Determinants of Poverty in Rural Households (The Case of Damot Gale Woreda: Wolayta Zone) a Household Level Analysis

DETERMINANTS OF POVERTY IN RURAL HOUSEHOLDS (THE CASE OF DAMOT GALE WOREDA: WOLAYTA ZONE) A HOUSEHOLD LEVEL ANALYSIS

MA THESIS

ZEGEYE PAULOS BORKO

JUNE, 2016 ARBA MINCH, ETHIOPIA

DETERMINANTS OF POVERTY IN RURAL HOUSEHOLDS (THE CASE OF DAMOT GALE WOREDA: WOLAYTA ZONE) A HOUSEHOLD LEVEL ANALYSIS

ZEGEYE PAULOS BORKO

A THESIS SUBMITTED

TO THE DEPARTMENT OF ECONOMICS,

COLLEGE OF BUSINESS AND ECONOMICS, SCHOOL OF GRADUATE STUDIES

ARBA MINCH UNIVERSITY FOR THE REQUIREMENT OF MASTER OF ART DEGREE IN DEVELOPMENT ECONOMICS

JUNE, 2016 ARBA MINCH, ETHIOPIA

DECLARATION

I hereby declare this MA thesis is my original work and has not been presented for a degree in any other university, and all sources of materials used for this thesis have been duly acknowledged.

Name: Zegeye Paulos Borko

Signature: ______

Date:______

SCHOOL OF GRADUATE STUDIES

ARBA MINCH UNIVERSITY

ADVISORS’ THESIS SUBMISSION APPROVAL SHEET

This is to certify that the thesis entitled “Determinants of poverty in rural households: The case of Damot Gale woreda, Wolayta zone” submitted for the requirements for the degree of Master’s with specialization in Development Economics, the Graduate Program of the Department/School of Economics, and has been carried out by Zegeye Paulos Id. No RMSc /006/07, under our supervision. Therefore, we recommend that the student has fulfilled the requirements and hence hereby can submit the thesis to the department for defense.

Name of Principal advisor Signature Date

Name of co-advisor Signature Date

iii

SCHOOL OF GRADUATE STUDIES

ARBA MINCH UNIVERSITY

EXAMINERS’ THESIS APPROVAL SHEET

We, the undersigned, members of the Board of Examiners of the final open defense by Zegeye Paulos have read and evaluated his thesis entitled “Determinants of poverty in rural households: The Case of Damot Gale woreda”, and examined the candidate’s oral presentation. This is, therefore, to certify that the thesis has been accepted in partial fulfillment of the requirements for the degree of Master of Art in Development Economics.

______

Name of Chairperson Signature Date ______

Name of Principal Advisor Signature Date ______

Name of Co - Advisor Signature Date

______

Name of External Examiner Signature Date

______

Name of Internal Examiner Signature Date ______

SGS Approval Signature Date

ACKNOWLEDGMENTS

The completion of this research paper and my academic success is not solely due to my effort. God and many people have helped me in one way or another. Above all, I would like to praise God, the Almighty, for the inner peace, health and strength he gave me and the help of whom was too immense to reach this day. My Special thanks then goes to my advisor, Tora Abebe (PhD), for his unwavering support and kindness, thoughtful insights, constant guidance, and constructive comments. My gratitude is also due to my co– advisor, Mr. Mohamed Beshir, for unreserved expert advice in the preparation of this paper. I would like to thank my office, Wolayta Zone Urban Development Department, for sponsoring my graduate studies.

I have special obligation to extend my heartfelt thanks to Mr. Tadele W/Michael, the vice head of SNNPR State Urban and Housing Development Bureau, who paved the way for my graduate studies when he was head of Wolayta Zone Urban Development Department.

I am grateful to household heads included in the sample, focus group discussions and key informant interviews for their kind and genuine response. I am also indebted to thank Development agents who supported me in collection of data.

Last, but not least, I would like to thank my wife Misrak Mantire for her understanding and love during the graduate study years. Her support and encouragement was in the end what made this thesis possible. My mother Marta Tantu, receive my deepest gratitude and love for her the many years of support during my undergraduate studies that provided the foundation for this work.

ACRONYMS OBS Observations CBN Cost of Basic Needs CSA Central Statistical Agency DGW Damot Gale woreda DGWARO Damot Gale woreda Rural and Agricultural office FAO Food and Agricultural Organization FDRE Federal Democratic Republic of Ethiopia FEI Food Energy Intake GDP Gross Domestic product HH Household Head IFAD International Fund for Agricultural Development IMF International Monetary Found LDCs Least Developing Countries MDG Millennium Development Goal MEDaC Ministry of Economic Development and Cooperation MFI Micro Finance Institution MoFED Ministry of Finance and Economy Development NGO Non Governmental Organization OECD Organization for Economic Co-Operation and Development OLS Ordinary Least Square TLU Total Livestock Unit PL Poverty Line PPP Purchasing Power Parity SNNPR Southern Nation Nationalities Region UN United Nation WBG World Bank Group WFP World Food Program

Table of Contents……………………..………………………….…Page No DECLARATION ...... I

ADVISORS’ THESIS SUBMISSION APPROVAL SHEET ...... iii

EXAMINERS’ THESIS APPROVAL SHEET ...... iv

ACKNOWLEDGMENTS ...... v

ACRONYMS ...... vi

LIST OF TABLES ...... viii

LIST OF FIGURES ...... ix

LIST OF APPENDICES ...... x

ABSTRACT...... xi

CHAPTER ONE ...... 1

1. INTRODUCTION ...... 1

1.1 Back Ground of the Study ...... 1

1.2 Statement of the Problem ...... 4

1.3 Objectives of the Study ...... 6

1.3.1 General Objectives ...... 6

1.3.2 Specific Objective ...... 6

1.4 Hypothesis of the Study ...... 7

1.5 Significance of the Study ...... 7

1.6 Scope and Limitation of the Study ...... 8

1.7 Organization of the Paper ...... 8

CHAPTER TWO ...... 9

2. REVIEW OF THE RELATED LITERATURE ...... 9

2.1 Theoretical Literature ...... 9

2.1.1 Definition of Poverty ...... 9

2.1.2 The Conceptualization of Poverty ...... 10

2.1.3. Theories of Poverty ...... 13

vii

2.1.4 Measuring Poverty ...... 16

2.3 Research Gap ...... 28

CHAPTER THREE ...... 30

3. METHODOLOGY ...... 30

3.1 The Study Area ...... 30

3.2. Data Type and Source ...... 31

3.3 Sample Size Determination ...... 32

3.4. Sampling Technique ...... 33

3.5 Approaches to Measuring Poverty and Unit of Analysis ...... 33

3.6. Data Collection Techniques and Instruments ...... 35

3.7. Method of Data Analysis ...... 35

3.7. Empirical Model ...... 36

3.7.1. Binary Logit Model ...... 36

3.8. Variable Description and Their Expected Sign ...... 39

CHAPTER FOUR ...... 44

4. RESULTS AND DISCUSSION ...... 44

4.1. Computing Poverty Line ...... 44

4.2 The Magnitude and Measure of Poverty ...... 44

4.2.1 Descriptive Analysis ...... 47

4.2.2 Econometric Analysis ...... 52

CHAPTER FIVE ...... 61

5. CONCLUSION AND POLICY IMPLICATIONS ...... 61

5.1. CONCLUSION ...... 61

5.2. POLICY IMPLICATION ...... 62

REFERENCE ...... 64

APPENDICES ...... 69

viii

LIST OF TABLES 1. Table 3.1 Sample Size of the Surveyed Kebeles------33 2. Table 3.2 Variables Description and their Expected Sign------43 3. Table 4.1 FGT Measure of Poverty in Four Surveyed Kebeles------46 4. Table 4.2 Annual Households’ Consumption Expenditure of the Study Area------47 5. Table 4.3 Households’ Age, Sex and their Economic Activity status------48 6. Table 4.4 Family Size and Poverty ------49 7. Table 4.5 Household Head Sex and Poverty ------50 8. Table 4.6 Cultivated Land Size and poverty------51 9. Table 4.7 Logit Model Maximum likelihood estimation ------54 10. Table 4.9 Marginal effect for Logit Regression------56

viii

LIST OF FIGURES

Figure 3.1 Map of the Study Area------31

Figure 4.1 Family Sizes and Consumption Distribution around Poverty Line------49

Figure 4.2 Household Ages and Consumption Distribution around Poverty Line------50 Figure-4.3 Ownership of Oxen and Poverty Status ------52

LIST OF APPENDICES

APPENDIX-1 Questionnaire for the Study------69

APPENDIX-2 Poverty Line Calculation table------77

APPENDIX-3 Adult Equivalence Conversion Factor------78

APPENDIX-4 Tropical Livestock Unit Conversion Factor------78

APPENDIX-5 Model Specification Test------79

APPENDIX-6 Multi-collinearity test for Continuous Variables------79

APPENDIX-7 Goodness-of-fit-test------79

APPENDIX-8 Multi-collinearity test for Discrete Variables------80

APPENDIX-9 Association between being Poor and Households Family Size------80

APPENDIX-10 Robust Logistic Regression Result------81

ABSTRACT

The study was carried out at Damot Gale Woreda of Wolayta Zone in Southern Nation Nationalities Regional State with the main objectives to describe correlates or determinants of rural poverty in the study area. In order to attain this objective the study made use of cross-sectional household survey data collected from 235 sample households .The data collected were analyzed and discussed applying FGT measure of poverty i.e. poverty head count index, poverty gap and severity. Using cost of basic needs approach; the study found that total poverty line (food and non food poverty line) of the study area was about 3612.151birr per year per adult equivalent consumption. Using this poverty line as bench mark the study indicated that 56.17 percent of the households were poor. The result of the logistic regression model revealed that out of 18 variables included in the model, 13 explanatory variables were found to be significant at 1%, 5% and 10% level. Accordingly, family size, household head sex, household age, dependency ratio and marital status were found to have positive association with poverty of the household and statistically significant. Mean while Age square, cultivated land size, oxen, access to credit, off farm activity, household health, remittance, and market access were found out to have strong negative association with the households poverty status and statistically significant up to less than 10% level of significance.

Key words: Binary Logit, Cost of basic need, Consumption approach, Determinants, Household, Rural poverty.

CHAPTER ONE

1. Introduction 1.1 Back Ground of the Study Poverty is a multidimensional and complex phenomenon and is related not only to the income or consumption, considered as monetary dimension of poverty, but also to non- monetary dimensions such as education, health, gender equality, water supply, etc. Poverty is caused by many factors and brings several effects which influence the lives of people considered to be poor. The influence of the factors varies from one place to another, because many countries have different development possibilities. The influential factors of poverty level are not only economical, but also social, political, cultural, geographical, etc (Spaho, 2014).

According to United Nations(1995): Extreme poverty, or absolute poverty, is defined 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. The 2014 release of a new set of purchasing power parity conversion factors (PPPs) for 2011 has prompted a revision of the international poverty line. In order to preserve the integrity of the goal posts for international targets such as the Sustainable Development Goals and the World Bank’s Twin Goals, the new poverty line was chosen so as to preserve the definition and real purchasing power of the earlier $1.25 line (in 2005 PPPs) in poor countries. Using the new 2011 PPPs, the new line equals $1.90 per person per day (Francisco H. G. Ferreira, 2015).

Using this new line (as well as new country-level data on living standards), the World Bank projects that global poverty will have fallen from 902 million people or 12.8 per cent of the global population in 2012 to 702 million people, or 9.6 per cent of the global population (World Bank, 2015).Therefore, extreme poverty widely refers to earning below the international poverty line of $1.9/day (in 2011 prices).The reduction of extreme poverty and hunger was the first Millennium Development Goal (MDG1), as set by 189 United Nations Member States in 2000 including Ethiopia. Specifically, MDG1 set a

target of reducing the extreme poverty rate in half by 2015, a goal that was met 5 years ahead of schedule (UN, 2014).With the expiration of the MDGs fast approaching, the international community, including the UN, the World Bank and the US, has set a target of ending extreme poverty by 2030(World Bank,2015).

World Bank Group (2013) proposed two goals to measure success in promoting sustainable economic development, and to monitor its own effectiveness in delivering results. The first goal is to essentially end extreme poverty, by reducing the share of people living on less than $1.25 a day to less than 3 percent of the global population by 2030. The second goal is to promote shared prosperity by improving the living standards of the bottom 40 percent of the population in every country. Critically, the goals need to be pursued in ways that sustainably secure the future of the planet and its resources, promote social inclusion, and limit the economic burdens that future generations inherit. (WBG & IMF, 2014/2015).

In most of developing countries larger population are living in rural than urban: About 70 percent of the world’s very poor people (around one billion) are rural, and a large proportion of the poor and hungry amongst them are children and youth. Despite massive progress in reducing poverty in developing countries the rural people are suffering from poverty resulted from lack of assets, limited economic opportunities, poor education and capabilities(IFAD, 2011).

In Ethiopia incidence of poverty declined markedly between 2004/05 and 2010/11. The headcount poverty rate fell from 38.7 % in 2004/05 to 29.6 % in 2010/11. This implies that Ethiopia is on the right track to achieving the MDG target of reducing poverty by half. Over the same period, poverty gap is also reduced, but not the severity of poverty. Headcount poverty fell in all regions of the country. The headcount poverty rate fell in rural areas from 39.3 % in 2004/05 to 30.4 % in 2010/11.Over the same period, in urban areas it declined substantially, from 35.1 % in 2004/05 to 25.7 % in 2010/11(MoFED, 2012).

Poverty in Ethiopia is highly correlated with the size and composition of households, the educational level of household head, the degree and extent of dependency within the household, asset ownership(particularly ownership of oxen in rural areas), the occupation of household heads, rapid population growth, major health problems, lack of infrastructure and extreme environmental degradation (MoFED, 2012).Thus identifying what characteristics are correlated with rural poverty can yield critical insights for policy makers.

Approximately 95% of Damot Gale Woreda households rely on agriculture for their major livelihood strategy ,all are smallholder farmers, they face constraints including shortage of land, land degradation and soil in fertility, lack of investment, erratic and unpredictable rainfall patterns, poor access to market, few off farm employment opportunities, low agricultural productivity and chronic illness .The major economic activity is rain fed farming and population density of 652 persons live in square kilometer which is more than regional average 163.9 and zonal average density 415.8 and an average of 4.3 persons live in a single household CSA (2011) estimation. Due to these and other demographic and socio-economic factors in habitants were chronically poor in the study area (DGWAO, 2014).

Since poverty is a major constraining factor among farming households, it is important to investigate the trend, structure and determinants of poverty in rural households. Therefore this study mainly focused on determinants of rural household’s poverty in Damot Gale Woreda by including the most crucial demographic and socio economic variables. Both qualitative and quantitative method of data collection method was employed and consumption per capital used to indicate the standard of living in study area rather than using income as welfare indicator. Cost of basic needs approach was used for setting poverty line .Data analysis was done by using descriptive analysis, FGT classes of poverty measures and econometric model binary logit regression analysis was employed to capture the influence of explanatory variables on dependent variable.

1.2 Statement of the Problem Poverty has multiple causes that exhibit economic, social and political characteristics and hence poverty reduction policies require multi-dimensional approaches and strategies. Poverty reduction policies have become one of the priority policy targets of governments in developing countries. The challenges to reduce poverty are formidable in developing countries where poverty is deep and widespread, income is extremely low, growth rate is weak and income distribution is uneven (Moges, 2013).

These features of the production and distribution of output create systemic tendency for the poverty elasticity of income to be weak, making the growth induced poverty reduction less effective (Besley and Burguess, 2003; Burgingnon, 2003) as cited in (ibid).

Understanding poverty in the Ethiopian context also needs to consider its multidimensional characteristics which go beyond mere income and food provision. Such characteristics include aspects of human capabilities, assets and activities necessary for sustainable livelihoods. A sustainable livelihood is one that can cope with and recover from stresses and shocks and maintain or enhance its capabilities and assets both now and in the future, without undermining the natural resource base (Carney 1998) as cited in (Asmamaw, 2004).

Ethiopia’s current high level of absolute poverty and food insecurity is primarily due to a low productivity in the Country’s huge agricultural sector. The high rate of population growth is also related to poverty, since people in absolute poverty have the incentive for high fertility to increase the number of potential income earners in the household and to provide for old age security (Smith, 1997) as cited in (Sisay Asefa et al, 2012).

The central challenge of poverty reduction in Ethiopia is essentially how to generate sustainable rise in the productivity of the labor force in agriculture, improve the application of modern technology and inputs in the sector, and reduce its vulnerability to shocks or falling to chronic poverty of rural residents. Capital investment, application of modern and improved agricultural production technology, secured landownership, and

effective financial services are some of the factors that could initiate and sustain improvement in productivity in agriculture or can help as an instrument to reduce poverty. The main impediments to poverty reduction in Ethiopia emerge from a complex wave of interaction of economic, political, demographic, social, geographic, and institutional factors and hence poverty reduction policies should address these underlying forces to develop strategies with lasting effect (MoFED, 2012).

Much of the studies on the correlates of rural poverty in Ethiopia had been confined on quantitative rather than qualitative methods using households as unit of analysis. The views and perception of households and the community at large on the manifestations and determinants of rural poverty have been overlooked. But there is mounting evidence that using quantitative and qualitative approaches together yield synergy. Furthermore, what have so far been studied in Ethiopia, much if not all, concentrate on and reflect the national picture. But studies and analysis at an aggregate level do not necessarily reflect the situation at grass root level. According to Dercon and Krishnan (1996) as cited in Metalign (2005) strongly advise that one should be careful about the implications derived from measurement and factors of poverty at national level, because it hides many important differences that exist in different locations, and hence, are likely to be reliable only for particular localities.

In economies where the initial pattern of income distribution is highly unequal and vertical mobility is restricted by economic, social and institutional hurdles, economic growth if it happens at all tends to have limited impact on reducing poverty (Besley Timothy & Robin Burgess, 2003).

Whereas income redistribution policies, when cautiously implemented, could be used to address immediate crisis situations, they have limited effectiveness in reducing poverty on a sustainable basis (ibid).

Even economies with remarkable growth rate could not achieve sustainable poverty reduction if the growth process does not generate productive job opportunities, mobility,

and accumulation of assets and capital for an increasing share of the population (Besley et al,2003). The pattern, characteristics and sector composition and sustainability of growth rate are therefore as important for poverty reduction as the pace of growth performance (Moges, 2013).

This research found the existing study gap of the study area on the determinants of poverty at a disaggregated level using households as unit of analysis, partially flit the methodological gap in measuring the poverty line, the time gap and used econometric model in order to explain the relationship between poverty and explanatory variables. Accordingly this study has been conducted with the main aims of measuring poverty in Damot Gale woreda and examined the relationship between poverty and different socio- economic characteristics among the community under study.

Major research questions in analyzing determinants of rural poverty in the study area are the following:

I. What are the major determinants of rural poverty in the woreda? II. What is the poverty line for Damot Gale woreda? III. What are the incidence, depth and severity of rural poverty in the study area?

1.3 Objectives of the Study

1.3.1 General Objectives The general objective of this study is to examine the profile and determinants of poverty in rural households in Damot Gale woreda.

1.3.2 Specific Objective  Identify factors that determine poverty in rural households in the study area.  Determine the proportion of households who live below and above the poverty line.  Analyzing the magnitude (incidence, depth and severity) of poverty in study area.

1.4 Hypothesis of the Study  The majority of rural households earn/consume below poverty line.  There is a significant negative relationship between rural poverty and its determinants: household age, education level, household head health access, cultivable land holding size, agricultural input use, household market access, off farm income, ownership of oxen, TLU, saving behavior, access to credit, access to remittance and community association.  There is a significant positive relationship between rural poverty and its determinants:-marital status, family size, dependence ratio and household head sex.

1.5 Significance of the Study In Ethiopia many researches concerning poverty have focused at national picture whereas, there is a limited grass root level (disaggregate) analytical study on rural poverty situations and determinants in Ethiopia. On the other hand any intervention to alleviate and ultimately eliminate poverty needs understanding of the extent, determinants and manifestations of poverty. Therefore, such studies are without hesitation important for the poverty reduction objective of the country, whose significant portion of population lives below the minimum acceptable standard of welfare indicator in general and the study area particularly.

This study shed light on direction for designing programs to reduce rural poverty in the study area and other areas that need the government interventions, NGO and other bodies concerned for the development and reduction of poverty. The results of the study provide information that helps to prioritize among the many possibilities depending on the relative extent of influences of its determinants. Finally, evaluating the determinants of poverty in rural households of Damot Gale will serve as a base for future study since no studies has been conducted in this area on correlates of rural poverty. Lastly this study attempts to make further contribution to the previous studies and can be used as a source material for further studies.

1.6 Scope and Limitation of the Study This study, which attempted to assess the main determinants and characteristics of rural poverty in Damot Gale Woreda based on the information obtained from farming households in the year 2016.Though welfare/standard of living differs from individual to individual even between a single households, due to difficulty of addressing each individual family member with a limited time ,we used the adult equivalence scale and assuming all individuals have the same standard of living in a given household. The study employed consumption as welfare indicator due to its role by better capturing long-run living standard, reflecting actual standard of living, its non-erratic and non-seasonality nature compared to income as an indicator. This study also employed CBN approach for setting minimum standard of living (poverty).

1.7 Organization of the Paper The studywas organized in to five chapters. The first chapter dealt with the introduction having back ground of the study, statement of the problem, objectives, hypothesis of the study, scope and limitation of the study, and significance of the study.The second part of the paper was about the theoretical and empirical literatures including basic conceptual and measurement issues related to the subject of the study available in Ethiopia and other countries.

The third chapter consists of the research methodology part that introduces data type and source, sample size determination, sampling techniques, data collection instrument, method of data analysis, and model specification (econometric model). Chapter four contains the descriptive and econometric analysis on the determinants of poverty in rural households, chapter five deals about conclusion and policy implication based on the empirical findings.

CHAPTER TWO

2. REVIEW OF THE RELATED LITERATURE

In this section of the study a review of the theories, concepts and definitions of related literatures were conceptualized and results from empirical studies were summarized.

2.1 Theoretical Literature

2.1.1 Definition of Poverty According to Haughton et al (2009), poverty is pronounced deprivation in wellbeing. It begs the questions of what is meant by well-being and of what is the reference point against which to measure deprivation. One approach is to think of well-being as the command over commodities in general, so people are better off if they have a greater command over resources. The main focus is on whether households or individuals have enough resources to meet their needs. Typically, poverty is then measured by comparing individuals’ income or consumption with some defined threshold below which they are considered to be poor. This is the most conventional view poverty is seen largely in monetary terms and is the starting point for most analysis of poverty.

A second approach to well-being (and hence poverty) is to ask whether people are able to obtain a specific type of consumption good: Do they have enough food? Or shelter? Or health care? Or education? Nutritional poverty might be measured by examining whether children are stunted or wasted; and educational poverty might be measured by asking whether people are literate or how much formal schooling they have received (ibid).

Perhaps the broadest approach to well-being is the one articulated by Sen (1979) who argues that well-being comes from a capability to function in society. Thus, poverty arises when people lack key capabilities, and so have inadequate incomes or education, or poor health, or insecurity, or low self-confidence, or a sense of powerlessness, or the absence of rights such as freedom of speech. Viewed in this way, poverty is a multidimensional phenomenon and less amenable to simple solutions. For instance, while higher average incomes will certainly help reduce poverty, these may need to be accompanied by measures to empower the poor, or insure them against risks, or to address specific weaknesses such as inadequate availability of schools or a corrupt health service.

Poverty is related to, but distinct from, inequality and vulnerability. Inequality focuses on the distribution of attributes, such as income or consumption, across the whole population. Vulnerability is defined as the risk of falling into poverty in the future, even if the person is not necessarily poor now; it is often associated with the effects of “shocks” such as a drought, a drop in farm prices, or a financial crisis. Vulnerability is a key dimension of well-being since it affects individuals’ behavior in terms of investment, production patterns, and coping strategies, and in terms of the perceptions of their own situations (ibid).

Farmers in rural Ethiopia live in a shock-prone environment. The causes of rural poverty are many including wide fluctuations in agricultural production as a result of drought, ineffective and inefficient agricultural marketing system, under developed transport and communication networks, under developed production technologies, limited access of rural households to support services, environmental degradation and lack of participation by rural poor people in decisions that affect their livelihoods. However, the persistent fluctuation in the amount and distribution of rainfall is considered as a major factor in rural poverty. Small-scale farmers are the largest group of poor people in Ethiopia. Their average land holdings are smaller, their productivity is low and they are vulnerable to drought and other adverse natural conditions. Poor people in rural areas face an acute lack of basic social and economic infrastructure such as health and educational facilities, veterinary services and access to safe drinking water (Regassa et al, 2007).

2.1.2The Conceptualization of Poverty Poverty affects many aspects of human conditions like economic, social, physical, moral, psychological, etc. As a result, there are different approaches in the conceptualization of poverty. One pair of approach comprises the “Welfarist” and the “Non-Welfarist” approach. While the former aims at defining the concept of well-being on the basis of the link that exists between income and utility or standard of living, the later approach focuses little on utility (Sen, 1979).

Following either of the two, different individuals defined poverty differently. For instance, Townsend (2002) defined poverty not just as a failure to meet minimum

nutrition levels but also as a failure to keep up with the standards of living prevailing in a society. According to Sen(1987) also relates poverty to entitlement failures to various goods and services. Rowtree (1901) has developed poverty standards on the basis of nutritional and other requirements.

On the other hand, Haughton and Khandker (2009),sees poverty in very broader terms as being unable to meet “basic needs” including food, health, education, shelter, etc. Economists, however, often prefer to view the concept of well-being in terms of the “Welfarist” approach. That is to say, they take expenditure on goods and services consumed by individuals valued at market prices so as to categorize a person as “poor" or “non-poor”. This money metric utility is derived from the neo-classical theory of consumer behavior. Therefore, in this case, poverty is said to exist in a given society when people are unable to obtain the minimum basic requirements necessary to sustain life of individuals.

Such kind of conceptualization of individual’s well-being in terms of standard of living measures seems pragmatic in developing countries, where much emphasis is given on food security and consumption deprivation. But, in developed societies, non-materialist aspects like the right to voting and/or participation could seem more realistic in understanding levels of societal wellbeing (ibid).

Another approach to define poverty is to see societal well-being from the perspective of severity as “chronic” and “transient”. Structural (chronic) poverty is defined as persistent or permanent socio-economic deprivations of the population whereas transitory poverty is temporary socio-economic deprivations. The former is linked to a host of factors like lack of skill, lack of productive resources, socio-political and cultural factors, etc. The later, on the other hand, is linked to natural and man-made disasters and is easily reversible. These all imply that there are various approaches towards the conceptualization and definition of poverty and well-being (Ravallion, 1998).However, there are essentially three broad categories to the definition of poverty. These are:

2.1.2.1Absolute Poverty According to MoFED (2012), absolute poverty can be viewed as the inability to secure the minimum basic needs for human survival and it is poverty in which people do not have access to basic necessities to fulfill their basic physical needs and pointed out that for the purpose of measuring poverty; the welfaristic framework does not provide a well- defined poverty line. Hence, the non-welfaristic approach, which is usually based on the basic needs or minimum caloric requirement, is often used to draw poverty line. The three most popular methods that use caloric requirement to set poverty lines are the Direct Caloric Intake, Food Energy Intake (FEI) and the Cost of Basic Needs (CBN). The Direct Caloric Intake method defines poverty line as the minimum caloric requirement for survival. Individuals who consume below a predetermined minimum level of caloric intake are deemed to be under poverty. Hence, this method equates poverty with malnutrition. The drawback of this method is that it does not take into account the cost of getting the basic caloric requirement and it totally overlooks the non-food requirement. If poverty has to be measured by a lack of command of basic goods and services, measuring poverty by caloric intake only is unlikely to reveal the extent of impoverishment of a given society (ibid).

2.1.2.2 Relative Poverty The relatively poor, therefore, are those whose incomes are lower compared to that of the rest of the community even if they are in a position to secure an adequate level for survival. Relative poverty means that some people are poorer than the rest of the community. Thus, the concept, relative poverty, is primarily concerned with the distribution of income and hence, inequality in living conditions among a population (MEDaC, 1999b). While almost everyone in the United States receives a higher income than almost everyone in, says Chad, there are still (relatively) poor people in the United States and (relatively) non-poor people in Chad (Gillis M., Perkins, H., Roemer M., and Snodgrass R., 1996).

Relative poverty is a global phenomenon. Because it is relative to the general standard of living rather than being based on the minimum set of basic goods, they are higher in richer than poor countries. This approach is suffering from major weaknesses. Firstly lacks clarity as to whether it is an indicator of poverty or measurement of income inequality. Secondly, the approach is entirely dependent on the value judgment of the

researcher that it is difficult to monitor poverty over time or space. Thirdly, the relative poverty line is essentially quite arbitrary and always assumes a constant per cent of the population in the bottom as poor, even if living standards for the whole population have risen over time. Fourthly, such a method is technically feasible only for developed countries (Ravallion M. , 1992). Furthermore, for LDCs including Ethiopia, where the largest share of the population lives in absolute poverty, the emphasis on relative poverty is not of primary relevance (Metalign, 2005).

2.1.2.3 Subjective Poverty In the subjective poverty definition, the identification of the poor and the non-poor depends on the subjective judgment of individuals about what constitutes a socially acceptable minimum standard of living in their own societies. Hence, unlike the above approaches, subjective poverty line depends directly on the opinion and feeling of the concerned individuals to determine the minimum level of income for themselves. The result of this approach may sometimes be misleading as it takes purely an account of individuals’ or groups’ own declaration about their position. In the conceptualization of poverty, the choice of income or consumption expenditure as best indicator for living standard measurement of households is another point of consideration. Most analysts, however, prefer current consumption to income as indicator of living standards for developing countries. This is because income of the poor often varies over time in fairly predictable ways. Particularly, this is true for underdeveloped economies that depend on traditional production systems (Ravallion, 1992).

2.1.3. Theories of Poverty

2.1.3.1 Poverty Caused by Individual Deficiencies

This theory mainly focus on explanations that the individual are responsible for their poverty situation. Typically, politically conservative theoreticians blame individuals in poverty for creating their own problems, and argue that with harder work and better choices the poor could have avoided (and now can remedy) their problems. Other variations of the individual theory of poverty ascribe poverty to lack of genetic qualities such as intelligence that are not so easily reversed. The belief that poverty stems from individual deficiencies is old (Ted K.Bradshaw, 2006).

According to Rainwater (1970:16), critically discusses individualistic theories of poverty as a “moralizing perspective” and notes that the poor are “afflicted with the mark of Cain. They are meant to suffer, indeed must suffer, because of their moral failings. They live in a deserved hell on earth.” Rainwater goes on to say that it is difficult to overestimate the extent to which this perspective (incorrectly) under-girds our visions of poverty, including the perspective of the disinherited themselves. Neo-classical economics reinforces individualistic sources of poverty. The core premise of this dominant paradigm for the study of the conditions leading to poverty is that individuals seek to maximize their own well being by making choices and investments, and that (assuming that they have perfect information) they seek to maximize their well being cited in (ibid).

2.1.3.2 Poverty Caused by Cultural Belief This theory suggests that poverty is created by the transmission over generations of a set of beliefs, values and skills that are socially generated but individually held. Individuals are not necessarily to be blamed because they are victims of their dysfunctional subculture or culture. Culture is socially generated and perpetuated, reflecting the interaction of individual and community. This makes the “culture of poverty” theory different from the “individual” theories that link poverty explicitly to individual abilities and motivation. Once the culture of poverty has come into existence it tends to perpetuate itself. By the time slum children are six or seven they have usually absorbed the basic attitudes and values of their subculture. Thereafter they are psychologically unready to take full advantage of changing conditions or improving opportunities that may develop in their lifetime (Scientific American, October 1966 quoted in Ryan, 1976: 120) as cited in (Ted K.Bradshaw, 2006).

2.1.3.3 Poverty Caused by Economic, Political, and Social Distortions or Discrimination The first “individualistic” theory of poverty is advocated by conservative thinkers and the second is a culturally liberal approach, this theory is a progressive social theory. Theorists in this tradition look not to the individual as a source of poverty, but to the economic, political, and social system which causes people to have limited opportunities and resources with which to achieve income and wellbeing.

The 19th century social intellectuals developed a full attack on the individual theory of poverty by exploring how social and economic systems overrode and created individual poverty situations. Much of the literature on poverty now suggests that the economic system is structured in such as a way that poor people fall behind regardless of how competent they may be. Partly the problem is the fact that minimum wages do not allow single mothers or their families to be economically self-sufficient. Elimination of structural barriers to better jobs through education and training have been the focus of extensive manpower training and other programs, generating substantial numbers of successes but also perceived failures (Jencks, 1996).

According to Chubb and Moe (1996) as cited in (Abdulaziz, 2014), a parallel barrier exists with the political system in which the interests and participation of the poor is either impossible or is deceptive. Recent research has confirmed the linkage between wealth and power, and has shown how poor people are less involved in political discussions, their interests are more vulnerable in the political process, and they are excluded at many levels. Coupled with racial discrimination, poor people lack influence in the political system that they might use to mobilize economic benefits and justice. A final broad category of system flaws associated with poverty relate to groups of people being given a social stigma because of race, gender disability, religion, or other groupings, leading them to have limited opportunities regardless of personal capabilities.

2.1.3.4 Poverty Caused by Geographical Disparities

Rural poverty, urban disinvestment, third-world poverty, and other framings of the problem represent a spatial characterization of poverty that exists separate from other theories. While these geographically based theories of poverty build on the other theories, this theory calls attention to the fact that people, institutions, and cultures in certain areas lack the objective resources needed to generate well being and income, and that they lack the power to claim redistribution according to (weber,2004) as cited in (Abdulaziz, 2014).

2.1.3.5 Poverty Caused by Cumulative and Cyclical Interdependencies

The previous four theories have demonstrated the complexity of the sources of poverty and the variety of strategies to address it. The final theory of poverty I will discuss is by far the most complex and to some degree builds on components of each of the other theories in that it looks at the individual and their community as caught in a spiral of

opportunity and problems, and that once problems dominate they close other opportunities and create a cumulative set of problems that make any effective response nearly impossible. The cyclical explanation explicitly looks at individual situations and community resources as mutually dependent, with a faltering economy, for example, creating individuals who lack resources to participate in the economy, which makes economic survival even harder for the community since people pay fewer taxes (Bradshaw, 2000).

According to Myrdal (1957:23), who developed a theory of interlocking, circular, interdependence within a process of cumulative causation that help explains economic underdevelopment and development. Myrdal notes that personal and community well being are closely linked in a cascade of negative consequences, and that closure of a factory or other crisis can lead to a cascade of personal and community problems including migration of people from a community. Thus the interdependence of factors creating poverty actually accelerates once a cycle of decline is started (Abdulaziz, 2014).

2.1.4 Measuring Poverty The conventional view links wellbeing primarily to command over commodities, so the poor are those who do not have enough income or consumption to put them above some adequate minimum threshold. This view sees poverty largely in monetary terms. Poverty may also be tied to a specific type of consumption; for example, people could be house poor or food poor or health poor. These dimensions of poverty often can be measured directly, for instance, by measuring malnutrition or literacy. The broadest approach to well-being (and poverty) focuses on the capability of the individual to function in society. Poor people often lack key capabilities; they may have inadequate income or education, or be in poor health, or feel powerless, or lack political freedoms (World Bank, 2005).

Three steps need to be taken in measuring poverty (Ravallion M. , 1992),

 Defining an indicator of welfare  Establishing a minimum acceptable standard of that indicator to separate the poor from the non-poor (the poverty line)  Generating a summary statistic to aggregate the information from the distribution of this welfare indicator relative to the poverty line.

2.1.4.1 Choosing an Indicator of Welfare

According to (Sen, 1979), the welfarist approach seeks to measure household utility, which in turn is usually assumed to be approximated by household consumption expenditure or household income; these may be considered as inputs into generating utility. When divided by the number of household members, this gives a per capita measure of consumption expenditure or income. Of course, even household expenditure or income is an imperfect proxy for utility; for instance, it excludes potentially important contributors to utility such publicly provided goods or leisure. A non-welfarist, approach might focus on whether households have attained certain minimal levels of, say, nutrition or health. Thus, such measures are useful in fleshing out a multidimensional portrait of poverty.

2.1.4.2 Establishing a Minimum Acceptable Standard or Poverty Lines

Households whose consumption expenditure falls below this line are considered poor. The choice of poverty line depends in large measure on the intended use of the poverty rates. The poverty line may be thought of as the minimum expenditure required by an individual to fulfill his or her basic food and nonfood needs. Once we have computed a household’s consumption, we need to determine whether that amount places the household in poverty, or defines the household as poor. The threshold used for this is the poverty line. The poverty line defines the level of consumption (or income) needed for a household to escape poverty. It is sometimes argued that the notion of a poverty line implies a distinct turning point in the welfare function (Ravallion, 1998).

A corollary is that it usually makes sense to define more than one poverty line. For example, one common approach is to define one poverty line that marks households that are poor and another lower level that marks those that are extremely poor. Another approach is to construct a food poverty line, which is based on some notion of the minimum amount of money a household needs to purchase some basic-needs food bundle and nothing more. If the cost of basic nonfood needs is estimated, the food poverty line added to the nonfood needs will equal the overall poverty line (ibid).

Therefore, poverty line for a household, zi, may be defined as the minimum spending or consumption ( income) needed to achieve at least the minimum utility level uz, given the level of prices (p) and the demographic characteristics of the household (x),

So, = (, , ) − − − − − − − − − − − − − − − − − − − − − − − − − − (1)

There are three methods that help to construct poverty line are: the cost of basic needs, food energy intake, and subjective evaluations: the cost of basic needs, food energy intake together known as objective poverty lines.

[ 2.1.4.2.1 Objective Poverty Line

A common and fairly satisfactory method of approaching capabilities is to begin with nutritional requirements. The most common way of making this operational is the cost of basic needs approach, while the food energy intake method has been suggested as an alternative when the data on price are more limited.

2.1.4.2.1.1 The Cost of Basic Needs Method

It first estimates the cost of acquiring enough food for adequate nutrition usually 2,100 Calories per person per day and then adds the cost of other essentials such as clothing and shelter (Rowntree, 1941). It proceeds as follows:

 Stipulate a consumption bundle that is deemed to be adequate, with both food and nonfood components.  Estimate the cost of the bundle for each subgroup

Although we begin with consumption bundle so much food, so much housing space, so much electricity, and so forth the poverty line is measured in money. We are therefore not insisting that each basic need be met by each person (a non-welfarist position), only that it could be met (a welfarist position). Operationally, the steps to follow are:

 Pick a nutritional requirement for good health, such as 2,100 Calories per person per day. This is widely used, and has been proposed by the Food and Agricultural Organization of the United Nations.  Estimate the cost of meeting this food energy requirement, using a diet that reflects the habits of households near the poverty line (for example, those in the lowest, or second-lowest, quintile of the income distribution; or those consuming

between 2,000 and 2,200 calories). Call food component and add a nonfood

component (). 18

Therefore, poverty line = + − − − − − − − − − − − − − − − − − −(2)

2.1.4.2.1.2 Food Energy Intake Method When price information is unavailable, the food energy intake method can be used. This method plots expenditure (or income) per capita against food consumption (in calories per person per day).To find the level of consumption expenditure (or income) that allows the household to obtain enough food to meet its energy requirements consumption will include nonfood as well as food items; even underfed households typically consume some clothing and shelter, which means that at the margin these “basic needs” must be as valuable as additional food. Given some level of just-adequate food energy intake k, one may use it to determine the poverty-line level of expenditure, z. = () − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − (3) So, given monotonicity, = () − − − − − − − − − − − − − − − − − − − −(4) Or, given a minimum adequate level of calorie, we have = () − − − −(5) This approach does not require any information about the prices of goods consumed. First one needs to determine the amount of food that is adequate at 2,100 calories per person per day, in line with UN Food and Agriculture Organization.

2.1.4.2.2Subjective Poverty Line We could measure poverty by asking people to define a poverty line, and using this to measure the extent of poverty. The concept of subjective poverty is based on the premise that people are the best judges of their own situation and that their opinions should ultimately be the decisive factor in defining welfare and poverty. This approach explicitly recognizes that poverty lines are inherently subjective judgments people make about what constitutes a socially acceptable minimum standard of living in their own societies (Yohannes, 1996).

Subjective poverty measures are therefore based on responses of individuals to attitudinal questions on household income and welfare like ‘what level of income do you personally consider as absolutely minimal? , In your opinion, is the household income sufficient to make the family’s ends meet?’ There is no guarantee for individuals similar in all respects to provide similar responses to the same question, and hence, does not ensure consistency.

Furthermore, the application of this approach has been confined to developed countries (ibid). 2.1.4.3 Poverty Indices It is based on information available on a welfare measure such as income or consumption per capita and a poverty line, for each household or individual. To construct a summary measure of the extent of poverty there are varies types of poverty indices but the most widely use are the following based on Foster, Greer, and Thorbecke (1984). 2.1.4.3.1 Headcount Index (P0) The headcount index (P0) measures the proportion of the population that is poor. It is popular because it is easy to understand and measure.

= − − − − − − − − − − − − − − − − − − − − − − − − − − − − − −( 6 ) 1 = I(y < ) − − − − − − − − − − − − − − − − − − − − − − − − − −(7 ) N Here, I (.) is an indicator function that takes on a value of 1 if the bracketed expression is true and 0 otherwise. So if expenditure (yi) is less than the poverty line (z), then I (.) equal to 1 and the household would be counted as poor. Np is the total number of the poor and N the total number of people in the population. The greatest virtues of the headcount index are that it is simple to construct and easy to understand.

According to (Dalton, 1920), however the measure has at least three weaknesses: First, the headcount index does not take the intensity of poverty into account. In that it violates the transfer principle an idea first formulated by that states transfers from a richer to a poorer person should improve the measure of welfare. Here if a somewhat poor household were to give to a very poor household, the headcount index would be unchanged, even though it is reasonable to suppose that poverty overall has lessened; second, the headcount index does not indicate how poor the poor are, and hence does not change if people below the poverty line become poorer and third, the poverty estimates should be calculated for individuals and not households.

But survey data are almost always related to households, so in order to measure poverty at the individual level we must make a critical assumption that all members of a given household enjoy the same level of well-being (Ravallion 1996).

2.1.4.3.2 Poverty Gap Index (P1) A moderately popular measure of poverty is the poverty gap index, which adds up the extent to which individuals on average fall below the poverty line, and expresses it as a percentage of the poverty line. More specifically, define the poverty gap (Gi) as the poverty line (z) less actual income (yi) for poor individuals; the gap is considered to be zero for everyone else. Using the index function, we have

= ( − ) ∗ ( < ) − − − − − − − − − − − − − − − − − − − − − −( 8 ) 1 G = − − − − − − − − − − − − − − − − − − − − − − − − − − − ( 9 ) N z The poverty gap index (P1) measures the extent to which individuals fall below the poverty line (the poverty gaps) as a proportion of the poverty line. The sum of these poverty gaps gives the minimum cost of eliminating poverty, if transfers were perfectly targeted. The measure does not reflect changes in inequality among the poor. 2.1.4.3 .3 The Squared Poverty Gap or Poverty Severity It is one of the Foster-Greer-Thorbecke (FGT) classes of poverty measures .To construct a measure of poverty that takes into account inequality among the poor, some researchers use the squared poverty gap index. This is simply a weighted sum of poverty gaps (as a proportion of the poverty line), where the weights are the proportionate poverty gaps themselves. Hence, by squaring the poverty gap index, the measure implicitly puts more weight on observations that fall well below the poverty line. Formally: 1 G = (α ≥ 0) − − − − − − − − − − − − − − − − − − − − − ( 10 ) N z where α is a measure of the sensitivity of the index to poverty and the poverty line is z, the value of expenditure per capita for the i-th person’s household is xi, and the poverty gap for individual i is Gi = z-yi (with Gi = 0 when yi > z) When parameter α = 0, P0 is simply the head-count index. When α = 1, the index is the poverty gap index P1, and when α is set equal to 2, P2 is the poverty severity index. For all α > 0, the measure is strictly decreasing in the living standard of the poor (the lower your standard of living, the poorer you are deemed to be). Furthermore, for α > 1 it also has the property that the increase in measured poverty due to a fall in one’s standard of living will be deemed greater the poorer one is. The measure is then said to be "strictly convex" in incomes (and "weakly convex" for α = 1). Another convenient feature of the FGT class of poverty

measures is that they can be disaggregated for population sub-groups and the contribution of each sub-group to national poverty can be calculated.

2.2 EMPIRICAL LITERATURE

Ethiopia has experienced different challenges in alleviating poverty, let alone meeting the Millennium Development Goal of halving the incidence of poverty by 2015. In fact, in the absence of agricultural growth, the country’s poverty rate would rise even higher, leaving as many as 10 million additional people in poverty by 2015. Modeling results indicate that, within agriculture, staple crops have the greatest capacity to contribute to poverty reduction. Based on annual growth of 3.4%, (1.5% additional productivity growth above baseline levels) staple food growth would support economic growth in the order of 4% and agricultural growth of about 3.5% per year. In response, the poverty rate in Ethiopia would fall from its 2000 level of 44.4% to about 37% in 2015.Yet this is insufficient. In the absence of improved market conditions, growth in staples will be difficult to achieve and increased grain production could harm farmers by depressing prices in the food surplus area. Thus, market development and access should be an integral component of agricultural development strategies (Xinshen Diao, Alejandro Nin Pratt, 2006).

Growth in the agricultural sector is combined with improved marketing margins through cross-sector linkage effects, GDP growth increases to 5.8% per year, and Agricultural share to GDP growth increases to 5.4% per year. Reducing marketing costs primarily benefits smallholders via the increased prices they receive for their goods, increasing their income from the same level of output. Moreover, improving market conditions creates a more efficient trading sector (as part of the service sector), which itself can generate greater nonagricultural income at constant costs. Due to such cross-sector linkages and positive price effects, the poverty rate under this scenario is significantly lowered, drawing the objective of halving poverty rate by 2015 within reach. Moreover, the pro- poor effect of the resulting growth is much stronger in rural areas, where simulation results indicate the poverty rate drops to 25% by 2015 from the 2003 level of 45.8 % (Ibid).

Remittances are increasingly becoming an important source of external financing for the developing countries. For some of the developing countries, it forms almost 40-50% of their GDP used to repay loans taken to finance migration or education, and insurance and strategic motives. It also directly contributes to household income, allowing households to purchase more assets; enables higher investment in business; and facilitate buying more goods, including education and health inputs. At the household level, remittances can spur entrepreneurial activity, emphasize the knowledge transfer and change in attitudes of the remaining family members of the migrants. The results of the study shows that remittances significantly reduce poverty in recipient countries but the results are more reliable for countries with remittances greater than 5% of GDP (Banga et al., 2009). Determinants and dimensions of poverty in Gulomekeda wereda rural kebeles total family size & dependency ratio are found to have positive association with poverty of the household and statistically significant. Meanwhile, farm size, total livestock owned (TLU), value of asset, educational status of the household head, access to credit and access to off farm income are found out to have strong negative association with the households poverty status. Outcome pertinent to Welfare inequality reveals that there is great variation in consumption expenditure of the households (Afera, 2015).

The regional distribution of total and food poverty in Ethiopia in 2010/11 shows, poverty head count index is the highest in Afar (36.1%) followed by Somali (32.8%) and Tigray (31.8%), while poverty estimates are lowest in Harari (11%) followed by Addis Ababa (28.1%) and Dire Dawa (28.3%). In terms of food poverty, the highest poverty is observed in Amhara (42.5%) followed by Tigray (37.1) and Beneshangul Gumuz (35.1%). The lowest food poverty is found again in Harari (5%) followed by Dire Dawa (21.7%) and SNNP (25.9%). Over all, compared to the previous years, the difference in poverty incidence among the regional states in 2010/2011 has narrowed substantially indicating a balanced growth among regional states (MoFED, 2013).

Moreover, absolute poverty is much lower than food poverty in all regions. The poverty results indicate that absolute poverty in 2010/211 compared to 2004/2005 have decline over the past five years in all regions except Dire Dawa (where absolute poverty incidence increased by (6%). Poverty gap in 2010/11 also declined in all regions except in rural Afar, rural SNNP, Addis Ababa and Dire Dawa. Poverty severity also declined in 2010/11 in many of the regions including Tigray, Amhara, Benshangul Gumuz, Harai,

urban Afar, urban Somali and rural Dire Dawa, but poverty severity increased in rural Afar, Oromia, rural Somali, SNNP, Addis Ababa and Dire Dawa(ibid). Food poverty incidence in 2010/11 compared to 2004/05 declined in all regions except in rural Amhara where food poverty incidence increased by 14%. Similarly, the food poverty gap in 2010/11 is lower than that of 2004/205 for all regions except for Afar region where food poverty gap increased by 14% in 2010 compared to 2004/05.The result for the food poverty severity index shows that the food poverty severity compared to that of 2004/05 declined in Amhara, urban Oromia, urban SNNP, Harari, and rural Dire Dawa. In the rest of the regions including rural SNNP, rural Tigray, Afar, rural Oromia and rural Somali, food poverty severity has increased in 2010/11 compared to 2004/05.The observed increase in poverty incidence, gap and severity in certain regions mentioned above is difficult to explain and further investigation may be necessary to know the exact reason why poverty has increased (MoFED, 2013).

Despite the few disappointing results in the changes of poverty, the overall reduction in absolute and food poverty incidences, gap and severity in majority of regional rural and urban areas is remarkable while the country has suffered from frequent domestic economic shocks such as inflation and drought and worldwide shocks. Registering substantial poverty reduction in times of such domestic shocks and global crisis show the appropriate policies put place and capability of the Ethiopian Government to protect its vulnerable people from the economic crises (ibid).

The study on identify determinants of rural poverty in Kersa Kondaltity woreda in Oromiya region the variables sex, age of the household head, health condition, agricultural input utilization at recommended rate, distance to the nearest market, number of chickens and traditional beehives owned and involvement in nonfarm activities were not found to be statistically significant at 1, 5 or 10 percent level, and hence do not emerge as major determinants of rural poverty. Whereas, the coefficients for household size, educational attainment of the household head, size of cultivable land owned, number of oxen and other animals owned, saving habit and access to credit were found to be statistically significant with odds ratio different from zero and hence were found to be determinants of rural poverty in the study area (METALIGN, 2005).

Determinant of vulnerability to poverty of households in rural Oromiya logit model was used to investigate the determinants of vulnerability to poverty. Estimate of vulnerability shows that 47.66 percent (1108) of households out of the total sampled households are highly vulnerable to poverty and 17.93 percent of the non-poor are highly vulnerable to poverty. The estimation of the model for determinants of poverty shows that larger household sizes significantly increase the probability of the household to be poor (Dereje, 2013).

Similarly the probability of being poor is on average higher for female headed households relative to the male headed households. On the other hand literate household head has negative effect on poverty. In general, households with large family size, illiterate are more likely to be poor than those with smaller family size and educated and household heads. It seems that the determinants of poverty and vulnerability are similar since those variables that have significant effect on poverty also have significant effect on vulnerability. A sizeable portion of households that are now non-poor are certainly vulnerable to falling into poverty in future. This has policy implications that ex ante measures should be enhanced to prevent as many households as possible from becoming poor and therefore such results should be taken into account, particularly when policy makers design social policy in addition to ex post measures to alleviate those already in poverty (ibid).

The study conducted in Boloso Sore district on determinants of food security status of farming households was collected primary data from 90 rural farm households from four villages, employed headcount index, food insecurity gap index, and food surplus gap index, food severity index and logistic regression model to analyze the generated data. The result of headcount index showed that only 34.5% of rural farm households were food secure while 65.5% were food insecure. The food insecurity gap and food surplus index showed that food secure households exceeded the food security line by 34.6%, while 27.8% of food insecure households fall below the poverty line. The severity of food insecurity gap among food insecure households was 11.7%. The logistic regression result revealed that family size, total cultivated land size, annual income, chemical fertilizer use, oxen owned, livestock holding, gender, access to extension services, market access,

access to credit, access to improved seed, literacy rate and age of households affected food security status of rural farm households (Tekle Leza & Berhanu Kuma, 2015).

The study made on measuring vulnerability to poverty on Ethiopia about 51% of households in Ethiopia are vulnerable to poverty that is significantly higher than the current poverty level of about 29%. While the Northern and the southern regions have the highest average vulnerability of approximately 52%, Oromia region has 49% vulnerability to poverty ratio. Household size, possession of livestock, farm size, and off- farm income, amount of rain fall, and basic goods and services received are the variables that significantly impact vulnerability to poverty (Dawit, 2015).

Determinants of poverty in Kenya: A Household Level Analysis the estimation results show that male-headed households are less likely to be poor. Similarly, the likelihood of being poor is smaller in urban areas than in rural areas. Probably to some extent related to this, people living in households mainly engaged in agricultural activities are more likely to be poor, compared to households in manufacturing activities. In all models the most important determinant of poverty status is the level of education. The effects of this variable are similar across the four models. The coefficient for household size is almost twice as high in the consumption-based as income-based models ones, while the impacts of the sector of employment, as well as the number of animals owned is insignificant in the consumption-based models (Alemayehu, 2005).

Analysis of poverty and its covariates among smallholder farmers in the Eastern Hararghe highlands of Ethiopia the study results depict that increase in household size by one adult equivalent would increase the probability of being extremely poor and moderately poor by 3.13 and 5.16%, respectively, where as it lowers the likelihood that a household will fall under category slightly non-poor and non-poor by 0.49 and 7.79%, respectively. Access to a non-farm source of income is also an important determinant of wellbeing in Eastern Ethiopia. For a given level of other regressers, the probability of being slightly non-poor and non-poor increases by 0.01 and 0.01, respectively. Non-agricultural activities complement agricultural sources of income by availing the household additional resources for both consumption and investment (Ayalneh, 2011).

Probability of being poor for male-headed households is higher than the female-headed households employing the per capita food energy consumption, female-headed households have higher incidence of poverty if household consumption expenditure is considered as a criterion, although the coefficient is not statistically significant (P > 0.10) in the latter case. That means, male-headed households have better capacity to comply with the minimum consumption expenditure required to meet the requirements, but fail to realize it in terms of actual food consumption (ibid).

The coefficient on education reflects the prime role that human capital plays in determining poverty. In fact, education is an important dimension of poverty itself, when poverty is broadly defined to include shortage of capabilities and knowledge deprivation. It has important effects on the poor children’s chance to escape from poverty in their adult age and plays a catalytic role for those who are most likely to be poor, particularly those households living in rural communities. Education is expected to lead to increased earning potential and to improve occupational and geographic mobility of labour. Therefore, it deserves an important place in formulating poverty reduction strategies (Ayalneh, 2011).

The odds ratio in favor of the probability of being poor increases, as household size increases. Other things remaining equal, the odds ratio in favor of poverty increases by a factor of 1.512 as household size increases by one. The possible reason is that with existing high rate of unemployment and less employment opportunity coupled with low rate of payment, an additional household member shares the limited resources that lead the household to become poor out of 60 households’ respondents, 22 are from poor households and the remaining 38 are from the non-poor households (Zewdu, 2011).

The average time for the nearest market is estimated as 1.675 hours as 1.25 hours for poor and 1.9 hours for the non-poor households. According to this result, the non-poor households traveled much distance to reach the nearest market center than the poor households. This implies that proximity to the nearest market does not affect poverty status of the community within the time set i.e. 1.675 hours (ibid).

Poverty by household characteristics such as age of the household head, divorce or separation of families, and region, the study found poverty incidence is the highest among families headed by a 30-64 years old person, which is 33%. Those headed by old people (greater than 65 year) have the next highest poverty incidence (29%), while those headed by young people (16-29) have a poverty incidence of 16%.The young people whose age is between 16-29 have the lowest level of poverty incidence, which is 11%. One possible reason for individuals to be absolutely poor is divorce or separation of families. The divorced families are not poorer than married in rural areas, found modest differences between married and divorced families in urban areas because in rural areas when families are divorced, families will retain their land rights and may be given better access to productive safety net to protect them from falling into poverty (MoFED, 2013).

Benjamin et al (2012), analysis of the determinants of poverty severity among rural farmers in Nigeria using data from randomly sampled 233 rural farmers in Benue State. The study showed that 87.63% variation in poverty severity was explained by variations in the specified explanatory variables. Furthermore, at 5% level of significance, the critical determinants of poverty severity among the respondents were economic efficiency, household income, dependency ratio, ratio of food expenditure to total household expenditure, farm size, access to credit, household production enterprise structure, extent of household production diversification, extent of production commercialization, expenditure on education, access to agricultural extension services, membership of cooperative societies or other farmers’ associations, market access, total value of household assets, household size and formal education(MoFED, 2013).

2.3 Research Gap

This research flit the study gap by including the most crucial demographic and socio economic variables to the previous studies new variables, such as household access to remittance, household involvement in community association, age square variable, and by focusing at grass root level poverty analysis rather than national picture. This is because poverty by its nature individual centered and its degree of intensity and manifestation differs from location to location.

Therefore, in this research unit of analysis was rural households of Damot Gale woreda, both qualitative and quantitative method of data collection method was used and consumption per capital was employed to indicate the standard of living in study area rather than using income as welfare indicator. Which is different from (Getaneh, 2011), (Regassa E. Namara, Godswill Makombe, Fitsum Hagos, Seleshi B. Awulachew, 2007), who used income as welfare indicator and aggregate level study and this study was also different to Ahmed Mohamed(2013) by its disaggregate nature.

We take consumption as an indicator of welfare because consumption better captures the long-run welfare, can reflect households’ ability to meet basic needs, ability of household’s access to credit and saving at times when their income is very low. Hence, consumption reflects the actual standard of living (welfare) (World Bank, 2005).In most developing countries, income report of households is likely to be understated compared to consumption expenditure report. Income is so erratic and seasonal that it may be very difficult for respondents to recall (FDRE, 2012).

CHAPTER THREE

3. Methodology

3.1 The Study Area

Damot Gale is one of 12 woreda’s in Wolayta zone of SNNPR in Ethiopia. It is located at 139km southwest of the Hawassa town which is the capital of Southern Regional State and 365km from Addis Ababa in the southern direction. Geographically, it is located between 60 53‘- 70 6’ 30” North latitude and 370 46’ – 370 58’ 40” East longitude. It has an altitude ranging from1501- 2950 meters above mean sea level. Mount Damota is the highest peak in the area. The study area covers an area of 24285.861 hectare.

Damot Gale woreda is divided in to agro-ecologic zones such as Dega or highland (25.3%), Woinadega or midland (61.2%) and Kola or lowland (13%) Damot Gale woreda agricultural and rural development office report (2014).Woinadega dominates the study area which has bimodal distribution of rainfall. Mean annual rainfall ranges between 1001-1400 mm (RFEDB, 2013) as cited in (Tesema, 2015).

The study area is bordered on the South West by Sodo Zuria, on the North West by Boloso Sore and Damot Pulassa, on the North by Hadiya zone, on the East by Duguna Fango, and on the South East by Damot Woyde. Based on the CSA (2011) estimation and Woreda Finance and Economic Development report, Damot Gale has a population of 177,570 out of this male 103,011 and female 74,559. The total households of the district are 30,767 of which male households, 26,417 and female 4,350 and has a total of 31 rural kebels1. Like other parts of the region agriculture is the main means of livelihood for the population both in terms of crop production and livestock.

1An administrative unit below the wereda are Kebele, which consists of a number of Communities (villages).Wereda is a local administrative unit, which together form Zones. In this study case in each kebele there are three communities(a total of 12 communities(villages)

Figure 3.1 .Map of the study area extracted from CSA, 2007.

3.2. Data Type and Source

In this research primary data which is collected from the study area (Damot Gale woreda) was used. Information on the demographic and socio economic condition of the households was collected through structured questionnaires by close ended elicitation format with open ended follow up questions. The structured questionnaires were posted to the heads of the households with face to face interviews. Interviews contained inquiries about demographic and socio-economic aspects - age, sex, marital status, household family size, household labor force, household head education level, household head health status, cultivable land size, improved agricultural input use, household market

access, off farm income, ownership of oxen, other livestock unit except oxen, saving behavior of household and access to credit, in the study area.

Secondary data obtained from the zonal and woreda administration offices, woreda agricultural and rural development department/office, finance office and CSA. Both quantitative and qualitative methods were used to analyze rural poverty as they complement each other. For the qualitative data key informant interview and focus group discussions were employed. Therefore, this research used both quantitative and qualitative research methods.

3.3 Sample Size Determination

The sample size in this study was determined by using the minimum sample size formula of Fowler(2001) and then adjusted for the total population of the study area by Cochran’s sample size formula (Cochran, 1977) as shown below

()() = − − − − − − − − − − − − − − − − − − − − − − − − − (3.1) The researcher has been decided to take that true margin of error may exceed the acceptable margin of 6%2 with confidence level 94% and estimated proportion of an attribute that was present in the population p=0.5.

1.88 ∗ (0.5) ∗ (0.5) = = 245 − − − − − − − − − − − − − − − − − − − −(3.2) 0.06 In order to calculate the final sample size, we have considered the total population of the study area. Therefore, Cochran’s (1977) correct formula was used to calculate the final sample size in the study area. = − − − − − − − − − − − − − − − − − − − − − − − −(3.3) 1 +

The total number of population in 31 kebele was about 177,570 male 103,011and female 74559 and the total household was about 30,767 male 26,417 and female 4,350.Total households in selected four kebeles 3779 (CSA,2011) estimation.

2Even though 3%, 5% and 10% precision levels are the most common,7% precision level with confidence level of 95% and P=0.5 is also used in the standard literature. Israel G., 2013 as cited in. (Meneyahel, 2015). Depending on these standard literatures this study was used 6% perception level with confidence level of 94% and p = 0.5.

Therefore, = = = = 235 − − − − − − − −(3.4) . .

Table 3.1 Sample Size of the Kebeles

Name of Kebele Total population Total household Sample households

Damot Mokonissa 5215 990 67

Gacheno 3702 718 45

Wandara Gale 6073 1238 75

Shasha Gale 4200 833 48

Total 19190 3779 235

Source: own computed proportion to size Therefore, the final sample size for the woreda was 235 which is the sum of four kebels

3.4. Sampling Technique In the study area, farming households were responsible for making day to day decision on farm activities. Thus, households were the basic sampling units in order to get quantitative and qualitative data on the determinants of rural poverty in the study area. A two-stage sampling technique was employed to get the required primary data. At the first stage, Damot Gale woreda was selected purposively because it was one of the food insecure woreda in wolayta zone. In the second stage, four villages (Damot Mokonissa, Wandara Gale, Shasha Gale and Gacheno) were selected by simple random sampling techniques out of 31Kebels in the woreda. The sample size was determined by using Cochran’s (1977) formula. A probability proportion to size (PPS) is employed to determine sample size from each kebele, accordingly 235 households were selected through systematic random sampling techniques. The first household was selected by lottery method and the rest survey points selected by interval.

3.5 Approaches to Measuring Poverty and Unit of Analysis

Based on the available information on welfare indicator i.e. consumption per adult equivalent and poverty line. The summery measure of poverty was made by using the

most common measure of poverty FGT (Foster, Greer, and Thorbecke (1984) classes of poverty measure. Consumption per adult equivalent was used as an indicator of welfare because consumption better captures the long-run welfare, can reflect households’ ability to meet basic needs, ability of household’s access to credit and saving at times when their income is very low. Hence, consumption reflects the actual standard of living (welfare). In most developing countries, income report of households is likely to be understated compared to consumption expenditure report. Income is so erratic and seasonal that it may be very difficult for respondents to recall (FDRE, 2012).

Cost of basic needs approach(CBN) was used to determine poverty line, it takes into account both the food and non-food requirements, is the most widely used method of estimating poverty line because the indicators were more representative and the threshold was consistent with real expenditure across time, space and groups.

Steps to establish poverty line  Food poverty Line  Stipulate a consumption bundle that is deemed to be adequate for the above mentioned level of minimum caloric requirement.  Pick a nutritional requirement for each food items which yield 2200 Calories per person per day.  Estimate a quantity of each bundle in gram that yield a calorie requirement of that food item per person per day.  Then convert quantity in gram of each item consumed to kilogram.  To obtain annual adult equivalent consumption of food items multiply food items in kg by 365 days.  Then these bundles giving the minimum caloric requirement were valued at local market prices to get a food poverty line of the study area. Which become minimum amount of money that a household needs to purchase basic-needs of food bundle.

 Non-Food poverty Line

After setting the food poverty line a specific allowance for the non-food goods was made i.e. with the spending of the poor was added to the food poverty line which will yield the overall poverty line.

According to World Bank (2005), there are different approaches to take part of measuring non-food poverty line Practice varies widely from one analyst to the next. But most studies set the poverty line as a share of mean expenditure/income or identified the poor using some percentage (e.g. 20%, 25%) of the income or expenditure distribution. Therefore, in this study to account for the non-food expenditure, the food poverty line is divided by the food share of the poorest quartile or quintile (Ravallion, 2003) & (MoFED, 2013).

Finally the analysis in this paper focused on poverty among households; if a household was deemed to be poor, all its members were counted as poor. The implicit assumption here is that all individual members of a household benefit equally (or in a constant proportion, depending on their age and gender, called adult equivalence scale) from the household's expenditure .we have used absolute poverty lines for the analysis in this paper.

3.6. Data Collection Techniques and Instruments

The researcher used kebele agriculture development workers to collect primary source data. Before entering to survey, the development agents were given a training mainly focusing on the contents of the questionnaire and procedure of survey. Observation and discussion with woreda as well as kebele governmental officials and expertise was held by the researcher. The key points were prepared for discussion with key informant, employers and governmental officials. Structured and semi-structured interview questionnaires were designed to collect quantitative data.

3.7. Method of Data Analysis

To achieve the objectives of this study, different methods of data analysis were used. The study used descriptive, FGT method and econometric analysis. The descriptive analysis uses percentages, graphs and tabulations to explain the magnitude of poverty in different socio economic characteristics of the rural households, Foster-Greer-Thorbeck class of poverty analysis was used to determine the proportion of households who were living below and above poverty line and the magnitude of poverty and binary logit econometric analysis was used to identify the effect of determinants of poverty on probability of being

poor in rural households in the study area. Tools and statistics used in descriptive and econometric are generated with the help of econometric software STATA and DAD.

3.7.Empirical Model

When the dependant variable in regression model is binary the analysis could be conducted using linear probability or index models i.e. logit or probit. But the result of linear probability model may generate predicted values less than zero or greater than one, which violate the basic principles of probability. In addition, the LPM is encountered with problem of non-normality of disturbance term and questionable coefficient of goodness of fit (R2). However, the index models logit or probit models generate predicted values between 0 and 1 inclusive (0 ≤ = ≤ 1)they fit well to the non-linear relationship between the probabilities and the explanatory variable. Each model has its own strength and weaknesses, but in this study logit model is preferable to probit model as it has more plausible feature such as simplicity: The equation of the logit CDF is very simple, while the normal CDF involves an unevaluated integral and interpretability: The inverse linearzing transformation for the logit model is directly interpretable as log-odds, while the inverse transformation probit model does not have a direct interpretation ( Gujarati & porter, 2009).

3.7.1. Binary Logit Model

The logit model is designed to analyze qualitative data reflecting a choice between two alternatives, which in this case are the poor and non-poor. The choice of the logit model is premised on the fact that ordinary least squares assumes a continuous dependant variable while in the case of poverty the response is a binomial process taking the value 1 for poor and 0 for non-poor. The parameters of this model was be estimated by using the maximum likelihood estimation rather than the movement estimation in which OLS regression technique rely on. The logit method gives parameter estimates that are asymptotically efficient, and consistent. Indeed, the logit approach is known to produce statistically sound results ( Gujarati & porter, 2009) Probability of being poor is specified as the value of the cumulative distribution function which is specified as function of the explanatory variables. The equation is of the form:

= + + + + ...... + + − − − − − − − − − −( 3.5)

Where,

Y= probability of a household being poor or non-poor

= ()

= Coefficient of the explanatory variables.

Xi= Explanatory variable.

ε = Distarbance(stochastic)term.

Noursis (1994), pointed out that in logistic regression model, we directly estimate the probability of an event occurring. For the case of a single independent variable, the logistic regression model can be written as

Pr() = Or equivalently Pr() = − − − −( 3.6) ( )

Where, β0 and β1 are coefficients to be estimated from data, Xi is the independent variable; e is the base of the natural logarithm.

For ease of exposition the model can be written as (for more than one independent variables)

Pr() = Or equivalently Pr() = − − − − − − − − − − − (3.7 )

This particular study was deal about the probability of being poor or not and this expression expressed in mathematical form as follows:

The probability of being poor (an event occurring) as the form:

1 ( = 1/) = ( = 1) = = − − − − − − − − − − − −( 3.8) 1 + 1 +

= ++ + − − − − + + − − − − − − − − − − − − − − − (3.9)

Note: - the error term also follows logistic distribution

For a non-event (non-poor) cumulative logistic distribution, representing the probability is just (1-pi) i.e.

1 − ( = 1⁄) = = − − − − − − − − − − − − − − − − − − − −( 3.10) 1 +

Therefore, by dividing equation (3.8) by equation (3.10 ) we can result in the odds-ratio in binary response, which is as stated below:

( = 1⁄) ( = 1) 1 = = = = − − − − − (3.11 ) [1 − ( = 1⁄)] 1 − ( = 1)

Equation (3.11) is simply the odd-ratio in favor of household falling below the poverty line. This is the ratio of the poverty that a household will be poor to the probability that it will not be poor.

When we take the natural logarisim of odd-ratio of equation (3.11) will result in logit model as we can see below

( = 1) = = 1 − ( = 1)

= + + + + + + + + +

+ + + + + + + +

+ − − − − − − − − − − − − − − − − − − − − − − − (3.12)

Assumptions of Logistic model

1. Assumes a linear relationship between the logit of the independent variable and dependant variables, however, does not assume a linear relationship between the actual dependant and independent variable 2. Independent variables were not linear functions of each other, i.e. perfect multi- collinearity makes estimation impossible. 3. The model was correctly specified i.e.  The true conditional probabilities are a logistic function of the independent variables;  No important variables are omitted;  No extraneous variables are included; and  The independent variables are measured without error.

Based on the above justification, we specified the logit model for probability of being poor or not-poor and determinants of poverty as follows:-

= + + + + + + + +

+ + + + + + +

+ + + + − − − − − − − − − −( . )

Therefore Yi = 1 if household is poor and = 0 if household is not poor, is regression parameters, is the error term and the explanatory variables will be defined under the variable description section (3.8) in the next section. The regression was estimated by Maximum likelihood technique.

3.8. Variable Description and Their Expected Sign

1. Dependant Variable (Y) or Probability of Being Poor: which is dummy and take value =1 if the household is poor and = 0 if the household is not poor.

2. Household head Age (Ag): It is continuous variable which represent the number of years of household heads. In this research it is assumed that as age increases rural households would acquire knowledge and experience through continuous learning which help them to actively participate in different activities that help to them escape from poverty. Therefore, it was hypothesized that this variable has negative relation to probability of falling to poverty.

3. Household age square (Ag2): It is age square of the household head which was used in order to capture the non linear relationship between poverty and age. In this study it was hypothesized that when age increases probability of failing to poverty decreases.

4. Household head sex (Sex): Dummy variable taking value = 1 if the household is male headed and 0 otherwise. According to Buvinic and Gupta (1997) as cited in (Rajaram, 2009) female headed households in general have more dependents and thus have higher non-workers to workers ratio compared to other households, they work for lower wages and have less access to assets and productive resources compared to men, owing to gender bias against women and bear the burden of household chores that result in time and mobility constraints compared to male-heads. Therefore, Male headed households are expected to have better chance of escaping poverty than women headed.

5. Household Family Size (FS): Represents the total number of individuals (in adult equivalent units) in the household. The expectation is that with large households, the number of dependents would be high and where there is limited chance for full

employment, household’s saving would decrease. Households with large family have more probability of falling to poverty than those with lesser family size.

6. Dependency ratio of the household (Dr):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 .It is important because it shows the ratio of economically inactive compared to economically active. The dependency ratio is thought to be negatively related to income of households. Therefore, in this study it is expected that it has positive effect on probability of being poor.

7. Household Head Marital Status (Ms): This refers to the marital status of household head. The assumption is that households headed by married individuals are supposed to be larger in family size .Large family in developed countries mean large labor force which in turn reduces the incidence of poverty but in developing country the reverse holds true. Therefore, in this study the expectation is household heads married are less likely to escape poverty. If the head of the household is married, it takes the value of 1, 0 otherwise.

8. Household Education (Edu): Education widens horizons of an individual. Since education creates awareness of the benefit of new technologies, a negative relationship is expected between the level of respondent’s education and poverty. In this study education is a continuous variable measured in years of schooling.

9. Household Heath (HE): health is the decisive factor for life, one with poor health conditions will have a poor living standard (welfare). Household labor is often one of the few means of earning income the rural people can rely upon. Lack of proper health will make the family survival challenging because poor health make the family unproductive. Therefore, 1 if the household had one or more seriously ill member 12 months prior to the administration of the survey and 0 otherwise. In this study it is hypothesized that if illness strikes, not only working days are lost, but also part of saving will be spent for medical treatment, as a result of which the probability of being trapped in poverty is likely to rise.

10. Household Cultivable Land Size (LS): Represents size of cultivated land the household owned. Since land an important asset and factor of production, in this study it is hypothesized that households with large holdings have better opportunity of obtaining more yields, and hence, could reduce the likelihood of being trapped in poverty.

11. Household Ownership of Oxen (OX): This represents the number of oxen the household owned for agricultural activates. For the draught reduction power it generates, no doubt that ox is most important in the agricultural systems of rural Ethiopia including the study area. Households with oxen cultivate significantly more land and obtain from others via share cropping or renting. Thus, it is hypothesized that the probability of the household being poor decreases with an increase access to ownership of oxen.

12. Households Total Livestock Unit of Livestock Other Than Oxen (TLU).It includes cows, steers, heifers, calves, sheep and goats, which in rural areas are amongst the dominant means of saving and the sale of which or their products would be a source of income during economic hardship. This variable also includes donkeys, mules and horses the household owned. Households having equine are able to transport their products to the market, flour mills, etc., by their own and even can generate income through renting. The expectation is that those households with more of these animals have better opportunity of smoothing their income overtime and hence of escaping out poverty.

13. Household Head Use of Improved Agricultural Inputs (IAgI): The expectation is that households that use improved agricultural inputs at a recommended rate obtain high crop yield, and hence, have wider opportunity of being better escape poverty. 1 if the household use inputs at a recommended rate and 0 otherwise.

14. Household access to Remittance (REM): It is the households’ access to remittances which is assumed to affect the per-capita income of the household positively and reduces the incidence of household poverty. The likelihood of a household receiving remittance increase choice of diversification in to off farm and nonfarm activities according to ( Bezemer& Lerman,2002) as cited in (Adugna eneyew & Wagayehu

Bekele, 2012).Therefore, it is dummy variable it takes value 1,if the household has access to remittance and 0, otherwise.

15. Household involvement in Community association (CA): It is household’s involvement in community associations to reflect social norms and relationships in a community it has a greater role when households/individuals face risk or different uncertainties (T.Bruck, 2001). To capture possible effects of community involvement, a dummy variable for whether or not someone in the household participates in community association. 1 if the household head belongs to a community group and 0 otherwise. 16. Household Head Participation in off-Farm Activities (OFI):Non-farm activities apart from diversifying and providing income are important source of employment during slack period in the agricultural calendar. Therefore, the expectation is that households engaged in non-farm activities have better chance of escaping from poverty.

17. Household Head Saving Habit (Sav): This refers that households that have the culture of saving could invest and even use their saving at times of economic hardship, and hence, have better chance of escaping from poverty. 1 if the household has the habit of saving and 0 otherwise.

18. Household Head Access To Credit (crd): Households having access to credit have better chance of involving in non-farm activities, purchasing ploughing oxen, etc. as a result of which households could increase and diversify their income and escape out of poverty.1 if the household has access to credit and 0 otherwise.

19. Household Market access (MAc): Represents distance to the nearest market (in hours). The assumption is that the nearer the rural area to the market, the better accesses to markets and the lower the chance of falling into poverty.

Table-3.2: Variables and their Expected Sign

Variable Description of variable Measurement Expected Name sign Poverty Probability of being poor Dummy (1=poor, 0 =non-poor) Dependant Ag Age of the household head Continuous variable measured in years - Sex Sex of the household head Dummy(1=male,0=female) -/+

FS Family size of the households in Continuous variable measured in + Adult equivalence number Dr Dependency ratio of household Continuous variable measured in + percent MS Marital status of household Dummy(1=married,0=unmarried+ + divorced + widowed ) Edu Education of household head Continuous variable measured in years - of schooling HE Health of the household Dummy(1=if the household has no - health problem,0 otherwise) LS Total size of cultivated land Continuous variable measured in - hectares OX The number of oxen owned Continuous variable measured in - number TLU Total livestock except oxen Continuous variable measured in - owned by farm household number IAgI Improved agricultural input Dummy( 1 if the household use - improved seed, 0 otherwise) REM Household access to remittance Dummy(1,if HH has access to remittance, 0 ,otherwise) CA Household involvement in Dummy(1 if the household - community participation participate,0 otherwise) OFI Household off farm activity Continuous measured in birr - Sav Saving behavior of household Dummy(1,if the HH has saving - behavior,0 otherwise) Crd Household access to credit Dummy(1 ,if the household access - credit, 0 otherwise) Mac Household market access Continuous variable measured in hours -

CHAPTER FOUR

4. Results and Discussion

4.1. Computing Poverty Line

Following the steps mentioned in methodology section 3.5 we established poverty line for the study area in the following form.

Food poverty Line:

 Total 3Adult equivalent food Expenditure = 2,637,882Birr  25% Adult equivalent population food share = 48063.5Birr  Percentage share of the lowest 25% population= 0.018220489  Food poverty line =2332.177 Birr Non-food poverty line:

To obtain this line we have divided the food poverty line by the food share of lowest 25 percent of expenditure distribution.

 Non food poverty= 2332.177/1.8220489= 1279.9749Birr.

Therefore, Poverty line in the study area = food poverty line plus Non-food poverty line

= 2332.177Birr + 1279.974 Birr

=3612.151Birr From the total of 235 household survey in Damot Gale woreda 132 (56.17%) households were below the poverty line (poor) =3612.151Birr and 103 (43.82%) of the households were above the poverty line (non-poor).

4.2 The Extent and Measure of Poverty

The magnitude and measure of Poverty in the study area is computed based on information available on a welfare measure, in this study case adult equivalent consumption level and a poverty line, for each household or individual. To construct a summary measure of the extent of poverty there are varies types of poverty indices but the most widely used and employed in this study was the well-known FGT (1984) class of poverty measures.

3Based on the assumption of equal welfare/living standard in a given household i.e. adult equivalence scale was used to adjust demographic differences on consumption either food or non food i.e. age and gender based classification.

Incidence of Poverty or Headcount Index (P 0) This is the share of the population whose consumption is below the poverty line in the study area, i.e. the share of the population that cannot afford to buy a basic basket of goods either food or non-food with the stated amount.

= − − − − − − − − − − − − − − − − − − − − − − − − − − − ( 1 )

Where, P0 = poverty head count ratio

Np = Number of households below the given poverty line

N = Total number of households in the sample

132 = = = 0.5617 = 56.17 − − − − − − − − − − − −( 2 ) 235 It shows that 56.17 percent of the sampled households in the woreda were below the poverty line. It is greater than the rural poverty of the country and Southern Nation Nationalities Regional State which is 30.4% and respectively MoFED (2013, pp 30).

Depth of poverty or Poverty Gap Index (P1) Poverty gap index measures individuals on average fall below the poverty line; and it is a percentage of the poverty line. More specifically, define the poverty gap (Gi) as the poverty line (z) less actual consumption (yi) for poor individuals; the gap is considered to be zero for everyone else. Using the index function, we have

= ( − ) ∗ ( < ) − − − − − − − − − − − − − − − − − − − − − − − ( 3) 1 Z − yi = − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − −( 4) N Z 1 Gi = − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − −( 5) N Z 1 Gi 1 = = (52.20007) = 0.2221 − − − − − − − − − − − − − − − − − −( 6) 135 3612.151 235 The result from the survey shows the poverty gap (consumption shortfall) of poor to reach poverty line in is 22.21 percent. In other words, it estimates the total resources needed to bring all the poor to the level of the poverty line consumption or the woreda needs to mobilizes resources equal to 22.21 percent of the poverty line for every adult equivalent

individuals and distributes these resources to the poor in the amount needed so as to move them to poverty line. Poverty severity or squared poverty gap (P2)

This index takes into account inequality among the poor, it is simply a weighted sum of poverty gaps (as a proportion of the poverty line), and hence, by squaring the poverty gap index, the measure implicitly puts more weight on observations that fall well below the poverty line. Formally:

1 G = (α ≥ 0) − − − − − − − − − − − − − − − − − − − (5 ) N z 1 2 = ( 25.8312) == 0.109 = 10.9 − − − − − − − − − − − −(6) 235 Even though households, whose consumption expenditure lies below poverty line have common name “poor” the degree of poverty varies from one to another. Therefore, poverty severity index measures variation in the poverty level of individual households. The result indicates that 10.9 percent variation among poor households in the study area.

Table-4.1 FGT measure of Poverty status of four surveyed Kebles

Keble Poor Non-poor OBS %Share of P0 P1 P2 poor

D/Mokonissa 36 31 67 27.27 0.1532 0.057 0.027

W/Gale 44 31 75 33.33 0.1872 0.072 0.0348

Gacheno 28 17 45 21.21 0.1191 0.0585 0.0304

Shasha Gale 24 24 48 18.19 0.1021 0.0335 0.0168

Total 132 103 235 100 0.5617 0.221 0.109

Source: Own computed from survey data. In table4.1.We can see distribution of poverty prevalence in four sampled Kebles. Wandara Gale has 33.3% share followed by Damot Mokonissa 27.27% share of poor from total households below poverty line. When we look at poverty head count index, 18.72 percent of the households from the total surveyed population poor in Wandara Gale stood first and followed by Damot Mokonissa (15.32%), Gacheno (11.91%) and Shasha

Gale (10.21%) respectively. For Kebles whose poverty proportion was high is due to their significant number of the dependant population with small cultivable land.

4.2.1Descriptive Analysis

In this section we discussed descriptive analysis of data to present the poverty level in different demographic structures, extent and proportion of poverty due to differences in rural households by using graphs, percentage, charts, figures and tables. A total of 235 households were surveyed in Damot Gale woreda and the results of the study revealed as follows.

Consumption Expenditure in Damot Gale woreda

In the study area inhabitants consumption expenditure on amount of food items differs based on livelihood zone. Most of the time the root and maize crop lively hood zone inhabitants consume maize as major crop. Besides it they consume other cereals and root crops .Where as the households living in wheat and Barley lively hood zone consume mainly wheat and barley and other crops in less quantity than their main production food items. Most of the food commodities were either produced or purchased from the local market and all of the non-food commodities are purchased from nearby town Boditi.

The average food and total consumption expenditure per adult equivalent per year were Birr 2540.47564 and Birr 4025.957977 respectively.

Table-4.2Annual households’ consumption Expenditure in Damot Gale Woreda, 2016

S.No Households consumption Expenditure Average

1 Per-capita food consumption Expenditure 2223.69353

2 Per-capita non-food consumption Expenditure 956.412337

3 Per-capita total consumption Expenditure 3180.10

4 Adult equivalent food consumption Expenditure 2540.47564

5 Adult equivalent non-food consumption Expenditure 1485.4831

6 Adult equivalent total consumption Expenditure 4025.957977

Source: Own computed from survey data.

Poverty and Household Characteristics

According to the CSA (2007), the Damot Gale woreda has an average of 4.3 persons per household, which was higher than the regional average family size and in this study we found 5.47 persons per household. While in poor household is average family size was 6.67 and a non-poor household is 3.95.This result shows that poor households have larger family than non-poor households. Therefore, it is a good clue to know the demographic factors influence on prevalence of poverty.

Table-4.3Households’ age and sex composition and their economic activity

Head of household Economic activity Age Family member

Househ

Perc Age Age age Male Femal

olds

0-14 15-64 >64 e

Total Working working Not Total Not working

Male Female Poor 102 30 132 337 543 880 61.76 346 475 66 440 440 Not poor 88 15 103 245 162 407 39.8 105 294 8 210 197 Total 190 45 235 582 705 1287 54.77 451 769 74 650 637

Source: survey data result, 2016

From the total surveyed households there are582 working and 705 not working4 family members which are about 54.77 percent of the family members were not working. Out of 880family members living in poor households 543were not working and 337 working i.e. about 61.76 percent were not working .Where as in non-poor only 39.8 percents were not working.

4 Not Working family members were the family members who didn’t contributed to the family income 12 months before data collection. When we consider age less than 14 & greater than 64 as not working (dependant), practically there are a lot of peoples contributing to their family income in the stated ages. Therefore, to know the exact contribution of each individual family member we collected data and resulted with the above table output. See page 70 questionnaire format.

Poverty and Family Size

Figure-4.1 Households’ family size and consumption distribution around poverty line Poverty Line Consumption= 20000 Household Family size and poverty

15000

Expen diture 10000 3612.151

5000

0 0 5 10 Family size

Source: own computed from survey data using stata.

The figure-4.1 shows that family size and poverty have direct relation. The distribution of poor below poverty line increases when family size increases from left to right (in x-axis) or line assigned by family size.

Table-4.4: Family Size and Poverty % Share of poor Family Size Poor Noon poor Total 1-5 43 87 130 33 6-11 89 16 105 67 Total 132 103 235 100 Source: Own computed from survey data

In above table we can see high poverty incidence in households whose family size greater than the average family size and relatively lower in families with family size less than the mean. Therefore, it indicates family size and poverty have positive relation.

Household head age and Poverty Figure-4.2Households age and consumption distribution around poverty line

Source: own computed from survey data using stata.

The average age of households in this study is 46 year. In the above figure as age increases, poverty increases at the initial stage and reaches a certain turning point 49.9. When a household head become mature enough (from young to old age), hold the most important assets such as land of his/her own, oxen, hold agricultural tool, become member of social security i.e. ‘Edir’,etc. Therefore, the finding of this study shows poverty is higher for younger households’ heads than their senior households (see: detail explanation in econometric analysis section).

Poverty and Household Head Sex

Table-4.5 Household Head Sex and Poverty Sex Poor Non-poor Total % poor Male Head 102 88 190 53.68 Female Head 30 15 45 66.67

Total 132 103 235 56.17

Source: Survey data result, 2016.

Given that women generally and traditionally have less access to land and other productive assets, it was hypothesized that women headed households are more likely to be vulnerable and hence are trapped in poverty.

As we can see in table 4.5 about 66.6%of female headed households were poor from the total surveyed female head households. Which shows they were more vulnerable to poverty than male headed households; about 53% were below the minimum living standard out of total male headed households. This is possibly due to a shortage of time spent on land preparation, cultivation and weeding in addition to their home responsibilities.

Poverty and Cultivated Land Size Table-4.6 Cultivated Land Size to Poverty Distribution

Size of land Below poverty line Above poverty line Total

Less than 0.75 hectare 108 (73%) 41 (27%) 149

Above 0.75 hectare 24 (28%) 62 (72%) 86

Total 132 103 235

Source: own computed from survey data, 2016.

The mean land holding in the study area was 0.75 hectare. The sum up the influence of land holding on poverty we divided land holding in to two i.e. above mean and below mean land owners. Therefore, when land holding is less than the mean 73% poor and only 27% non-poor. Whereas land holding greater than average, poor 28% and non-poor.

72%.

Health and Poverty

In this study we found that family health as critical determinants of rural poverty. Out of 235 surveyed households 31% responded as their family members were frequently sick. Out of these respondents 67% were below the minimum threshold (poor) and 33% households were non-poor. The finding clearly states that as household family member frequently face health problem the probability of failing to poverty increases.

Ownership of Oxen and Poverty

Figure-4.3 oxen ownership and poverty

150 27 100 above PL 50 107 51 below PL

25 25 0 0 Oxen 0‐1 Oxen2‐3 oxen >4

Source: Own computed from survey data.

In the figure 4.4 above 79.85%households were below poverty line(poor)and 20.15% were above poverty line(non-poor) out of the 134 households in category “Oxen zero to one” and the reverse output was revealed that the number of oxen increased to category “oxen two to three” which is 32% were poor and 68% were not poor. Therefore, when oxen ownership increases poverty decreases.

4.2.2 Econometric Analysis 4.2.2.1 Model Specification A model specification error can occur when one or more relevant variables are omitted from the model or one or more irrelevant variables are included in the model. It can substantially affect the estimate of regression coefficients. Moreover, in this study, the model specification errors were checked by linktest, the test of hat and hatsq were 0.000 and 0.001 respectively which are significant. Therefore, it shows that the linktest has failed to reject the hypothesis that the model is specified correctly. Accordingly, it seems to us that we don’t have a specification error (Appendix-5).

In addition to the basic descriptive statistics, the logistic regression model was employed to identify the determinants of household poverty in the study area. The variables included in the model were tested for the existence of multi-co linearity, if any. Contingency coefficient and variance inflation factor were used for multi-collinearity test of discrete and continuous variables, respectively (see Appendix-6 and Appendix-8).

Contingency coefficient value ranges between 0 and 1, and as a rule of thumb variable with contingency coefficient below 0.75 shows weak association and value above it indicates strong association of variables. The contingency coefficient for the discrete variables included in the model was less than 0.75 that didn’t suggest multi-collinearity to be a serious concern. As a common practice continuous variable having variance inflation factor of less than 10 are believed to have no multi-collinearity and those with VIF of above 10 are subjected to the problem and should be excluded from the model (Gujarati, 2009, pp340).

So as to identify the major determinants of rural poverty in Damot Gale woreda the dependent variable, probability of being poor was regressed against various explanatory variables. The regression table revealed that binary logistic model managed to predict 78% of the responses correctly.

Apart from percent correct predictions, the model Chi-Square with “n” degrees of freedom and Hosmer and Lemeshow’s are used to test goodness-of-fit test. Accordingly, p-values associated the Chi-Square with 19 degrees of freedom. The value of .0000 indicates that the model as a whole is statistically significant that shows the model fit the data well (see Table-4.7).

Another commonly used test of model fit is the Hosmer and Lemeshow’s goodness-of-fit test. Hosmer-Lemesshow goodiness-of-fit statistic is computed as the Pearson chi-square from the contingency table of observed frequencies and expected frequencies. Similar to a test of a two-way table, a good fit as measured by Hosmer and Lemeshow’s test yield a large p-value. Therefore in this study the test result show that p=1 this suggests that the model is correctly fitted with the data (see Appendix-7)

Robust logistic regression was used to control for hetroscedasticity in binary outcome models. Hetroscedastiscity in binary outcome models will affect both the “Betas” and their standard errors (Wooldridge, 2001) .In this particular study both regression i.e. earlier regression and robust logistic regression have the same result. None of the

coefficients changed, but the standard errors and Z values are a little different. Had there been more heteroscedasticity in these data, would have probably seen bigger change. Therefore, this model is free from heteroscedasticity problem (see Appendix-10).

4.2.2.2 Estimation of Determinants of Rural Poverty in Damot Gale Woreda The binary logit model was used to estimate the determinants of rural poverty in Damot Gale woreda. The estimation result of the model is presented in the following table:

The variables that are positively related with the probability of being poor are household head sex, age, marital status, household family size and dependency ratio. Negatively correlated with the probability of being poor were Age square, credit use, head education, family health, and cultivated land size, off farm income, total livestock unit, community association, remittance, agricultural input use and saving behavior of the rural households.

In the table 4.7 above out of 18 explanatory variables, 13 of the variables: household age, age squire, household size, dependency ratio, off farm income, remittance, family health,

credit access, market access, household sex, oxen owner ship and cultivated land size have a significant effect on the rural households failing in to poverty at the significance level at 1%, 5% and 10%. The negative values of explanatory variables in the table above indicate that when the unit change in independent variable lead to decrease in probability of being poor.

Marginal Effect for Logit Regression

Since the logit model we employed for regression analysis is not linear, the marginal effect of each independent variable on the dependant variable is not constant but it depends on the value of the independent variables. Thus, marginal effects can be a means for summarizing how change in a response is related to change in a covariate. For categorical variables, the effects of discrete changes are computed, i.e., the marginal effects for categorical variables show how P(Y = 1) is predicted to change as Xk changes from 0 to 1 holding all other Xs equal.

Whereas for continuous independent variables, the marginal effect measures the instantaneous rate of change, i.e. we compute them for a variable while all other variables are held variables constant .That means in this study change in the probability of being poor with a unit change in continuous independent variable (Greene, 1993).Thus, opposed to linear regression case, it is not possible to interpret the estimated parameters as the effect of the independent variable up on being poor. However, it is possible to compute the marginal effects at some interesting values of the significant explanatory variables. We can see in table 4.8 below.

Household Size and Poverty The size of house hold was positively related and the coefficient is statistically different from zero at 1 percent significance level. Holding all other variables constant at their mean values, it was expected that household family size increase by one adult equivalent individual, the probability of a household to be poor increase by about 22.0%. This is attributed due to the fact that the average number of children age less than 15 is 2.625) and old aged greater than 64 is 0.503 were larger in poor households than non-poor households with age less than 15 family members average size 1.0288 and age greater than 64 members 0.076.With existing high rate of fertility in rural area, less employment opportunity, weak off farm income participation, member of the family become unemployed and coupled with low rate of payment. Therefore, additional household member shares the limited resources that lead the household to become poor. The result was consistent with the study of (Benjamin et al, 2012), (Abebe, 2011), (Metalign, 2005) and (MoFED, 2013).

Dependency Ratio

Dependency ratio and probability of being poor positively related and the coefficient is different from zero at 5% significance level .Holding all other variables constant on their mean values, as the dependency ratio of household’s increases by one dependent individual, the probability of a household to be poor increases by 29.05%. The possible explanation for it is that as dependency ratio increases, households saving will be low which limits the chance of consumption smoothening during bad agricultural production season. The result was consistent with the findings of (Nsikak et al, 2012), (Maru, 2004) & (MOFED, 2013) .The positive relation of dependency ratio on households' living standard is also consistently confirmed by the community members during focus group discussion.

Age of Household Head

The result of this finding shows that age is significantly different from zero at 5% significance level and positively related with probability of being poor. Whereas the coefficient of age squared is statistically significant at 5%and negatively related with probability of being poor. It indicates that the relationship between age and being poor is not linear.

The positive coefficient for age and the negative one for age squared could indicate a monotonic increasing function of poverty by age until a turning point is reached, after which point the function starts to decrease. The turning point is obtained by -βt / 2βq= . . = = 49.90 years. This shows that the function turns at 49.9years of (.) . age. Therefore, the finding shows as age increases by a unit (one year), the probability of being poor increases at about 7.379% up to the turning point of 49.9 years(49 year, 10 months and 24 days),holding other variables at their mean values. After this turning point holding other variables at their mean values unit increase in age, the probability of being poor decrease at about 0.07394%.

The possible reason for this findings young household heads age was positively related to probability of being poor, while their counterparts (embodied in age-squared variable) are less likely to be failed to poverty for households in study area. These results appear to

reflect the often ignored reality that younger adults face challenges of having limited command on vital resources of farming such as land, oxen and other social assets, which increases their likelihood of falling into poverty. Whereas aged households due to the presence of strong bond among extended families, unreserved cooperation from the families of their independent children that could decrease the probability of failing to poverty with aged household heads. The finding was consistent with finding of (Abrham et al., 2012) i.e. poverty and vulnerability dynamics an empirical evidence from Smallholders in northern highlands of Ethiopia.

Household Market Access

The results of the survey revealed that the variable under consideration is negatively related and significant at 1% significance level with poverty. Holding other things constant, a unit increases in access to market (a decrease in time spent) to the nearest market the probability of the household to be poor decrease by about 1.485%. The possible explanation is that access to markets gives the household an opportunity to be involved in income generating activities or off farm activates, obtain their basic needs at reasonable prices from the competitiveness, it will give the opportunity to sale their agricultural products with fair price. The result obtained is consistent with studies done by (Abebe, 2011), (Getaneh, 2011) & World Bank (2005).

Cultivated Land Size

Land holding is negatively correlated with rural poverty and statistically significant at 1% level. As it was expected, holding other variables on their mean values, land holding increase by a unit (1 hectare) the probability of household to be poor decrease by about 87.8%. The possible reason is income and consumption may go up for a rural household as land holding is increased .This indicates that a household’s ability to generate sufficient economic livelihood depends on the environment in which the land exists for agricultural use. This result is consistent with the findings of (Getaneh, 2011), Dercon (1999), World Bank (1998) and Maru (2004).

Ownership of Oxen

Ownership of oxen was negatively correlated with the probability of a household being poor and the coefficient is significantly different from 0 at10% significance level.AS owner ship of oxen increases by a unit (one ox increase), probability of the household

being poor decrease by about 11.09%.The possible reason is 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 agricultural land and labor resources of the household.

Off-farm income

Off-farm income is negatively related to probability of being poor and the coefficient is significantly different from zero at 10% level as. Other variables being constant at their mean values, a one birr increase in off farm income, the probability of the household to be poor decreases at about 0.0177 percent.

Credit access

Credit access was negatively related to the probability of being poor and the coefficient is significantly different from zero at 10% level. Holding other variables of the model at their mean values, a discrete change in credit access from 0 to 1 ( no access to access), probability of being poor decrease at about 44.52 percent.

The possible reason is credit gives an opportunity to be involved in income generating activities in addition to farm income, so that derives revenue and purchasing power of the household to escape from risk of poverty. Moreover, it helps to smooth consumption when household face with temporary problem.

Remittance

The coefficient of remittance is significantly different from zero at 10% significant level and has negative relation to probability of being poor. Holding other variables at their mean values, a discrete change in remittance from 0 to 1 (“no-access to remittance” to “access to remittance”), probability of failing to poverty decrease at about 44.87%. Because remittance may be used as additional income source to agriculture, used as starting capital to finance off farm activity, to purchase more assets, enables higher investment in business; and facilitate buying more goods, including education and health inputs. The finding was consistent with (Banga et al., 2009) impact of Remittances on Poverty in Developing Countries and (Abrham et al., 2012).

The variables sex of the household head, household marital status and household family health were significantly different from zero at 5%, 10% and 5% respectively. Keeping other model variables at their mean values, a discrete change in sex of head from 1 to 0 (male to female),the probability of falling in to poverty increases for female heads at about 54.83%.When discrete change in households marital status from 0 to 1 (unmarried to married),probability of failing in to poverty increase at about 29.4% for married households and a discrete change in family health from 0 to 1 (“no-health problem” to health problem), probability of failing to poverty for a household increase at about 48.9%, keeping the rest of variables constant at the model.

CHAPTER FIVE

5. Conclusion and Policy Implications

5.1. CONCLUSION The study was conducted to identify determinants of rural poverty in households in Damot Gale woreda. Data was collected from 235 sample households from four kebeles (Wandara Gale, Shasha Gale, Gacheno and Damot Mokonissa).The study used cost of basic needs method to compute the poverty line of the study area by using consumption as an indicator of welfare or standard of living.

Based on the information on welfare indicator i.e. adult equivalent consumption we computed poverty line, which is the combination of food and non food poverty expenditure, Birr 3612.151 i.e. the minimum amount of money required to purchase the consumption bundle in the study area. The poverty incidence, poverty gap and poverty severity were calculated in accordance with the poverty line; and found 0.56, 0.22 and 0.109 percent respectively. Headcount index shows that 56.17% of the households were poor and 43.83% were not poor, poverty gap result implies 22% consumption shortfall from the poverty line and severity result indicate 10.9% variation among poor households.

The descriptive analysis shows highest poverty incidence in households among family size greater than average family size i.e. 6-11 (67%),and lowest in family size less than average 1-5 family size (33%). 66.6% of female households were poor from the total surveyed female head households and 53.68% of male headed households were below poverty line. Out of total surveyed households, 73 (31%) house households responded as their family members were frequently sick;49 (67%) were below the minimum threshold (poor) and 24 (33%) were non-poor and from the total poor households in the study area 87% were holding land size below one hectare.

Determinants of poverty in the study area were computed by employing the econometric analysis i.e. binary logit model and the regression result revealed that family size, household age, age square, marital status, household health, total cultivated land size, off

farm income, oxen owned, head sex, market access, access to credit, remittance and dependency ratio affects poverty status of rural farm households of the study area significant at 1%, 5% and 10%. As a result the following policy recommendations were made in section 5.2 below.

5.2. POLICY IMPLICATION Following the results from descriptive and econometric analysis, the following policy implications are forwarded as alternatives for the effective poverty reduction.

 Implementation of family planning and related measures should be taken to limit household family size.  Land size owned by households was found affecting significantly households’ poverty; as cultivable land size is limited it is important to reduce number of households depending on farm income by introducing agro industries and other nonfarm job opportunities into rural areas. Family planning can also play its role here too. Intensive agricultural practices should also be intensively promoted to enhance productivity on the limited land available.  Access to credit is also negatively correlated with poverty in the study area. It helps the poor households to improve their productivity, create jobs, smooth consumption flows but with a prior saving used as pre requisite to qualify for credit in the form of group lending hinders credit access to the poor in the area. However, poor farmers find group lending inconvenient to access credit from MFI since they are rejected from the group by better offs on one hand and pre requisite saving requirement on the other. Therefore, accommodative credit policy should be employed; meaning that MFIs and other development agencies need to introduce credit policies targeting poorest of the poor.  Market access improves household’s income and reduces probability of becoming poor; hence efforts should intensify to create some sort of market in the vicinities of households and improve road and other infrastructure facilities to established markets.  We have found that female-headed households were more likely to be poorer than men headed households. Promoting female education, empowering the female heads of households, introduction of agricultural packages that can be easily managed by women may enhance such household’s income among other factors.  Negative correlation between poverty and off farm income observed indicates that when off farm income increases poverty decreases, hence we recommend creation of job and business opportunities that can generate off-farm income for the households.

Poverty reduction strategies should target specific locations and specific households as most of the time poverty by its nature is individual centered rather than aggregate. Therefore schemes that can improve incomes of individual households and certain localities should be employed selectively.

REFERENCE

Abdulaziz, Y. (2014, june). Determinantes of Urban Poverty in Ethiopia,Evidence From Arba Minch Town: Ahousehold Level Analysis. Arbaminch, SNNPR, Ethiopia. Abebe, Z. (2011). Dimensions and Determinantes of poverty among Rural Households the case of Itang woreda Gambela. Masters Thesis .

Abrham et al. (2012). Poverty and Vulnerability Dynamics :Empirical Evidence from Smallholders in Northern Highlands of Ethiopia. Quarterly Journal of International Agriculture, No. 4.

Adugna Eneyew & Wagayehu Bekele. (2012, June). Determinants of Livelihood Strategies in Wolaita, Southern Ethiopia. Vol. 1(5) , pp 153 -161. SNNPR, Ethiopia: Agricultural Research and Reviews. Afera, N. (2015). Determinants of Poverty in Rural Tigray:Ethiopia Evidence from Rural Households of Gulomekeda Wereda,. Journal of Poverty, Investment and Development, Volume 10, 99-100. Ahmed Mohamed (2013). Determinants of Poverty in Rural Ethiopia: A house hold level analysis. Masters Thesis. Lund University. Alemayehu, G. (2005). Determinants of Poverty in Kenya: A Household level analysis. Economics working paper . Asmamaw, E. (2004). Understanding Poverty : The Ethiopian Context ( A Paper presented at The Gambia AAPAM Roundtable Conference,. Addis Ababa. Ayalneh, B. (2005). Determinants of poverty in rural Ethiopia. Quarterly Journal of International Agriculture , No. 2: 101-12. Ayalneh, B. (2011). Analysis of Poverty and its Covariates among Smallholder Farmers in The Eastern Hararghe Highlands if Ethiopia. Journal of Development and Agricultural Economics Vol. 3(4) , pp. 157-164. Banga et al. (2009). Impact of Remittances on Poverty in Developing Countries . Benjamin et al. (2012). Analysing the Determinants of Poverty Severity among Rural Farmers in Nigeria:Censored Regression Model Approach.American International Journal of Contemporary Research, Vol. 2 No. 5. Besley Timothy & Robin Burgess. (2003). “Halving Global Poverty” The Journal of Economic Perspectives, 17(3): 3-22.

Bradshaw, T. K. (2000). Complex Community Development Projects: Collaboration, Comprehensive Programs and Community Coalitions in Complex Society . Community Development Journal , 133-145. Cochran. (1977). Sampling techniques. New York: John Wiley& Son. CSA. (2007). Population and Housing Census of Ethiopia. Addis Ababa. Dalton. (1920). The Measurement of the Inequality of Incomes. Economics Journal 30 (1920 , 348-61. Dawit, M. (2015). Measuring vulnerability to poverty an Empirical Evidence from Ethiopian Rural Household Survey. Master’s Thesis . Dereje, F. D. (2013). Assessing Households Vulnerability to Poverty in Rural Oromiya,Wollega University,Ethiopia. Journal of Economics and Sustainable Development, Vol.4, No.8,, 116-117. DGWAO. (2014). Annual Report on Emergency,Food Security and Early Warining. Damot Gale Woreda Agricultural Office. Esubalew, A. (2006). Determinants of Urban Poverty in Debre Markos, Ethiopia. Masters Thesis . FDRE. (2012). Ethiopia’s Progress Towards Eradicating Poverty:An Interim Report on Poverty Analysis Study (2010/11). Addis Ababa: Federal Democratic Republic of Ethiopia. Foster, G,T. (1984). A class of decomposable poverty measures. Econometrica 52 , 761- 766. Francisco H. G. Ferreira, S. C. (2015). A Global Count of the Extreme Poor in 2012:Data Issues, Methodology and Initial Results. Policy Research Working Paper 7432 , 6-7. Getaneh, A. K. (2011). The Impact of Selected Small-Scale Irrigation Schemes on Household Income and the Likelihood of Poverty in The Lake Tana Basin of Ethiopia. Gillis M., Perkins, H., Roemer M., and Snodgrass,R. (1996). Economics of Development. W.W.Norton & Company, Inc. New york. Greene. (1993). Econometric Analysis, Fifth Edition. New York: Pearson Education, Inc., Upper Saddle River,New Jersey, 07458. Gujarati & porter. (2009). Basic Econometrics (Fifth Edition ed.). New York, America: McGraw-Hill/Irwin. Hansen et al. (n.d.). Financial Satisfaction in Old Age: A Satisfaction Paradox or a Result of Accumulated Wealth?

Haughton and Khandker. (2009). Hand Book Poverty and Inequality. Washuington,Dc: World Bank. IFAD. (2011). Rural Poverty Report(2011):Enabling Poor Rural Peopl,Erome, Italy. Rome, Italy: International Fund for Agricultural Development (IFAD). Jencks, C. (1996). Can we replace welfare with work? in m. R. Darby (ed), Reducing Poverty in America (pp. 69-81) Thousand Oaks: Sage. united states of amerca. Maru, S. (2004). Rural Poverty and its Determinants in Zeghe Peninsula, Bahir Dar Zuria Woreda. Masters Thesis , pp 68. MEDaC. (1999b). Poverty Situation in Ethiopia. Welfare Monitoring Unit of The Ministry of Economic Development and Cooperation (MEDaC). Addiss Ababa. Meneyahel, Z. (2015). Farmers Willingness to Pay and Choice of Smallholder Water Lifting Technologies: Evidence from Adami Tulu, Dangela and Lemo Weredas, Ethiopia. Pp 46. Metalign, A. T. (2005). Rural Poverty Situation and Determinants: The Case of Kersa Kondaltity Woreda, South West Shewa,Ma.Thesis Submitted to The School of Graduate Studies, Addis. Addis Ababa. MoFED. (2002b). Develoment and Poverty Profile of Ethiopia. Addis Ababa: Ministry of Finance and Economic Development. MoFED. (2012). Development and poverty. Addis Ababa: Ministry of Finance and Economic Development. MoFED. (2012). Ethiopia’s Progress Towards Eradicating Poverty:An Interim Report on Poverty Analysis Study (2010/11). Addis Ababa: Ministry of Finance and Economic Development. MOFED. (2013). Development and poverty in Ethiopia 1995/96-2010/11. Addis Ababa: Ministry of Finance and Economic Development. Moges, A. G. (2013). The Challenges and Policies of Poverty Reduction in Ethiopia, Special Issue on the Ethiopian Economy Institute of Humanities and Social Sciences, University of Tsukuba, Japan. Ethiopian E-journal for Research and Innovation Foresight , Vol 5, No 1 (2013) ‐pp (94-117. Noursis, j. Marija(1964). SPSS Advanced Statistics 6.1. SPSS Inc, U.S.A. Nsikak et al. (2012). The Determinantes of household poverty in Nigeria. International Journal of Agricultural Management & Development (IJAMAD) , pp-6. Rajaram, R. (2009). Female-Headed Households and Poverty: Evidence from the National.

Ravallion & Bidani. (1994). How Robust is a Poverty Profile? World Bank Economic Review. Volume 8,No. 1. Jakarta, Indonesia. Ravallion, M. (1992). Poverty Comparisons: A Guide to Concepts and Methods. LSMS Working Paper no. 88. World Bank. Ravallion. (1998). “Poverty Lines in Theory and Practice.” LSMS Working Paper No. 133,World Bank, Washington, DC. Ravallion. (2003). ”Assessing the Poverty Impact of an Assigned Program" in Francois Bourguignon and Luiz A.Pereira da Silva (eds.) The Impact of Economic Policies on Poverty and Income Distribution:. Oxford University Press . Razak, N. F. (2007). Determinants of Poverty in Malaysia. Regassa E. Namara, Godswill Makombe, Fitsum Hagos, Seleshi B. Awulachew. (2007). Rural poverty and inequality in Ethiopia: does Access to Small-scale irrigation make a difference? Addis Ababa. Rowntree, B. S. (1941). Poverty and Progress: a Second Social Survey of York. London:. London: Longmans, Green. Sabina Alkire, Mihika Chatterjee, Adriana Conconi, Suman Seth and Ana Vaz . (2014, june). Poverty in Rural and Urban Areas direct comparesion. Sen, A. (1979). “Personal Utilities and Public Judgements: Or What’s Wrong with Welfare Economics?” . The Economic Journal , 537–58. Sisay Asefa et al. (2012). Rural Poverty and Food Insecurity in Ethiopia. Vol 4, No 1 ‐pp (68-98). Spaho, A. (2014). Determinants of Poverty in Albania. Journal of Educational and Social Research , Vol. 4 No.2 p157. T.Bruck. (2001). Determinantes of Rural Poverty in Post-War Mozambique;Evidence from a Household Survey and Swamy,"Social Capital,093/wbro/6.1.57/ Implication for Government and Donor Policy" . Queen Elisabeth House Working Paper Series No. 67,Oxpord University,Oxpord. Ted K.Bradshaw. (2006). Theories of Poverty and Anti-Poverty Programs in Community Development RPRC Working Paper No. 06-05. California, Davis: RUPRI Rural Poverty Research Cente. Tekle Leza & Berhanu Kuma. (2015, January 26). Determinants of Rural Farm Household Food Security in Boloso Sore District of Wolayta zone,SNNP State,

Ethiopia. (A. U. Chrysanthi Charatsari, Ed.) Asian Journal of Agricultural Extension,Economics & Sociology , 61-66. Tesema, A. (2015). Contribution of wotershade Manegment for climate Change.The Case Study of Wolayta Zone.Master Thesis. Addis Ababa,Ethiopia. Townsend P.and D. Gordon (2002).World Poverty, New Policies to defeat an Old Enemy, Bristol, Policy Press. UN. (2014). The Millennium Development Goals Report 2014. New York: United nations. WBG & IMF. (2014/2015). Ending Poverty and Sharing Prosperity:Global Monitoring Report 2014/2015 ,Washingitan DC,USA. Washingten Dc: A joint publication of the World Bank Group and the International Monetary Fund. WFP. (2014). Comprehensive Food Security and Vulnerability Analysis (CFSVA) in Ethiopia. Addis Ababa: World Food Program. Wooldridge. (2001). Introductery Econometrics. World Bank. (2005). World Bank Institute Poverty Analysis Report. World Bank. World Bank. (2015). Ethiopia Poverty Assessement. World Bank. World Bank. (2015). New 2015 Global and Regional Poverty Forecasts. Washingiten: World Bank. Xinshen Diao, Alejandro Nin Pratt. (2006, May 9). Growth Options and Poverty Reduction in Ethiopia:an Economy-Wide Model Analysis. 22-23. Yohannes, K. (1996),Demographic Characteristics of Poor Households in Urban Ethiopia,the Ethiopian Economy: Poverty and Poverty Alleviation. Proceedings of ohe Fifth Annual Conference on The Ethiopian Economy,Ethiopia. Addis Ababa, Ethiopia. Zewdu, A. (2011). Dimensions and Determinants of Poverty Among Rural Households the Case of Itang Special District of Gambella in Ethiopia: Maters Thesis. Addis Ababa.

APPENDICES

APPENDIX-1: Questionnaire for the Study

Dear Respondent:-

My name is Zegeye Paulos; I am a graduate student at Arba Minch University. I am doing research entitled “Determinants of poverty in rural households in Damot Gale Woreda, Wolayta zone” as part of the requirements for the award of Master of Art degree (MA) in development economics.

This interview questionnaire designed for a study whose overall objective is to identify and analyze the magnitude and determinants of rural poverty in Damot Gale woreda. The output of the study is beyond doubt important for the poverty reduction endeavor of the woreda. Therefore, the information you are going to provide will be exclusively used for academic purposes only. In answering the questions, please remember that there are no correct or wrong answers. So I urge you, to answer as fully as possible because your perceptions and opinions are valuable information.

Thank you in Advance!!

Section 1. Household Characteristics 1. Kebele name ------Community/village name ------2. Household head name ------Code------3. Household head Age ------Sex: 1. Male 2. Female 4. Marital status: 1. Married 2. Single (unmarried) 3. Widow 4. Divorced 5. Household size(Family size) ------6. Number of household in the ranges of age A. 0 – 14 years old ------B. 15 – 64 years old------C. 65 years old and above------7. Household head highest educational level 1. Illiterate 4. Junior (7-8) 7. First degree holder 2. Read and write 5. Secondary (9-12) 8. Above first degree holder 3. Primary (1-6) 6. Diploma Holder 8. Have any of your household members frequently suffered from diseases?1=yes 2=No 9. If #8 “yes” Degree of illness 1. Very critical 2. Critical 3. Moderate 4. simple 10. Annual farm income of household ------Birr , 11. Income from Off-farm activities ------Birr 12. Remittance/ Transfer you get from your children/relative monthly/annually------Birr 13. Total monthly/annual income that you and other members of your household in Birr------14. Monthly Expenditure of your household for food items in Birr------15. Monthly/annual Expenditure of your household for non-food items in Birr------16. How do you rate your family monthly income against monthly expenditure? 1. Income covers expenditure 2. Income does not cover expenditure 17. If #16 answer “does not cover”, how do you cover monthly income and expenditure? 1. Sale of assets 2. Support from relative’s 3. By sending children to work 4. by borrowing from lending institutions 18. Do you involved in any community association(Edir or any other) that support you in case of unexpected economic/social shocks your family face? 1= yes 2= No 19. Do you have support/remittance from your relatives/children outside local area 1=yes 2=No

Section 2: Socio-economic characteristics of the Respondent

Was the family member engaged in Relationship to the Marital status Educational If “No” reasons Sex Age productive work If “Yes” type of work? household head status during 12 for not working months?(yes/No) Name of household members 1 = Head 0 = can not 2 = wife 1=Married read and 1 = yes 1 = student 1 = own farm 3 = Son/Daughter 0=unmarried write 0 = No 2 = no land 2 = petty trade

r 4 = Mother/Father 1 = read 3 = too old 3 = sale of fuel wood

a e e l

y 5 = Brother/Sister 4 = disabled 4 = contractual farm a

and write d m e

e 6 = Niece/Nephew 5 = others 5 = domestic work t f For the rest e = l 7 = Uncle/Aunt 6=sale of local 0 p write grade

e 8 = Grand parent beverage o l c

a of formal 9 = Servant 7 = others n m i

education = e 10 = other relative 1 g

A completed

11= non relative

Section 3: Households Asset and Economic Profile 1. Who do you think the main breadwinner of the household?

A. Head B. Wife C. Son/Daughter D. Mother/Father E. Brother/sister F. Uncle/Aunt G. Grandparent H. Niece/Nephew 2. Do you have land of your own for cultivation? 1=yes, 2=N0 3. If your answer to #2 is “yes” how many hectares? ---- land cultivated in hectares- 4. Do you cultivate all your land currently? 1=yes, 2=No 5. If your answer to #4 is” No”, what is the reason?

1=shortage of oxen 3=shortage of agricultural input

2= shortage of labor 4= other specify------

6. If your answer to #4 is “No”, what is your decision on the land?

1=renting out 3= Leaving unploughed

2=share cropping 4= others (specify it) ------

7. Do you cultivate more land than what you have for example by renting from others or via share cropping? 1=yes, 2=No 8. If your answer to #7 is “yes” how many hectares do you cultivate? ------9. Compared to that of last year how is yield of 2007? 1=increased 2=decreased 3= unchanged 10. Do you think that the yield obtained in 2007 is good? 1=yes, 2=No 11. Have you used agricultural inputs in 2007E.C 1=yes, No=2 A. Fertilizer B. Herbicides C. Improved seed D. Insecticides 12. Have you ever used improved tools of agricultural production? Yes=1 No= 2 13. Access to market: ------hours from the nearest surrounding market and ------hours from the woreda’s Capital, Boditte Tuesday market

14. Do you have animals? Yes=1 No=2

If your answer to #14 is “yes” fill the following table

Type of Oxen cows steers Heifers Calves Sheep Goat Horses Mules Donke Chicken livestock ys s

No of animals owned currently

15. Do you involved in off-farm activities? 1=yes 0=No 16. If your answer to #15 is “yes”, what are the activates 1=weaving 7= Making/selling charcoal 13=Carpentry

2= Blacksmithing 8=selling fuel wood 14= “Mesobsifet”

3 = Tannery 9 = Grain trade 15 = Retailing

4 = Basketry 10 = Livestock trading 16 = Employ of local institution

5 = Pottery 11 = selling labor 17 = others ------

6 = Tailoring 12 = selling local beverage

17. Do you have habit of saving? 1=yes 2=No. If “yes” amount you saved------

1= purchasing animals 2= putting money in the house 3= saving in bank accounts

4= storing part of the harvest 5= ‘Equb’ 6= others (specify) ------

18. Do you have access to rural credit? 1=yes 2=No. If “yes” amount borrowed---- 1= formal 2= Non formal high interest rate (‘Arata’) 3 = Non formal with no interest rate 19. If you do not have access to rural credit, what are the reasons?

1 = Absence of micro Finance 2 = saving requirement

3 = Marginalized to get organized in to groups for group collateral

4 = For fear of defaulters in the group 5 = For fear of risk of not paying back

6 = Do not know what to do with the credit

7= Too old to work with and pay back the credit.

Section 4. Household Expenditure and Income I. Household Income How much income does the household generate from the following sources? No Sources How much Income did the household receive(in Birr) Per month if any Birr Last 12 months in Birr 1 Crop production 2 Sale of animals 3 Milk, butter and cheese 4 Egg production 5 Honey production 6 Sale of durables 7 Received from ‘Equb’ 8 Remittance/transfer received 9 Wage of head 10 Wage of spouse 11 Wage of children 12 From non-farm activities 13 Others(Specify)

II. Household Expenditure 1. How much does the household spend for the following food and drink items? No Items Measure Quantity Expense Expense Expense Source ment per Week per month per year 1=own produce ( in Birr) ( in Birr) (in 2=purchase 4=gift Birr)

1 Wheat KG 2 Teff KG 3 Barley KG 4 Maize KG 5 Sorghum KG 6 Beans KG 7 Chick peas KG 8 Cow pea KG 9 Lentils KG 10 Vetch KG 11 Vegetables KG 12 Fruits KG 13 Coffee KG 14 Tea Cup 15 Meat KG 16 Milk/Cheese Litter 17 Chicken Number 18 Egg Number 19 Oil Litter 20 Salt Kg 21 Sugar Kg 22 Honey Kg 23 Pepper Kg 24 Tella Litter 25 Tej Litter 26 Kinito Litter 27 Araqi Litter 28 Others

2. How much does the household spend for the following Non-food items No Items Amount of Expense in Birr

Per month Per year

1 Clothing and footwear 2 Utensils 3 School fee 4 Medical care 5 Transportation 6 Transfer to others 7 Social affairs (for church, Idir, etc) 8 Matches 9 Kerosene 10 Soap 11 Batteries 12 Payment for Equb 13 Repayment of credit(for Agricultural inputs and others)

Thank you very much for your cooperation!!

APPENDIX-2: Poverty Line Calculation Table Consumption basket used to compute food poverty line for Damot Gale woreda in 2016

Food items Per-capita Normative daily requirement market price of the local area Calorie Gm per day Kg per year/AD per.AD *365

Wheat (un-milled and milled) 577.5 159.09 58.06 522.54

Teff (un-milled and milled) 82.5 23.23 8.47895 110.226

Barley (un-milled and milled) 181.5 49.32 18.0018 198.0198

Maize (un-milled and milled) 660.76 174.80 63.802 319.01

Beans 148.5 41.95 15.31175 229.67625

Bread(kita5), and other prepared 31.66 15.89 5.79985 4.26.099325 food

Potatoes and other tubers 392.07 244.58 89.2717 196.39774

Vegetables 36.62 99.75 36.40875 291.27

Fruits 1.27 2.45 0.89425 10.7343

Coffee and tea 22.36 18.76 6.8474 205.422

Meat 7.2 3.65 1.33225 133.225

Milk, Cheese and Egg 15.5 18.06 6.5919 26.3676

Oil and fat 13.63 1.68 0.6132 15.33

Salt, Sugar and others 28.93 16.21 5.91665 73.958125

Total food poverty line 2200 869.42 317.3305 2332.177Birr6

Source: Own construction by following the Method World Bank institute August, 2005 poverty Manual and Development and poverty report MoFED (2013)

5 Local bread which is made up of Maize/wheat powder 6 Minimum food consumption required for adult equivalent individual.

APPENDIX-3; Adult Equivalence Scale

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 14-16 1.06 0.86 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: MOFED (2013) Development and poverty report and AESE, 2006

APPENDIX-4.TLU Conversion Factor

Sr.No Livestock Type TLU 1 Ox/Cow 1 2 Bull 0.8 3 Heifer 0.75 4 Calf 0.2 5 Donkey 0.7 6 Donkey (Young) 0.35 7 Horse/Mule 1.1 8 Camel 1.25 9 Sheep/Goat 0.13 10 Sheep/Goat (Young) 0.06 11 Chicken 0.013 Source: Storck et al (1991)

APPENDIX-5:Model specification Test

Source: Own computed using stata

Source: Own computed using stata.

APPENDIX-9: Association between being Poor and Household Size

Source: Own computed using stata