Determinants of Urban Poverty: The Case of Debre Markos

Esubalew Alehegn[1]

Abstract[2]

The heterogeneity of urban poverty in a country is attributed to the high monetization of economies. Unlike in rural areas, urban poverty is reflected at an individual rather than communal level. Accordingly, poverty in such context is usually described in terms of income, consumption level, and employment. In Ethiopia the challenge of urban poverty has become daunting. However, analytical works that scrutinize urban poverty profile in the country in general and in medium towns in particular are at best scanty.

In light of this the study assessed determinants of urban poverty in Debre Markos town. Data to carry out the study came from surveys conducted in the year 2006. A total of 260 household heads were selected by a systematic random sampling from six Kebeles: Kebele 01, 03, 04, 05, 08, and 12 of the town.

A Logistic regression model was employed and estimated based on the primary data with the probability of a household being poor as a dependent variable and a set of demographic and socioeconomic variables as the explanatory parameters. By making use of Cost of Basic Needs (CBN) approach the study identified respondents as poor and non-poor. Based on this, out of the 260 surveyed household heads, 172(66%) of them were found poor. The study obtained the head count, poverty gap, and severity index 0.66, 0.21 and 0.09 respectively.

The variables that are positively correlated with the probability of being poor are: sex, household size, and health status of the household. Variables negatively correlated with the probability of being poor are: income, educational level, marital status, employment, age, housing tenure, water source, electricity connection and telephone service. Variables which affected significantly incidence of poverty at 99% confidence interval are: average monthly income, family size, educational level, and health status of the households where as sex, age, marital status, water source, housing, telephone service, and electric connection were found statistically insignificant indicators of urban poverty.

That incidence of poverty is rampant among the surveyed households- 0.66 the head count ratio, 0.21 poverty gap, and 0.09 the severity index in the town respectively calls for urgent interventions aimed at curbing the fate of the poor.

1. Introduction

Poverty amidst plenty is a daunting challenge to developing countries in the 21st century. Ethiopia as one of the developing countries in the world is the poorest of the poor by any standard. Poverty in Ethiopia is a longstanding problem. It affects a significant portion of the rural and urban population. Based on estimates of International Poverty Lines, USD 1 per day in the country is 26.3 percent. The percentage will increase to 80.7 if the poverty line is raised to USD 2 per day per person in the year 2007 (World Bank, 2005). Based on the National Poverty Line of the year 1999, 44 percent of the population is poor (MoFED, 2002). World Bank (2000) and UNDP (2003) cited in Yared (2005) reported the country's lowest GNP Per-capita in the world with its Purchasing Power Parity (PPP) adjusted GNP ranking 200^thfrom 206 countries. The HumanDevelopment Index (HDI) and Human Poverty Index (HPI) respectively ranked Ethiopia the 91^st and 6^th out of 175 and 94 developing countries (UNDP, 2003).

Poverty in Ethiopia manifests in a number of ways and this, in fact, is attributed to a multitude of interrelated factors. Abu (2006), for example, has identified potential factors as insufficient source of income, lack of asset, poor health status, poor educational level and backward attitude of people towards work. These factors in one or another way have direct or indirect effects on the life standard of people. Lack of income, for example, will result in reduction of expenditure pattern; poor health leads to being unproductive, absence from work, less energetic; lack of education results in lack of skill, helplessness and so on. Although such factors are common, there are, obviously, some variations among the causes, processes, and consequences of poverty in the urban and rural setup. In the urban setup while life is predominantly a monetized economy and complex that of rural is basically dictated by land assets and number of livestock available to the farmer.

Urban poverty in Ethiopia is pervasive. Using a per-adult equivalent measure, the headcount (P₀), the poverty gap (P₁), and severity (P₂) of consumption poverty in 1994 were 39, 15, and 8 percent. In 1995, this figure became 38,15,8 percent in order. Two years latter (1997), the headcount, poverty gap, and severity figure of urban poverty in the country were 36, 15, and 9(Bigsten et al, 2002) respectively. Recent studies put overall urban poverty around 40,much higher than the previous years. Given this figure, the government of Ethiopia provides unduly emphasis to rural societies with the notion that the rural part is conceived as the hard hit section of the country in terms of poverty. Central to the rapid development policy of the country is the introduction of the Poverty Reduction Strategy Paper (PRSP). This, it is hoped, will bring and guides poverty reduction efforts. One major weakness in the government's PRSP is lack of detail information for implementing and monitoring the strategy. It particularly does not give much attention to urban areas. This probably is due to the adherence of agricultural development led industrialization strategy. The issue as regard to medium towns is even more neglected. This paper should help government realize urban poverty reduction goals by laying the foundation for analytical work aimed at an in-depth understanding of medium towns' poverty and by establishing benchmark conditions for poverty monitoring.

The rest of the paper is organized as follows. Section two presents problem statement and objectives. Section three highlights profile of the study area. Section four introduces methodology. Section five brings literature review. Section six discusses the estimation results and finally conclusion and policy implications are made.

2 Problem

In Ethiopia poverty is the general feature for the nation and causing many sufferings and anguish to the largest proportion of the population. It is high agenda of the government, donor agencies, NGOs and other actors. The government has been formulating and implementing various policy interventions and programs that are in one way or another related to poverty reduction.

Currently, though poverty is taken as the country’s rural phenomena there is a diffusion and growth of urban poverty. Indeed, the number of urban poor is increasing at unprecedented rate. This is due in part to the highest rural-urban exodus and alarming internal population growth (Dessalegn and Aklilu, 2002). In effect, the urban economy has limited capacity to accommodate the populous. In such a situation employment in the formal sector is tough and the probability of getting commendable job opportunities, in fact, could be daunting.

Strategies aimed at poverty reduction, whether in rural or urban, need to identify factors that are strongly associated with poverty and that are amenable to modification by policy (Alemayehu et al, 2005). However, albeit the effects of urban poverty in Ethiopia are getting severe, the factors that account for the results are not studied very well. Debre Markos, which is one of the oldest and medium towns of Ethiopia, faces no exception to this.

This research focuses on one of the least studied areas in poverty reduction strategy-"Determinants of Urban Poverty: The Case of Debre Markos Town" and its relevance to other medium towns urban poverty analysis. Not a few studies have already been done about determinants of urban poverty in old and medium towns of Ethiopia. The few studies done on urban poverty have largely been about large towns. These include the primate -Addis Ababa or secondary towns like Nazreth, Bahir Dar, Mekelle and Awassa(EEA, 2004/05). Most of the available studies are descriptive and focus on measurement issues. Still fewer are the studies specifically at the institutional approach. There are only perhaps a handful of studies discussing contemporary medium urban poverty determinants in old medium towns of Ethiopia. The dearth of studies on the subject of medium urban poverty determinants at household level doesn't much the government's present agenda of poverty reduction strategy, and of course, rural-urban linkage paradigm. The scantiness of materials doesn't fit too the magnitude of problems and issues on urban poverty reduction.

With this in mind, the general objective of this paper is to assess determinants of urban poverty in Debre Markos town. Specific objectives include identifying households who live below the poverty line, examining determinants that play significant roles on urban poverty in the study area, and forward possible research and policy implications.

3 Study Area Profile

Debre Markos is one of the oldest and medium towns of Ethiopia. It is found 300 kilometers Northwest of the capital-Addis Ababa and 265 kilometers Southeast of the Amhara National Regional State city-Bahir Dar. The geographical coordinates of the town are 10^o 21¹ latitude north and 37^o43¹ longitude east. Its total municipal area is approximately 60 kilometers square. Situated at 2420 meters above sea level, the weather condition, most of the time is, 'Woinadega'[3]. The town enjoys a tropical climate with a mean annual rainfall of 1308 mm, temperature 16^oc, while the maximum and minimum recorded temperature being 24^oc and 4^oc respectively (Planning and Economic Development of East Gojjam, 2004).

The 1994 Population and Housing Census of Ethiopia enumerated the population of Debre Markos 49, 297. Disaggregated by sex, 22,652 were males and 26,645 females with the sex ratio of 85. The age structure of residents in the town is similar to most developing countries. Young citizens, particularly, less than fifteen years of age dominate the town. A population projection for Debre Markos, the median variant, puts total population of the town about 117,816 in the year 2006. The sex ratio is very high in the peripheral than the inner urban areas of the town.

Figure 1: Debre Markos Town

The economic activity and social infrastructure of the town is low and the overall life standard of the inhabitants is not in good condition (Planning and Economic Development East Gojjam, 2004). This is due to lack of diversified opportunities such as, absence of commercial crops in the nearby areas, homogenous culture, same language, religion, lack of commerce, and of entrepreneurship.

Residents are engaged in occupations which have limited returns. These include small trade and industry, government employee, and urban agriculture. Most females predominantly engaged in preparing and selling the traditional ‘popular’ drink-Tella. A small number of residents are employed in civil services, while others in trading, small-scale industries (woodwork, metalwork), handicrafts (like weaving, and sewing) and other petty businesses. A large number of households also earn their livelihood by brewing and selling local beverages like Tella, Arekie, and Tej (East Gojjam Administrative Zone, 2001)

Despite paucity of data problems of maladjustments are increasingly felt. There is lack of occupation, affordable education, health, and other psychosocial problems. The rate of unemployment is increasing and the number of job seekers is growing fast (East Gojjam Administrative Zone, 2001). This further is aggravating the existing social problems. The absence of affordable recreational centers in the town is another problem faced by the people. Assaults, thefts, cases of law negligence, and burglaries are some of the common features.

The quality and distribution of education service in the town is still not remarkable (Debre Markos City Service, 2005). Regarding the number, there are nine elementary schools, two secondary schools-first cycles, one preparatory school, eight kindergartens, two basic alternative schools, thirteen basic adult educations, one technical and vocational training school, one college-Debre Markos College of Education, and one National University undergoing construction (Debre Markos City Service, 2005). As to the quality it is poor for the education facilities like students to class ratio, students to teacher ratio, books and availability of qualified professionals are in poor conditions.

In the town, the supply of water usually falls short of demand. The quality is also by no means atrocious. There are eight deep wells, most of which are old, except the one developed this year. These wells together produce twenty-five litters per second, daily on average 2070 cubic meter or 25 liters per second. Surprisingly, 45 percent of the total produced water is wasted, only 20.25 liters consumed in a second, due to different reasons, and one of the most frequently cited is found to be leakage (Debre Markos Water service, 2004). Only 13,200(76%) households have private tap water and the rest-4200 do not have. This shows that there is no adequate water consumption in the town. Households, government and public institutions, and NGOs together consume 55 percent of the produced water, 45 percent being the leakage (Debre Markos Water Service, 2004).

The town gets a 24-hour electricity from the plant of Fincha hydroelectric power with 132-kilo volt line producing capacity (Deber Markos District Electric Power, 2003). During the study period, there were 8259 (47.5%) households, 139 government organizations, 125 private sectors, and 4 small industries, in total 8527 electricity subscribed customers. The rest-52.5 of the households does not have their own electricity. This implies for every household who has own electricity there are about 1.12 households without electricity witnessing the fact that its distribution is the least.

According to Debre Markos Telecommunication Customer Service Office (2004), of the town's total 17,400 households there are only 4,000 subscribers, though the establishment dates back to 1955. This tells for each household with own telephone there are 43.5 who are without. Matching up the town’s telephone density with the household size, it is the lowest by any standard. The services rendered include, among others, fixed telephone, fax, and internet dial-ups. It also has supplementary services such as call waiting, call transfer, teleconference, and internet services.

With regard to transport, (excluding the main road from Addis Ababa to Bahir Dar) only 5 kilometers of the town is asphalted, 42 kilometers inner roads graveled 8.2 kilometers infantry stone road and 95 kilometers dusty /mud roads. During the study period, there were only 22 private taxies (Minibuses and Ladas) serving the customers (East Gojjam Zone Transport Office, unknown year). Contrasting the existing demand of the inhabitants to the existing poor road quality and quantity coupled with the small number of taxies makes one aware to see the problem further.

Although the presences of adequate social services are fundamental to the overall development of urban centers, the number and distribution of social institutions in the town are not in commendable rates. In general, the social services of the town cover only 11.8% of the total allocated area. This shows, on average, about ninety percent of the area is to be covered in the years ahead. Not only is the coverage minimal but also the quality of the distribution. In sum, though 153 years passed since its foundation, the growth and development of the town has remained slow and sometimes started even declining. Accordingly, the incidence of poverty has been increasing due to various reasons (Debre Markos City Service, 2005). Some of them include:

Lack of adequate social services, infrastructures and investments,
Governments' little attention to urban areas,
Lack of good governance,
Lack of adequate financial sources,
Inefficient municipal administration,
Lack of responsibility and accountability, and
Inefficient qualified human power.

All the above problems in one or another way have implications on urban poverty in the town.

4 Methodology and the Data

Data Source

The data of this study came from a cross-sectional survey made in 2006. Information was gathered on household demographic characteristics, employment, asset, income, education, household size, health status, and consumption expenditure. To obtain data on poverty in the town responses were collected through structured questionnaires. The structured questionnaires were posed to household heads. The rationale behind the choice of a household as a unit of analysis is the assumption of sharing resources among households. The numbers of household heads, sample size, were determined by making use of Fowler formulae (2002).

[Za/2]² P [1-P]

D²

Where n= number of surveyed population = sample size

Za/2 = the two-tailed critical value at 95 percent confidence interval (1.96)

P = assumed incidence of urban poverty [4]in Debre Markos (0.22) by taking the 1994 case study of Bahir Dar.

D = Marginal error between the sample and population size (0.05)

The result gives:

(1.96)² 0.22 (1-0.22)

n = (0.05)²

n = 260 households

Therefore, the sample size is 260.

Sampling Technique

The study comprised six kebeles[5] of the town. They are kebele 01, 03, 04, 05, 08, and 12. The total number of households in the surveyed kebeles adds up 9270. Numbers of households selected in each kebele were determined proportion to Kebele population. In the absence of official documents tracing the resemblance and differentiation of the town's socioeconomic status selection criteria of kebeles were made based on two premises: poverty categorization and spatial distribution. In lieu of this, Kebele 01, 04, and 05 were taken as areas of the poor dwellers whereas Kebele 03, 08 and 12 are for those of the relatively well-off residents. Spatially, Kebele 03, 04, and 05 are center whereas Kebele 01, 08 and 12 taken as center-periphery.

Households were randomly selected from each of the 6 keblels based on sampling frames prepared from the housing registry available at the kebele administrative offices. To be more precise systematic random sampling adopted every 25^th of the household. In each of the surveyed kebeles simple random sampling made selection of the first household.

In the survey, the questions were posed to the head of the household and the responses, therefore, represent an individual's evaluation about the poverty of the entire household. Since most household heads were busy in weekdays the survey was conducted in weekends. A possible reservation against the response of the heads of the household is that other members of the household may have different evaluations. This is not likely to be a serious problem in the research since the head is usually the sole or the main breadwinner and his/her evaluation tends to be more authentic. The following table shows the number of households taken from the selected kebeles.

Table 1: Household Size of Surveyed Kebeles

Surveyed Kebele

Total Households

820

1711

1284

1141

2995

1319

Sampled Households

Source: Respective Kebele List and Own Computation

Note N=9270 and n=260.This can be obtained by using the above table. N refers to total households in six Kebeles and n tells proportionately sampled households.

Method of Data Analysis

The presentation and interpretation of the study is quantitative. This is done by making use of Logit model. In this model LIMDEP software was employed to determine the coefficients of the determinants-odds, odds ratio, and marginal effects and test the statistical significance relationships between determinants and the dependent variable-urban poverty. A significance level of 0.05 was adopted to accept or reject the hypothesized assumption. The hypothesized assumption is found in table 3 of the annex.

Model Specification

In order to analyze the correlates of urban poverty a Logistic regression model was employed with the dependant variable being the dichotomous of whether the household is poor (1) or not poor (0). The explanatory variables considered in the analysis are demographic characteristics (sex, age, family size), educational level, occupation, household health, water, and house tenure.

The Logit Model

Logit model is appropriate when we assume the random components of response variables follow binomial distribution & when most variables have categorical responses. Put differently, it is suited when the dependent variable is dichotomous and of the type that have a yes or no response. The form of the Logit model following Gujarati (2006) is:

(1)

Where,

Ý = Probability of a household being poor or non-poor

= Intercept (constant) term

=Coefficients of the predictors estimated using the maximum likelihood method

X_i= Predictors (independent variables)

= Random effect (error term)

Aggregating the value yields

Ý = (2)

In practice Y is unobserved, and is symmetrically distributed with zero mean and has cumulative distribution function (CDF) defined as F (). What we observe is a dummy variable y, a realization of a binomial process defined by

y= (3)

From equation (2) leaving the constant term and rewriting the model yields

Prob(Y=1) = Prob

= Prob

(4)

The Logit model usually takes two forms. It may be expressed in terms of Logit or in terms of event probability. When expressed in Logit form, the model is specified as

Log (5)

Using equation 4 and 5 can be transformed into a specification of the Logit model of event probability by replacing the general CDF, F, with a specific CDF, L representing the Logistic distribution

(6)

The above equation represents the probability of an event occurring. For a non-event, the probability is just 1 minus the event probability.

(7)

5 Literature

Researches in the past indicated variations in the forms and dimensions of poverty in categories such as rural-urban settings (Todaro & Smith, 2003). Poverty literature, often times, in rural areas is marked by its connection with agricultural productivity, land possession and livestock number while urban poverty is associated with heterogeneous economic and social factors (Ravallion 1995,1993,1992), All too often, the poverty of the rural populace does have an impact on urban poverty. In most cases, rural poverty is one of the many factors that stimulate massive exodus among the productive segments of the rural population to cities. In such cases, the poor economic performance of the rural areas will be the major contributing factor to the persistence of urban poverty (Tizeta, 2001).

The heterogeneity of poverty in urban settings is attributed to the high monetization of economies. Unlike in rural areas, urban poverty is defined at an individual rather than communal level. Poverty in such context is usually described in terms of occupation, income, consumption level, and employment status. The above-mentioned aspects, therefore, can serve as bases of urban poverty analysis (Department for International Development, 1997).

World Bank cited in Shewaye (2002) sees urban poverty as a multi-dimensional phenomena characterized by cumulative deprivation where one form of deprivation leads to another. The various dimensions of urban poverty include income, health, education, tenure insecurity, personal insecurity and disempowerment among others. The multi-faceted nature of urban poverty is also noted in Tizita (2001). In her contribution, the various features of poverty that characterize most of the urban poor are: unemployment, lack of wage employment, failure to send children to school, lack of access to health facilities, sanitation, potable water, electric services and good housing. Above all, lack of employment is one of the greatest economic challenges that incapacitate poor people to meet their basic needs.

A study by Christensen (2004) examined the evolution of urban poverty. On the causes of urban poverty, Christensen's findings point to such factors as high urban population growth, rural- urban migration and migration from small to big towns. Rural-urban migration is a coping mechanism devised by the rural poor, but migration adds to the existing burden of urban poverty. Unlike findings elsewhere in sub-Saharan Africa, Christensen's results indicate that the rate of urban poverty is strikingly similar to that of rural poverty in Ethiopia. Although the service sector has shown some growth and is believed to alleviate poverty in Ethiopia, this study did not show that the increased potential for employment has translated into a decline in urban poverty.

Urban areas in Ethiopia are expanding without the necessary preconditions and this is, in fact, paving the way for rampant urban poverty. There is, indeed, ample evidence that urban areas are unable to cope with the increasing population, and delivery of services has deteriorated markedly over the years. Access to housing, health, and education services continues to be seriously limited. Basic sanitary conditions are atrocious by any standard. Transportation facility, energy availability and access to job, labor market, skill reproduction, work, entitlements and finance are also at their lowest level (Dessalegn and Aklilu, 2002).

Urbanization is increasingly pausing a major issue of concern not only in the primate city- Addis Ababa but is leading to daunting challenges among the secondary (regional) cities such as Bahir Dar, Nazareth, Dire Dawa, Awassa, Mekelle, Jimma and presently even to the medium and small towns of the country. Given the high rural - urban migration, fertility rate and natural increase within the urban area, the structure of the population is largely dominated by higher proportions of the lower age group. Of the country's total population, 44 percent were under 15 and 3 percent more than 64 years (CSA, 2003).

This implies that the burden of the dependency ratio for the 53 percent active labor force (aged 15- 64) would be 88 percent. The young population, therefore, dominates the main feature of the Ethiopian urban population with the children (0-14 years) and youth (15-24 years) together accounting for almost 65 present of the total by 2000 (CSA, 2003).

The visible urban poverty signs are everywhere-malnourished citizens with dirty and torn clothes, beggars, shanty homes, scattered garbage, small items exchange sites, idle persons and the like. These poverty symptoms are likely to aggravate with increased urbanization that the country is undergoing (EEA/EEPRI, 2004/05).

Abbi and Andrew (2003) analyzed the status of chronic poverty in urban Ethiopia. They conducted the study in the primate city -Addis Ababa and other secondary cities- Bahir Dar, Nazereth, Dire Dawa, Mekelle, Awassa, Jimma, and Dessie. They conducted their study in three waves of panel data set on 1500 households collected through the Ethiopian Urban Household Surveys from 1994 to 1997. By making use of both descriptive and econometric evidence, their study showed the extent of chronic and transitory poverty in urban Ethiopia, identified the characteristics of the poor, and determinants that explain chronic and transitory poverty. They examined the robustness of the pattern and trends of poverty suggested by the quantitative evidence by linking the subjective evaluation of welfare changes.

They used total household consumption expenditure as proxy for analysis. The rationale behind leaving aside the income approach is due to their sensitivity nature and the less credible responses it would come if used.

Using this, they found out that during 1994-1997, median consumption expenditure per adult declined for the total sample from 100.46 Ethiopian Birr (ETB) to 73.4 Birr. This decline, according to their study, is evident in all regions, is monotonic over the period, and is particularly apparent between 1994 and 1995. Overall, their result suggested deterioration of poverty.

In the second and third waves of their study (1995 &1997) they provided households questions related to changes in household income, expenditure, and living standards since 1994 interview. The three questions asked to households were (a) how has the households’ income changed since 1994 interview? (b) how has households expenditure on basic needs changed since 1994 interview? and (c) to what extent did the living standard of the households change since 1994 interview? This was a perception approach, however, the analysis were of income based instead of quantitative consumption expenditure.

The study confirmed that 40 percent of the case indicated is a significant match to the quantitative evidence that households' income has generally reduced. The study further revealed the congruence between the subjective responses based on income and quantitative approaches through consumption expenditure. Overall, the finding showed an increase in the incidence of urban poverty.

Bigsten et al (2003) reported poverty trends using consumption poverty lines on urban Ethiopia between 1994 and 1997 and found the decrease in the consumption level in the years considered. For all the result showed an increase in poverty from 1995 to 1997. Likewise, in the case of Tadesse (1998), the trends vary by city. Between 1994 and 1995, it declined in Addis Ababa, Awassa, Bahir Dar, and Jimma while it increased in Dessie, Dire Dawa, and Mekelle.

A study by the Federal Democratic Republic of Ethiopia in the period 1995/96 to 1999/2000, based on consumption measure, revealed increased poverty in urban areas more than 11 percent. This study is consistent with the findings of Abbi and Andrew (2003). A study by Ministry of Finance and Economic Development (MoFED,2002) based on 1999/2000 Household Income and Consumption Expenditure (HICE) and Welfare Monitoring Survey(WMS) indicated that incidence of poverty in Ethiopia is higher in rural than urban areas with poverty head count ratio of 45.4 and 37 percent respectively. Though this figure points us to the conclusion that rural poverty is crudely higher than urban, the latter, per see, is significant.

Access to food has also deteriorated in urban areas as measured by real food expenditure per capita and/or adult, which also resulted into decline in calorie consumption per day per adult. The percentage change of food consumption, according to the survey in the urban areas, is negative in all the cases whereas that of the rural areas is positive. This tells that the consumption expenditure of the urban dwellers is in bad scenario.

Tesfaye's (2004) analysis, using panel data collected by Economics Department of Addis Ababa University, has generated different results from the analysis made by MoFED based on the 1999/2000 Household Income Consumption & Expenditures data. Generally, while both analyses confirm poverty increment in urban areas, the level of changes in poverty incidence across different towns made by the two studies is not consistent. The type of methodology adopted and the data analyzed could partly explain this divergence.

Clox (2003) in Ghana found the correlation of household head and spouses’ educational level with a significant positive influence on the likelihood that a household was never poor. The spouses having been educated to primary level or the head to secondary level both had strong negative influences on the likelihood that the household was chronically poor. Djavad (2002) in Yared (2005) found the effect of education for long-term poverty but for short-term poverty, its effect was only significant with high school and above.

Mekonnen (2002) studied determinates and dynamics of urban poverty in Ethiopia by using panel data of households drawn from the Ethiopian urban socio-economic survey conducted by the Economics Department of Addis Ababa University. The study used multivariate regression model to capture factors that determine changes in the standard of living and mobility of households in and out of poverty. He employed total household expenditure per adult equivalent as the dependent variable in the model with the exogenously predetermined household characteristics as the explanatory variables. Grootaert (1997) in Garza (2001) studied determinants of poverty in Cote d'Ivore by using Probit model. He used the data from Cote d'Ivore living standards survey conducted annually from 1985 to 1988 for analysis. He estimated the Probit model for both urban and rural areas separately. Both researchers (Mekonnen and Grootaert) found that the probability of being poor decreases as the age of the household head increases. With regard to the correlates of employment to urban poverty (Abbi and Andrew, 2003, Eyob and Mark, 2004, Mekonnen 2002) found that there is a negative and significant relationship between employment level of the household head and incidence of poverty.

Most empirical evidences suggest that there is a positive correlation between households' size and poverty. For instance, Djavad in Yared 2005 for Iran concludes that households with larger number of members tend to be poor. Likewise, Grootart for Cote d' Ivor, International Food Policy Research Institute (IFPRI) for Malawi, Herrer for Peru, Garza for Mexico, Eyob and Harris for Eritrea, Nigatu, Mekonnen and Ethiopian Economic Association for Ethiopia also reached at similar conclusions-household size is positively correlated with poverty.

Lawson et al (2003) analyzed poverty transitions and persistence in Uganda. The study used the Uganda National Household Survey conducted in 1999/2000.In the study, household movements relative to the poverty line were considered by means of a multivariate nominal Logit model. The study obtained an increase in household size had significant and positive influence on the likelihood that household was chronically poor or fell into poverty.

Garza examined determinants of poverty in Mexico. The data used in the study came from the 1996 National Survey of Income and Expenditure of Households. A Logistic regression was estimated based on the data with the probability of a household being extremely poor as the dependent variable and a set of economic and demographic variables as the explanatory. The study obtained that there is no evidence why female-headed households are more likely to be poor than male-headed households. Using a Logistic regression and the 1992 National Survey of Income and Expenditure, Cortes (1997) revealed that if a woman heads the household the probability of being poor decreases by six percent. These studies are not in conformity with the Ethiopian case. A case in point will suffice to mention the works of (Shewaye, 2002, Mekonnen, 2002) in which female-headed ones are those who experience hard-core poverty.

Michael adopted average odds of participation to analyze how households in different socioeconomic levels shared the benefits from public sectors expenditures on health. The study assumed that access to health service would increase a household welfare in so doing reducing poverty. His findings indicated that households in the bottom quintile have managed to utilize health services relatively more than those in the upper expenditure intervals, which is, in fact contrary to the commonly held assumptions.

All in all, the crucial determinants of poverty among the majority of primate, and big urban areas and nowadays even to medium towns of the third world countries including Ethiopia can be summarized as: low levels of physical and human capital, unequal distribution of productive assets, inadequate access to social services, high fertility especially amongst the poor, and urban development strategies which are biased against labor absorption (Oberia, 1993).

6 Results and Discussion

Descriptive Analysis

Economists and development specialists agree on the perplexities of setting genuine poverty lines. For instance, the minimum calorie intake requirements for households, which are believed to play crucial roles for individual cases, in a specified period, though popularly utilized, are still flawed with debates. This is because households are composed of family members with different age and sex categories leading to differences in needs, consumption habits, and preferences. It is also true that the same level of income cannot serve equally the needs of households that are different in composition.

To minimize such problems scholars have been busy probing for a number of alternatives among which the adult equivalent scale, which establishes on equivalence in the consumption of an adult, a child, and extra, is used eminently. This study has adopted the adult equivalent consumption method in general and the cost of basic needs approaches in particular. The method of calculation for arriving this can be found in the annex part. Following this approach this part discusses major findings of the study.

The survey found the head count, poverty gap and severity indices as 0.66, 0.21, and 0.09 respectively. The incidence of poverty in the surveyed town is, therefore, rampant. When this value is disaggregated the following table depicts the situation.

Table 2: Surveyed Kebeles