1.00 0.82 60 plus 0.84 0.74 Source: Dercon and Krishnan (1998)

Table A3: Conversion factor for Tropical Livestock Unit (TLU) Livestock Type TLU Ox 1.10 Cow 1.0 Heifer 0.50 Bull 0.6 Calves 0.20 Sheep 0.01 Goat 0.09 Donkey 0.5 Horse 0.80 Mule 0.7 Poultry 0.01 Source: Abdinasir, Ibrahim (2000)

116 APPENDIX- B: SURVEY QUESTIONNAIRE CORNELL-BAHIRDAR UNVERSITIES MPS PROGRAME IN INTEGRATED WATERSHED MANAGEMENT AND WATER SUPPLY

The questionnaire is prepared to undertake a study on the effect of selected small-scale irrigation on poverty reduction at house hold level. The purpose of the questionnaire is to gather information on irrigating and non-irrigating household‟s socio-economic, agricultural and non-agricultural activities, access for services and other important information. Dear respondents, the result of this study will help different stakeholders and policy makers to make appropriate measures on irrigation development in the future. Your responses are confidential. Therefore, you are kindly requested to provide genuine responses. Thank you for your time and cooperation!

Identification a. Code ______b. Peasant Association name (PA) ______c. Village______

1. Demographic Characteristics of the Household 1. Household head name: ______2. Age of the household head ______

2. 1. Age, sex and education of all household members including permanently employed laborer, husband and wife. Age Male Female Educatio Age Male Female Education n/Grade /Grade 0-1 12-14 1-2 16-18 2-3 18-30 3-5 30-60 5-7 60-64 7-10 above 64 10-12 Total

1. Sex of the household head ______1= Male 2 = Female 2. Education of the household head ______0 = illiterate 1 = Read and writ 1 = Elementary complete 3 = Junior complete 4 = High school complete & above

117 3. Education of the spouse______0 = illiterate 1 = Read and writ 2 = Elementary complete 3 = Junior complete 4 = High school complete & above

4. Religion______1 = Christian 2 = Muslim 3 = Traditional 4 = other

7. Total family numbers of the household______8. Do you face labor shortage? 0 = No 2 = Yes 9. Type of your house 0 = Grass roofed 1 = Corrugated iron roofed house 2. Resource endowments 2.1. Farmland and other assets No Size in Owned? How much it timed 1. Yes cost now Land Type Other assets 2. No (Value) in ETB 1 Currently 5 Agricultural farmed assets a. Own a. Motor pump b. Rented in b. Pedal pump c. Shared in c. Well d. Shared out Rented Out 6 Business assets 2 Fallow land a. Mill 3 Pasture land b. Shop 4 Woodlots c. Carts

118 3. Crop production 3.1. Crops production in rainfed, from Nov 2009-June2010 N Type of Crops Plot Size Total Consumed Sold o (Timade) production at home Amtt Value Ave. (Kg) (Kg) (kg) (ETB) price

1 Teff 2 Maize 3 Wheat 4 Barely 5 Sorghum 6 Finger millet 7 Rice 8 Beans 9 Peas 10 Chickpea 11 Vetch 12 Lentil 13 Noug 14 Sesame 15 Rapeseed 16 Linseed 17 Garlic 18 Fennfreek 19 Other crops

3.2. Vegetable, Fruit, and woodlot production in rainfed No Type of Crop Plot Size Total Consumed Sold grown (Timade) production at home Amount Value Ave. (Kg) (Kg) (kg) (birr) price Vegetable 1 Tomato 2 Potato 3 Pepper 4 Onion 5 Cabbage 6 Kosta 7 Others Fruit 1 Avocado 2 Orange 3 Lemon 4 Others

Woodlots 1 Eucalyptus 2 Gesho

119 1. Are you irrigation user? 0 = No 1 = Yes a. If the answer is No, what was the reasons not using irrigation? 1 = No farmland in surface water access 2 = No awareness about it 3 = Sufficient rain and moisture 4 = others

2. Which small-scale irrigation type do you use? 1 = Modern micro dam 2 = Traditional river diversion 3 = Motor pump 4 = Treadle pump 5 = others specify______

3. How long do you use irrigation farming? ______years 0 = No irrigation farming before 1 = 0-2 years 2 = 2-4 years 3 = 4-6 years 4 = 6-8 years 5 = 8 and above years

4. What are the major problems you face/observe in your irrigation farming ______and what actions do you take to solve the problem ______5. The major problems encountered in small-scale irrigation schemes you use per crop season. 5.1. a. In the upgraded small-scale irrigation, how much payment /one crop season, if any, ______b. What do you feel about the payment, if you feel the cost is high how much is the reasonable price ______c. What are the major problems encountered in the use of upgraded small-scale irrigation, what is your opinion about the solution______

5.2. a. What are the major problems in using Motor pump for small-scale irrigation? What is your opinion about the solution ______5.3. a. What are the major problems in using Pedal pump for small-scale irrigation? What is your opinion about the solution______

120 5.4. Well: a. For what purpose the well water used? If the water was used for irrigation how the water drawn out from the well ______b. What is the depth of the well in meter? ______, for how long do you use the well once constructed ______, if there are a frequent failure what was the main reasons ______c. On average for how many months the water available in the well within a year______? 1. How much of your land is used by irrigation______

2. Have you cultivated the total of your irrigable land during the last dry season 0 = No 1 = Yes If your answer to question is No, what are the reasons? (Circle the answers) 1 = Shortage of family labor 2 = Lack of seed 3 = Lack of oxen 4 = Enough production rainfed 5 = Lack of credit 6 = Others specify

8. Have you rented in or rented out any cultivable land 0 = No 1 = Yes If yes, how much is the cost per Timad per crop season for: a) Irrigable land______b) Non-irrigable land______9. What is the source of water for your irrigation?

1 = River

2 = Well 3 = Spring 4 = Pond 5 = Others 10. What is the distance between the sources of water to your irrigated land? ______11. What is the system of water sharing with others? ______

12. Is the amount of water is enough to irrigate your land? 0 = No

121 1 = Yes

If your answer is no, what mechanisms do you suggest to solve the scarcity of water______13. Is there irrigation water use association around your area?

0 = No 1 = Yes

If your answer is yes, what are the benefits obtained from the association and your role______4. Production in the irrigation agriculture: 4.1 Crop production in irrigation (Nov 2009- June 2010) No Type of Plot Size Total Consumed Sold Crops (Timad) production at home Amt Value Ave. (Kg) (Kg) (kg) (birr) price 1 Teff 2 Maize 3 Wheat 4 Barely 5 Sorghum 6 Finger millet 7 Rice 8 Beans 9 Peas 10 Chickpea 11 Vetch 12 Lentil 13 Noug 14 Sesame 15 Rapeseed 16 Linseed 17 Garlic 18 Fennfreek 19 Other crops

122 4.2. Vegetable, Fruit, and woodlot production in irrigation No Type of Crop Plot Size Total Consumed Sold grown (Timade) productio at home Amt Value Av. n (Kg) (Kg) (kg) (birr) Price Vegetable 1 Tomato 2 Potato 3 Pepper 4 Onion 5 Cabbage 6 Kosta 7 Others

Fruit 1 Avocado 2 Orange 3 Lemon 4 Others Woodlots 1 Eucalyptus 2 Gesho 3 Others 16. Why do you select the above type of Vegetable /crops for your irrigation farming?

1 = Better price 2 = Good production 3 = High disease tolerance

4 = Easiest to cultivate 5 = Seed availability 6 = Other

17. Did you get reasonable price for your produce at the place you used to sell to?

0 = No 1 = Yes

18. What help do you need from the government or any organization on your irrigation farming______

123 5. LIVESTOCK PRODUCTION (Nov. 2009 - June2010) No Type of No of If there is any sold Rema animal animals animal rk Total How much if you Sold Income owned want to Sell (Nov- amount gained June) (Birr) 1 Cow 2 Bull 3 Heifer 4 Calf 5 Ox 6 Mules 7 Horse 8 Donkey 9 Camel 10 Goat 11 Sheep 12 Poultry 13 Bee colony

5.2 Livestock output N Commodity type Amount Consumed Sold Remark o produced (liter, Kg, no) (Birr) (liter, Kg, no ) 1. Dairy output (Dec.1- 15, 2010) 1.1 fluid milk 1.2 Butter 1.3 Yoghurt 1.4 Cheese 1.5 Others 2. Poultry (December1- 30, 2010) 2.1 Egg 2.2 Chicken 3. Honey bee (from the last one harvest season) 3.1 Honey 3.2 Bees wax 3.3 Bee colony 4. Animal by-products Hide and skin Manure/Dung

124 6. Credit, input and extension service supports in production (2009/10) 6.1 Credit support service 1. Did you need credit for the production of your agricultural products? 0 = No 1 = Yes 2. If yes, did you have access to credit for the production of the Commodities? 0 = No 1 = Yes 3. What is the source of your Credit?

1 = Banks 2 = Friends/relatives 3 = Traders

4 = Microfinance 4. Is credit timely and adequately available for agricultural commodities development? 0 = No 1 = Yes 6.2 Extension services 1. Do you receive any sort of extension services available in your locality? 0 = No 1 = Yes 2. If yes, did you gain any knowledge from the extension agents that could help you to do things differently on the specific commodities? 0 = No 1 = Yes 3 = If no, specify your reason______6.3 Access to other Services 1. Do you get market information about prices and demand conditions of agricultural inputs and out puts? 0 = No 1 = Yes, if yes indicate the source of information______

1. Do you use input for the last one cropping season? 0 = No 1 = yes 3. How far do you travel to get local market ______km? 4. How far do you travel to get to the nearest school in your vicinity? ______Km 5. How far do you travel to get the services of all weather roads? ______km

125 7. NON-FARM AND OFF – FARM INCOME

7.1. Do any member of your family has involved last year on non/off farm activities? 0 = No 1 = Yes, if the answer is yes, in which one from the next table. No Off-Farm/Non-farm activities Amount Remark (birr) 1 Working on other‟ farm 2 Rent from motor Pump, Pedal pump, Rent from draft power, and others 3 Daily laborer on construction or other non- farm activities 4 Self employment in manufacturing e.g. Artisan (blacksmith, weaving, pottery, ) 5 Sales of wood (Charcoal) 6 Sales of local drink 7 Transporting using Carts 8 Hair dressing 9 Sales of stone/sand 10 Salary from temporary or permanent employment Remittance Trade Aid Any comments: ______

126