




































 
 

 

65 
© 2019 by the authors; licensee Asian Online Journal Publishing Group 
 

Economy 
Vol. 6, No. 2, 65-75, 2019 

ISSN(E) 2313-8181/ ISSN(P) 2518-0118 
DOI: 10.20448/journal.502.2019.62.65.75 

© 2019 by the authors; licensee Asian Online Journal Publishing Group 

    
 

 

 
 
 
Urban Poverty and its Determinants in Debre Markos City: A Household Level 
Analysis 

 
Debeli Chala Biyena1    
Endegena Asmamaw Beyene2    

 

 
( Corresponding Author) 

 
1Ethiopian Biodiversity Institute, Ethiopia. 

 
2Arba-Minch University, Ethiopia. 

 
Abstract 

This study is aimed at examining the extent of urban household poverty using the FGT approach 
and the national poverty line as a reference. The data were collected from a representative of 316 
sample households drawn from each respective kebeles using a combination of simple random 
sampling and systematic sampling techniques. The findings of the study showed that, the 
proportion of the poor people is estimated to be 46.8% indicating that almost half of the people in 
the study area were unable to meet the monthly per adult equivalent consumption expenditure of 
315 ETB. The poverty severity index was also 37.4% in the study area while it was 2.9% and 3.2% 
at national level and in the region. The estimation result of the logit model also showed that, level 
of educational achievements, and household income was negatively and significantly correlated 
with the probability of being poor at 10% and 1% significance level respectively. Whereas, the 
variables that were positively and significantly correlated with the probability of being poor were 
larger family size, unemployment, not owning a house, and household health status/disease at 1%, 
10%, 10%, and 1% level of significance respectively. Based on the result of the study, the following 
recommendation was made. Efforts should be made to raise the real income of households through well-
paying and steady job creation by the setup of micro and small scale enterprises, with the increased 
provision of economic and social infrastructure of houses, education, and better water sanitation services for 
poverty reduction. 

 
Keywords: Urban poverty, Poverty headcount, Poverty gap, Poverty severity, Logit model. 

 
Citation | Debeli Chala Biyena; Endegena Asmamaw Beyene 
(2019). Urban Poverty and its Determinants in Debre Markos City: 
A Household Level Analysis. Economy, 6(2): 65-75. 
History:  
Received: 9 August 2019 
Revised: 12 September 2019 
Accepted: 15 October 2019 
Published: 2 December 2019 
Licensed: This work is licensed under a Creative Commons 

Attribution 3.0 License  
Publisher:  Asian Online Journal Publishing Group 
 

Acknowledgement: The authors would like to acknowledge all the local 
people for actively participating in the study. Government officials and civil 
servants at different levels, who directly or indirectly involved in this study, 
are highly appreciated for their assistance especially during the data collection 
period. 
Funding: This study was sponsored by the Arba Minch University. 
Competing Interests: The authors declare that they have no conflict of 
interests. 
Transparency: The authors confirm that the manuscript is an honest, 
accurate, and transparent account of the study was reported; that no vital 
features of the study have been omitted; and that any discrepancies from the 
study as planned have been explained. 
Ethical: This study follows all ethical practices during writing.   

 

 

Contents 
1. Background of the Study ................................................................................................................................................................ 66 
2. Statement of the Problem .............................................................................................................................................................. 66 
3. Methodology of the Study ............................................................................................................................................................. 67 
4. Result Analaysis and Disccussion ................................................................................................................................................. 68 
5. Conclusions and Policy Implications ........................................................................................................................................... 74 
References .............................................................................................................................................................................................. 74 
 

 

 

 

 

http://crossmark.crossref.org/dialog/?doi=10.20448/journal.502.2019.62.65.75&domain=pdf&date_stamp=2017-01-14
http://creativecommons.org/licenses/by/3.0/
http://creativecommons.org/licenses/by/3.0/
https://www.asianonlinejournals.com/index.php/Economy/article/view/1137
https://orcid.org/0000-0002-5891-9250
https://www.asianonlinejournals.com/index.php/Economy/article/view/1137
https://orcid.org/0000-0002-5891-9250
https://www.asianonlinejournals.com/index.php/Economy/article/view/1137
https://orcid.org/0000-0002-5891-9250


Economy, 2019, 6(2): 65-75 

66 
© 2019 by the authors; licensee Asian Online Journal Publishing Group 

 

 

Contribution of this paper to the literature 
This study contributes to existing literature by examining the extent of urban household poverty using 
the FGT approach and the national poverty line as a reference. 

 
1. Background of the Study 

Poverty is prevalent in large parts of the world and is one of the largest challenges of mankind in the 21st 
century. Despite changes in development paradigms in the last half of the 20th century, the promise to bring well-
being to all human beings remained unfulfilled. One of the major development problems facing the world today is 
growing phenomenon of poverty. It is estimated that over 1.3 billion people live on less than US$1.25 per day, and 
1 billion people cannot meet basic requirements. Furthermore, 415 million people (one in every two people) in sub- 
Saharan Africa survive on less than US$ 1.25 per day and 184 million people (16 percent of African population) 
suffer from malnutrition (Sharma et al., 2016). 

Ethiopia is among the lowest income countries in the world with an average per capita income of merely US$ 
550, a very low Human Development Index (HDI) of 0.435 positioning the country at 173rd out of 187 countries, a 
poverty incidence (Head count Ratio) of 26 percent at US$ 1.25 a day in Purchasing Power Parity (PPP), and urban 
unemployment rate of 16.5 percent (Ministry of Finance and Economic Development (MoFED), 2013; United 
Nations Development Programme, 2014). 

In 2015, Ethiopia was facing drought due to the effects of El Nino, and about 10.2 million persons were in need 
of emergency aid into 2016. According to the United Nations Development Program of 2015, about 23 million 
Ethiopians live in conditions substantially below the basic poverty line and food insecurity remains a major 
challenge. In the same report 44.2 percent of children under five are malnourished and stunted. 

According to the report by Oxford University Poverty and Human Development Index (OPJDI) in 2011, 87 
percent of the Ethiopian population was poor as measured by Multidimensional Poverty Indicators (MPI), which 
means they were deprived in at least one-third of the weighted MPI indicators. This put Ethiopia as the second 
poorest country in the world when using the MPI approach (OPHDI, 2013). 

The multidimensional character of poverty in Ethiopia is reflected in many respects such as, destitution of 
assets, vulnerability, and human development. Poverty in Ethiopia is also associated with certain household 
characteristics. For instance, as compared to richer households, poor households in urban centers tend to have a 
larger proportion of dependents, older household heads, more unemployed family members and more female-
headed households (Aredo, 2005). Moreover, vulnerability to shocks is more serious in urban areas than in rural 
areas. This is partly due to the fact that urban households lack assets such as land and livestock, which are available 
in rural areas. The world is rapidly experiencing urbanization. As urbanization is increasing so also the incidence of 
urban poverty is increasing in depth and intensity in cities of most of developing countries of the world including 
Ethiopia. The dimension of urban poverty that manifests in various ways may also include unemployment, poor 
housing condition characterized in overcrowding, lack of basic services and environmental degradation. In 
Ethiopia, many urban people do not meet their basic needs (Sumner and Meera, 2015). In 2010, 11 percent of 
Ethiopia’s poor lived in cities, but this percent rose to 14 percent in 2010/11. An estimate by MoFED (2013) 
pointed out that 27.8 percent of Ethiopian population was absolutely poor (unable to meet basic needs) of which 
25.7 percent was urban. The number of poor people stayed almost constant between 2004/05 and 2010/11 at 3.2 
million (MoFED, 2013). Even though, the government of Ethiopia has tried to address some problems related to 
poverty, the focus given to urban areas are not gone with the extent of the problem. High population growth due to 
rural to urban migration and other internal factors have making life hard in urban Ethiopia. This really can lead to 
high crime, strikes and other socio-economic and political problems. Therefore, the issues that were addressed in 
this study can help the federal and local governments to design strategies for sound poverty reduction and to tackle 
the real problem of urban areas in general and the study area in particular. 
 

2. Statement of the Problem 
For decades, both rural and urban poverty in Ethiopia has remained pervasive and ever deepening. Despite the 

decline in the level of poverty currently  in Ethiopia, there are about 22.6 million poor people in 2013/14 who are 
living under the poverty line (which is very close to PPP US$1.25 a day on food and non-food items) and who are 
unable to satisfy their basic needs (Sumner and Meera, 2015). Besides, the severity of poverty is also increased 
between 2005 and 2011. Therefore, reducing poverty among the poorest of the poor, insuring food security and 
reducing the number of poor people in the country remain priority agenda and area of intervention in the post- 
Millennium Development Goals period (Sumner and Meera, 2015). 

In the developing world, little is known about urban poverty from quantitative evidence mainly due to lack of 
data tracking from the same households overtime (Kedir and McKay, 2003). Similarly, in Africa the analysis of 
urban poverty dynamics hampered by similar problems and there is diminutive substantiation on such an important 
dimension of poverty.Earlier studies of poverty in Ethiopia mainly focused on rural areas rather than urban areas 
(Bigsten et al., 2003; Dercon and Pramila, 2003). There are few studies that explored the poverty situation of urban 
households in Ethiopia in both a static and dynamic context (Dercon and Taddesse, 1997; Disney et al., 2003). 
However, they concentrated on assessing the poverty condition of primate and large cities of Ethiopia. In line with 
the above points, the urban sector in Ethiopia has been neglected and unnoticed by researchers and policy analysts, 
and also ignored in debates on poverty issues. This has resulted in a’ rural bias.’ 

Even though the effect of urban poverty in Ethiopia is getting severe, the factors that account for the results 
are not studied very well. Most of the studies have been conducted in rural areas of Ethiopia and attempts on urban 
sectors are still scanty. Even the studied ones were confined to the primate city Addis Ababa and secondary cities 
like, Bahir Dar, Nazareth, Awassa, and Mekelle. Poverty status in the Amhara Region remains multifaceted and 
complex. The 2010/11 HICS survey result showed that in terms of food poverty index, the highest incidence was 
observed in Amhara region (42.5%) and the total poverty index was 30.5% in the region. East Gojjam Zone is one 
of the sixteen (16) zones of the region and like other Ethiopian urban areas poverty is a major challenge. According 
to the 2010/11 Ethiopian government poverty assessment report poverty trend in the zone was persistence. 



Economy, 2019, 6(2): 65-75 

67 
© 2019 by the authors; licensee Asian Online Journal Publishing Group 

 

 

There were a number of issues that are considered as a limitation of the previous study and make the current 
study different from the previous one. Among the issues that were considered as a research gap were; one, there is 
only one research attempt regarding poverty in the study area before 12 years ago by Esubalew (2006). Therefore, 
we try to estimate and compare the poverty situation of the study area before 12 years ago and the current ones 
based on different indicators like demographic characteristics, socio-economic indicators, and other households’ 
characteristics, it is quite different. Second, the researcher applied the FEI approach (which shows food poverty 
only) to address the situation of poverty and to identify poor and non – poor households in the study area. 
However, the study was conducted in one of the old medium – sized cities, Debre Markos and since cities have a 
monetized economy. Therefore, as a big limitation in the previous study the researcher does not provide allowance 
for non – food items in measuring the situation of poverty which is the most important part. Therefore, so as to fill 
the gap  in this study  the National poverty Line  set by the government was applied, which is based on the Cost of 
Basic Needs (CBN) approach that provide allowance for non- food items in measuring urban poverty. The last is 
the application of sampling techniques. For instance, the researcher purposely took 6 kebelles from 12 kebelles, and 
in that process he had selected 3 kebelles having high socio-economic status based on different indicators and 3 
other kebelles having low-socio- economic status and such action may result in sampling bias which might have a 
negative impact on the finding of the study. 
 

3. Methodology of the Study 
Debre Markos, the capital city of East Gojjam Administrative Zone is located in the north west of the capital city of 

Ethiopia, Addis Ababa at a distance of 300kms and 265 kms to the capital city of Amhara National Regional State, 
Bahir Dar. Specifically, it is located in the Amhara Regional State, East Gojjam Zone. Until 1995, Debre Markos was 
the capital city of the province of Gojjam and currently the town served as the capital city of East Gojjam Zone.  The 
city is named as Debre Markos after its principal church, which was constructed in 1869 E.C and is devoted to Saint 
Mark. Debre Markos is one of the oldest and medium sized cities of Ethiopia and currently, the city has seven (7) 
kebele administrations (Debre, 2016).  It has latitude and longitude of 100 20’ N, 370 43’ E and altitude of 2,446 meters 
above sea level and it has moderate temperature (Debre, 2016). The area of Debre Markos city is 6,160 ha and has oval 
shape; its average annual temperature is 18.5 0C; Mean annual rain fall is 1,380mm and the existing wind direction is 
from North to South. The city is administration and commercial center. There are 61 manufacturing industries, 37 
wholesale trades, 2101 retail trades, 168 service trades and 6 fuel stations and 6 garages in the town. There are four 
government and three private banks and one micro- finance giving services in the town. The major 
investment opportunities in the town are: agro-industry processing, constriction industry, hotel and tourism & 
social services. According to CSA, 2007, economic activity rate was 44.2 for sexes, 48.3 and 40.4 for males and females 
respectively. The rate of unemployment in the same period was 11.8% for both sexes. The male and female 
unemployment rate was 9.1 and 14.1 respectively. Debre Markos town has economic linkages with the surrounding 
areas and Addis Ababa. The town gets grain products, livestock supply, natural resources (fuel wood and charcoal), and 
labor from surrounding areas manufactured and commercial products from Addis Ababa.   

According to CSA (Central Statistical Agency) (2007), the population of the city was 62,497. Out of this 
29,921(47.87%) were males and 32,576(52.1%) were females; 16,325(26.14%) were within the age group of 0-15 years, 
42,185(67.49 %) were 16-60 years, and 3,987(6.37 %) were 61 years and above. The population growth rate at low 
variant was 2.4% while household family size in the city is calculated to be 3.2. According to CSA (2013), the 
population projection of the city had been estimated 38,291 male and 41,689 female inhabitants which is a total of 
79,980 populations. Area of the city is expected to be 1214.9 Sq.km and 65.85 km/sq. density. 

The primary and secondary data sources were used to carry out the study. The sources of the primary data are 
cross-sectional data collected from the sample that intends to represent the population. On the other hand, secondary 
data were collected from previous working literatures, the findings stated in published and unpublished documents and 
literatures related to the research problem. In this study, household survey was the main tool used to gather the 
necessary data from the sample households. To undertake the survey, structured questionnaire was prepared as 
instrument and an interview was held to gather information at a household level. The prepared questionnaire was 
prepared in English and then translated into Amharic. The Amharic version questionnaire was also pre-tested on 20 
randomly selected households in similar communities. This was done purposely for clarity, acceptability, flow and 
reduction of repetition. Finally, the survey was conducted by four selected interviewers and the researcher supervised 
these four interviewers. A multi-stage sampling technique was employed to get the required primary data. At the first 
stage, based on the current kebele administration structure, Debre Markos city has 7(seven) kebeles, therefore, in this 
study 4 kebeles were selected randomly. Secondly, the 316 households were drawn using systematic random sampling 
proportionate to household head techniques. The household head list in each representative kebele was used as 
sampling frame to select sample households. Finally, systematic sampling method was employed to select the first 
households from each of the kebeles. To make it precise, every 38th of the household were selected from the registration 
of each kebele based on their house number. 
 

Table-1. Number of sample households taken from sampled kebeles. 

No Name of kebeles Total No. of  households in each kebele Sample households in each kebele 

1 Kebele  01 1222 32 
2 Kebele 02 4529 118 
3 Kebele 03 3916 102 
4 Kebele 04 2463 64 

Grand Total 12,130 316 

 
In this study, logistic regression model was employed. The dependent variable takes a value of 1 if the 

household is being poor with the probability of Pi, otherwise takes a value of 0, if the household is non poor with 
the probability of 1-Pi. Therefore, Specification of the logit regression model can be shown as follows: 

     ⌊
  

    
⌋                                                                    (1) 



Economy, 2019, 6(2): 65-75 

68 
© 2019 by the authors; licensee Asian Online Journal Publishing Group 

 

 

Logit (Pi) scale ranges from negative infinity to positive infinity and is symmetrical around the logit of 0.5 

(which is zero). The formula below shows the relationship between the usual regression equation (        + 

…+      ), which is a straight line formula in the logistic regression equation. The form of the logistic regression 
equation is: 

     ⌊  ⌋     ⌊
  

    
⌋                                                 (2) 

   Where, Pi = 1 is the Probability that the household is being poor, 0 otherwise.  
             1-Pi = the probability that the household is not poor. 
              Xi = explanatory variables. 

The probability of one becoming poor or non-poor based on the explanatory variables was calculated using the 
formula given below, which is simply another rearrangement of the formula given above.   
 

   
                         

                           
        (3) 

 
In this study, the explanatory variables (Xi) were selected based on available related literatures on the subject 

at issue. The variables include: age of the household head, sex, marital status, family size, education, main 
occupation, unemployment, income, home ownership, dependency ratio, health status, number of working hours, 
seasonality of work, and access to basic services. Therefore, the model for household poverty status can be 
represented by: 
POVERTY =  

f(AGE,SEX,MARTS,FAMSIZ,EDUC,UNEMP,DEPR,OCUP,INCOME,HOMOWN,HHHS,NWKRS,SESOWK,ABS, )                (4) 

Where,   =  is the random variable LogitPi=β0+β1age+β2sex+β3marst+β4famsiz+β5educ+β6unemp+β7depenr+β8occup 

+β9income+β10homown+β11hhhhs+ Β12nwkhrs+β13sesowk+β14abs+                                     (5) 
 

Table-2. Description of the explanatory variables used in the logistic regression model. 

Variables Description Measurement Expected sign 

Age Household head age Number +/- 
Gender Household head sex Male=1,0 otherwise + 
Marital status Household head maritalst Married=1, 0 otherwise + 

Family size Number of people in household Number + 
Occupation Household head main occpn Gov’t employee =1, 

otherwise=0 
+/- 

Education Educational status Illiterate=1,otherwise=0 - 
Income Level of income Birr - 

Unemployment Unemployed members Number + 
Dependency Family dependents Number + 

Household health Health status Sick=1,otherwise=0 +/- 
Number of working 
hours 

Working hours/day Number - 

Seasonality of work Season impact Yes=1, No=0 +/- 
Home ownership Housing status Own house=1, 

otherwise=0 
+/- 

Access to basic services Credit access yes=1,  No=0 - 

 

4. Result Analysis and Discussion 
In this section, the socioeconomic and demographic characteristics of the data obtained in household survey 

were analyzed using descriptive and econometric methods of data analysis. The whole description takes National 
poverty line as a reference to identify the poor from the non-poor households. 
 
4.1. Poverty Indices 

To measure the extent of poverty and poverty situation of households, Headcount index (P0), Poverty gap 
index (P1), and Poverty severity index (P2) are the most important and widely used indices. In this study, based on 
the poverty line (315 ETB per adult equivalent per month) and the data collected from sampled households, the 
three (3) poverty indices of the study area were computed as follows: 

Headcount Index (P0):  

 

               P0   = 
 

  
  

   

   
          

Poverty Gap Index (P1)  
 

  
∑ (

    

 
)

 

   

1 

                    =  
 

    
∑ (

              

   
)

   

   
  

                    = 
 

    
∑ (

           

   
)

   

   
 

                = 
 

    
(          ) 

       
               = 0.3438 



Economy, 2019, 6(2): 65-75 

69 
© 2019 by the authors; licensee Asian Online Journal Publishing Group 

 

 

Poverty Severity Index (P2) = 
 

    
∑ (

              

   
)

   

   

2 

                               =  
 

    
(          )2 

                                                    =   37. 3639 
   

Table-3. Poverty indices in the study area. 

Poverty Variables National 
Urban(2010/11) 

Regional 
Urban(2010/11) 

Study Area 
Debre Markos (2017) 

Headcount Index(P0) 0.279(27.9 %) 0.292(29.2 %) 0.468(46.8 %) 
Poverty Gap Index(P1) 0.073(7.3 %) 0.080(8.0 %) 0.344(34.4 %) 

Poverty Severity Index(P2) 0.029(2.9 %) 0.032(3.2 %) 0.374(37.4 %) 

 
As can be seen in Table 3 using real per adult consumption expenditure the levels of total urban poverty 

indices at National level, Regional level and in the study area are provided. As apparent from the table above, the 
proportion of poor people (poverty headcount index) in the study area is estimated to be 46.8 percent. However, the 
proportion of poor people at Regional level stood at 29.2 percent and at National level stood at 27.9 percent. 
Similarly, the Poverty Gap Index in the study area was estimated to be 34.4 percent while it was 8.0 percent in the 
region and 7.3 percent at national level. The Poverty Severity Index in the study area was estimated to be 37.4 
percent while it was 3.2 percent at regional level and 2.9 percent at national level.   
 

4.2. Descriptive Data Analysis Based on Sample Households Characteristics 
4.2.1. Household Family Size and Poverty 

The average family size of the sampled households in the study area was 4.5 which were below the national 
average of 5 persons (CSA, 2010). The incidence of poverty is invariably noted to be higher among those with 
larger family sizes. Large households tend to associate with poverty (Lanjouw and Martin, 1995). There is also a 
widely held view that larger families tend to be poor in developing countries like Ethiopia. As evidenced in Table 4, 
the highest incidence of poverty which is 27.53% is observed in those households having 3-5 family members 
followed by households having 6-8 family members with poverty incidence of 16.77% and the lowest incidence of 
poverty  0.6 % is observed  in those households having 1-2 family members.  
        

Table-4. Poverty incidence by households family size. 

Household family size Household Poverty Status Headcount/ 
P0 Poor households(N=148 Non poor households(N=168)  

Total Frequency percent Frequency Percent 

1-2 2 1.35 35 20.83 37 0.006 
3-5 87 58.78 115 68.45 202 0.275 

6-8 53 53.81 18 10.71 71 0.167 
≥ 9 6 4.05 1 0.59 7 0.019 

Grand Total 148 100 168 100 31 0.468 

 

4.2.2 Level of Educational Achievements and Poverty 
As human capital theory predicts, the best investment of all is the one made in people and therefore, greater 

educational attainment may imply a larger set of employment opportunities thereby increased earning potential 
and improve occupational and geographical mobility of labor. Education can serve as an important tool for escaping 
from poverty. As Table 5 reveals, the incidence of poverty is higher for those households having low level of 
educational achievements like illiterate households and those households able to read and write with 13.9 % and 
11.4% of poverty incidence followed by households having primary and secondary education with poverty incidence 
of 7.3% and 5.4. In the case of those households having Diploma/TVET and Degree the incidence of poverty is 
lower than others and they have equal footings of 4.4%. 
      

Table-5. Poverty incidence by level of educational achievements. 

 
Level of Education 

Household Poverty  Status  
 

Total 

 
Headcount/ 

P0 
Poor 

households(N=148) 
Non-poor 

households(N=168) 

Frequency Percent Frequency Percent 

Illiterate 44 29.73 13 7.74 57 0.139 
Read & write 36 24.32 4 2.38 40 0.114 

Primary 23 15.54 23 13.69 46 0.073 

Secondary 17 11.48 29 17.26 46 0.054 

Diploma/TVET 14 9.46 40 23.81 54 0.044 

Degree& above 14 9.46 59 35.12 73 0.044 

Grand Total 148 100 168 100 316 46.84 

 

4.2.3 Income and Poverty 
In this study the average incomes of households per month in birr was taken as a base to classify households’ 

income as lower, middle and higher. Therefore, the monthly income of households was divided as follows: the lower 
income ranges from 0-4040 birr, the middle income ranges from 4041-8000 birr, and the higher income ranges 
from 8001-1600 birr. As can be presented in Table 6 in the study area about 87.16% of the poor households and 
44.64% of the non-poor households were found in the low income level range of 0-4040 birr, 12.16% of the poor 
and 45.24% of the non-poor were found in the middle income range of 4041-8000 birr, and 0.67% of the poor and 



Economy, 2019, 6(2): 65-75 

70 
© 2019 by the authors; licensee Asian Online Journal Publishing Group 

 

 

10.12% of the non-poor were found in the higher income range of 8001-1600 birr. As apparent from the table, the 
highest incidence of poverty (87.16%) was observed in those households who were in the lower range of 0-4040 
birr.            
  

Table-6. Poverty incidence by level of household income. 

 
Income range 

Households Poverty Status  
 

Total 

 
Headcount/ 

P0 
Poor 

households(N=148) 
Non-poor 

households(N=168) 

Frequency Percent Frequency Percent 

0 – 4040 129 87.16 75 44.64 204 0.408 
4041 – 8000 18 12.16 76 45.24 94 0.057 
8001-16000 1 0.675 17 10.12 18 0.003 
Grand Total 148 100 168 100 316 0.468 

    
Table-7. Monthly income–expenditure gaps. 

Does your monthly 
income cover your 

monthly expenditure? 

Households poverty status  
 

Total 
 

 
Headcount/ 

P0 Poor 
households(N= 148) 

Non- poor 
households(N= 168) 

Frequency percent frequency percent 

Yes 56 37.84 110 65.47 166 0.177 
No 92 62.16 58 34.53 150 0.291 

Grand Total 148 100 168 100 316 0.468 

 
The households’ monthly income-expenditure gaps in the study area showed that from the poor category, 

56(17.72%) of the households can able to cover their monthly consumption expenditure and 92(29.11) of the 
households were unable to cover their monthly consumption expenditure whereas in the non-poor category, 
majority of the households 110(34.81%) of the households can able to cover their monthly consumption 
expenditure and the remaining 58(18.35%) of the households were unable to cover their monthly consumption 
expenditure see Table 7. 
 

Table-8. Way of filling income – expenditure gaps/coping strategy of the poor. 

 
Ways of filling income-expenditure 
gaps/coping strategy of the poor 

Household Poverty Status  
 

Total 
 

Poor 
households(N=148) 

Non-poor 
households(N=168) 

frequency percent frequency percent 

By reducing amount of household consumption 51 55.43 28 49.12 79 

By selling valuable assets 5 5.43 3 5.26 8 
By purchasing low price &low quality  items 36 39.13 26 45.61 62 

Grand Total 92 100 57 100 149 

 
The Table 8  provides, the majority of the households 51(55.43%) said that they fill the  income-expenditure 

gaps by reducing the amount of households monthly consumption, 36(39.13%) fill the gap by purchasing and 
consuming low price and low quality items, and the remaining 5(5.43%) fill the gap by selling their valuable assets 
whereas in the non-poor category, 28(49.12%) of the households fill the gaps by reducing the amount of households 
monthly consumption, 26(45.61%) fill the gap by purchasing and consuming low price and low quality items, and 
the remaining 3(5.26%) fill the gaps by selling their valuable assets. Thus, the outcome of the survey shows that 
94.56% of the poor households fill their income–expenditure gaps by reducing the amount of consumption and by 
purchasing low price and low quantity items as a coping strategy for survival otherwise leading a meager life. 
 

4.2.4 Unemployment and Poverty 
From the total poor households, 58(39.19%) have zero unemployed family members, 66(44.59 %) of the 

households have one unemployed family member in their house indicating that majority of the households have 
zero or one unemployed family member, and 24(16.22 %) of the households have 2 unemployed family members. 
Similarly, from the total non-poor households, 131(77.97 %) have zero unemployed family member, 30(17.86%) 
have one unemployed family members, and the remaining 7(4.17 %) have 2 unemployed family member. As 
apparent from Table 9, the incidence of poverty was highest for those households having 1 unemployed family 
member with poverty incidence of 20.8 %, followed by households with zero unemployed family members and 2 
unemployed family members with incidence of 18.4% and 7.6% respectively. Even if majority of the households 
have zero or one unemployed family members 124(83.78 %) of the households couldn’t escape from poverty and 
they were in the poor category see Table 9. 

 
Table-9. Poverty incidence by number of unemployed household members. 

Unemployed 
family members 

Household  Poverty  Status Total Headcount/ 
P0 Poor households(N=148) Non-poor 

households(N=168) 

Frequency percent Frequency percent 

0 58 39.19 131 77.97 189 0.184 
1 66 44.59 30 17.86 96 0.208 
2 24 16.22 7 4.17 31 0.076 

Grand Total 148 100 168 100 316 0.468 



Economy, 2019, 6(2): 65-75 

71 
© 2019 by the authors; licensee Asian Online Journal Publishing Group 

 

 

4.2.5 Home Ownership and Poverty 
As Table 10 reveals, 94(29.75%) of the sampled households were lived in their own house, 215(68.04%) of the 

households do not have their own house and lived in private/rental houses, and the remaining7 (2.22%) of the 
households were lived in their relative’s houses.  Majority of the sampled households were lived in private/rental 
houses and the incidence of poverty was also highest for these households with poverty incidence level of 35.4%. 
Though, there was no equal proportion in the distribution of households, the lowest incidence level was observed 
for those households who were lived in their relative’s houses and those households who were lived in their own 
house with incidence of 1.9% and 9.5 % respectively. 
   

Table-10. Poverty incidence by households homeownership. 

 
Household Homeownership 

Household  poverty  status 

Poor 
households(N=148) 

Non-poor 
households(N=168) 

 
 

Total 

Headcount 
P0 

frequency percent frequency percent 

Own house 30 20.27 64 38.09 74 0.095 
Private/Rental 112 75.67 103 61.31 215 0.354 
Relative’s house 6 4.05 1 0.59 7 0.019 

Grand total 148 100 168 100 316 0.468 

 

4.2.6 Health and Poverty 
Household health status is one of the determinants for the aggravation or improvement of poverty as many 

literatures proved from theoretical underpinnings. In this study households were asked to say Yes or No whether 
any of their household member frequently suffered from disease or not. As apparent from Table 11 from the poor 
households 107(33.86%) said Yes and the remaining 41(12.97%) said No, implying that there was no any household 
member who frequently suffered from disease. Whereas in the non- poor category, 92(29.11%) of the households 
said Yes and the remaining 76(24.05%) of the households said No, indicating that there was no any household 
member who suffered from disease. The incidence of poverty was also highest for those households whose family 
member frequently suffered from disease with incidence level of 33.8% and the lowest level of incidence was 
observed for those households that any of their family member were not frequently suffered from disease with 
incidence level of 12.9% indicating that households’ health status/disease is the major determinant for the 
aggravation or improvement of the level of poverty in the study area. Moreover, the finding of the study revealed 
that household health/disease has significant impact on the incidence of poverty. 
 

Table-11. Poverty incidence by household health status. 

Household members frequently 
suffered from disease? 

Household poverty status  
 
 

Total 

 
Headcount 

P0 
Poor 

households(N=148) 
Non - poor 

households(N=168) 

frequency Percent frequency percent 

Yes 107 72.3 92 54.76 199 0.338 

No 41 27.7 76 45.24 117 0.129 
Grand Total 148 100 168 100 316 0.468 

 
As regarding the extent of city medical services provision status, the sampled households were asked about 

their evaluations about the quality and reliability of the medical services in the city. As apparent from Table 12 
44(13.92%), 201(63.61%), 68(21.52%), 3(0.95%) of the respondents said that the medical service provision in the 
study area was poor, good, very good, and excellent respectively. As apparent from the table, majority of the 
households’ evaluation in the case of extent of medical services provision disclosed that city medical services 
provision was in good condition. 
 

Table-12.  Status of Debre Markos city health centers service provision. 

Household poverty status 
 

City health centers service provision  
Total Poor Good V.good Excellent 

Poor 
households (N=148) 

22 
(14.86%) 

88 
(59.46%) 

37 
(25%) 

1 
(0.67%) 

148 

Non -Poor households( N=168 22 
(13.09%) 

113 
(67.26%) 

31 
(18.45%) 

2 
(1.19%) 

168 

Grand Total 44 201 117 3 316 

 
As Table 13 reveals, regarding the degree of city health service intervention to solve health related problems 

households were asked about their evaluations. Accordingly, 19(6.01%), 97(30.69%), 168(53.15%), 32(10.12%) of 
the households said no attention given, little attention, some attention, and a lot attention respectively.     

   
Table-13. Health centers intervention/attention to solve health related problems. 

Household poverty status 
 

City health centers attention to solve problems Total 

No attention Little 
attention 

Some 
attention 

A lot 
attention 

Poor households (N=148) 9(6.08%) 54 (36.48%) 77 (52.03%) 8 (5.41%) 148 
Non - poor households(N=168) 10 (5.95%) 43(25.59%) 91(54.17%) 24(14.28%) 168 

Grand Total 19 97 168 32 316 



Economy, 2019, 6(2): 65-75 

72 
© 2019 by the authors; licensee Asian Online Journal Publishing Group 

 

 

 
 
 

4.2.7 Electricity and Poverty 
In this study, to assess household’s access to electricity a question was posed to the households to check 

whether they have their own metered electricity or not. As can present in Table 14 the majority 274(86.71%) of the 
households in the study area have their own metered electricity. Specifically, from the poor category 125(84.46%) 
said yes and the remaining 23(15.54%) of the poor households said that they have no their own metered electricity. 
Whereas, from the non-poor category, 149(488.69%) of the households responded that they have their own 
metered electricity and the remaining 19(11.31%) said that they have no their own metered electricity.  The 
incidence of poverty was also highest for those households who have their own metered electricity with incidence 
level of 39.5% while for those households who have no their own metered electricity they have a 7.2% of poverty 
incidence implying that access to electricity has no any meaningful impact to escape from poverty in the study area.  
  

Table-14. Poverty incidence by access to electricity. 

Household have 
metered electricity? 

Household poverty status  
 
 

Total 

 
Headcount 

P0 
Poor households(N=148) Non - poor households(N=168) 

frequency percent frequency percent 

Yes 125 84.46 149 88.69 274 0.395 
No 23 15.54 19 11.31 42 0.072 

Grand Total 148 100 168 100 316 0.468 

 

4.2.8 Water Supply and Poverty 
From theoretical underpinnings and as many literatures proved, water and sanitation contributes to poverty 

alleviation and to improvements in the standard of living in several ways. In this study to assess household’s access 
to water and their main sources a question was posed to households to check whether they have access or not. As 
Table 15 reveals, majority of the households have their own pipeline in their compounds. From the total survey 
209(66.14%) of the households have their own pipeline, 48(15.19%) of the households used public pipelines as their 
main source of water, 50(15.82%) of the households used private pipelines and the remaining 9(2.85%) used their 
own dug-well as a main source of water for the household. Specifically, 81(25.63%) and 128(40.51%) of the poor and 
non- poor households have their own pipelines in their compounds as a main source of water, 29(9.17%) and 
19(6.01%) of the poor and non-poor households used public pipelines as their main source of water, 35(11.07%) and 
15(4.75%) of the poor and non-poor households used private water sources, and the remaining 3(0.95%) and 
6(1.89%) of the poor and non-poor households used their own dug-well as a main source of water. As apparent from 
the table, the highest incidence of poverty 25.6 % was observed in those households who have their own pipelines 
followed by households who used private water sources and public water as their main sources of water with 
incidence level of 11% and 9% respectively. 
 

Table-15. Households main source of water supply. 

 
Household  main source of water 

supply 

Household poverty status  
 
 

Total 

 
Headcount 

P0 
Poor 

households(N=148) 
Non-poor 

households(N=168) 

Frequency percent frequency percent 

Own pipeline 81 54.73 128 76.19 209 0.256 
Public pipe line 29 19.59 19 11.31 48 0.0917 
Private pipeline 35 23.65 15 8.93 50 0.111 
Own dug-well 3 2.03 6 3.57 9 0.009 
Grand Total 148 100 168 100 316 0.468 

 
Regarding water service provision, to assess the status of city water service provisions in terms of quality and 

reliability households were asked about their evaluations. As can be presented in Table 16 majority of the surveyed 
households 170(53.79%) responded that the provision was in poor condition indicating that water service provision 
in the city was not in line with demand of the society. 108(34.17%) said satisfactory, 36(11.39%) said good and the 
remaining 2(0.63%) of the households said that city water service provision was in a very good position. 
 

Table-16. City water service provision status in terms of quality and reliability. 
Household poverty status 

 
City water  service provision  quality and reliability  

Total V. good Good Satisfactory Poor 

Poor households (N=148) 0 17 
(11.48%) 

50 
(33.78%) 

89 
(60.14%) 

148 

Non - poor households( N=168) 2 
(1.19%) 

19 
(11.31%) 

58 
(34.52%) 

81 
(42.21%) 

168 

Grand Total 2 36 108 170 316 

 
In this study to assess the extent of water supply service problem in Debre Markos city households were asked 

about their evaluations.  
As apparent from Table 17 almost half of the surveyed households 151(47.78%) said that there was a serious 

shortage of water in the city and the supply of water was not equipped with the demand of the society. While 
106(33.54%) said less serious, 58(18.35%) said too serious, and the remaining 1(0.31%) said that there was no 
problem regarding water supply. 
 
 
 

 



Economy, 2019, 6(2): 65-75 

73 
© 2019 by the authors; licensee Asian Online Journal Publishing Group 

 

 

  Table-17. Extent of Debre Markos city water service provision problem . 

Household poverty status 
Extent of city water service provision  problem 

Total 
Too serious Serious Less serious No problem 

Poor households (N=148) 27(18.24%) 66 (44.59%) 54 (36.48%) 1 (0.67%) 148 

Non - poor households( N=168) 31(18.45%) 85(5.59%) 52(30.95%) 0 168 
Grand Total 58 151 106 1 316 

 

 

4.3 Econometric Analysis and Model Evaluation Results 
Heteroscedasticity test: A situation in which the variance of the dependent variable varies across the data. 

Many methods in regression analysis are based on the assumption of homoscedasticity or equal (homo) spread 
(scedasticity), that is, equal variance (Gujarati, 2004). In logit analysis there is no equal variance or homogeneity of 
variance assumptions and the variance of the error terms is not constant. In this analysis, Cook Weisberg test for 
heteroscedasticity (hettest) using fitted values of poverty is carried out in stata software. The result showed that 
the value of Chi-square x2(1) = 1.55, and Prob>value = 0.21. Thus, the dependent variable varies across the data. 
 

Table-18. Logistic regression result. 

Variable Coefficient Odd ratio Z Std. Err dy/dx 

Age -0.359 0.97 -0.92 0.28 -0.0065 
Sex -0.887 0.9236 -0.14 0.49 -0.0184 
Ms -0.39 0.8112 -0.85 0.19 -0.0511 

Educ** -0.01 0.7023 -2.41 0.10 -0.0863 

Occup -0.31 0.9263 -1.01 0.07 -0.0187 
Fs*** 0.00 5.37 5.13 1.75 0.4109 

Income*** -0.00 0.9989 -5.60 0.0001 -0.0002 
Depen -0.53 0.8196 -0.62 0.262 -0.0486 

Unemploym* 0.08 1.947 1.74 0.746 0.1629 
Homown* 0.09 2.243 1.69 1.07 0.1974 

NWK -0.40 0.9156 -0.83 0.097 -0.0215 
Season 0.75 1.183 0.31 0.64 0.0410 
Credit 0.17 1.771 1.35 0.75 0.1397 

HHHS*** -0.00 0.2092 -3.47 0.094 -0.3822 
Water 0.17 1.420 1.36 0.36 0.0858 

Elt 0.25 1.989 1.13 1.20 0.168 
Tel -0.57 0.5214 -0.56 0.610 -0.1591 

Cons -0.75 0.4326 -0.32 1.14  
                           Source: Own survey data, 2017***, **, * significant at 1%, 5%, and 10% respectively. 

 
The variable that are negatively correlated with the probability of being poor are age, sex,  marital status, 

education, occupation, income, dependency, number of working hours, household health status. Positively 
correlated with the probability of being poor were family size, house ownership, unemployment, seasonality of 
work, access to credit, water source, and electricity. In the Table 18 of 17 independent variables, six of the variables 
family size, education, income, unemployment, house ownership and household health status have a significance 
level at 1%, 5%, and 10%. The negative values of explanatory variables indicates that when the unit change in 
independent variable lead to decrease in probability of being poor. 

The better educational achievements diminish the probability of a household being poor and it is found 
statistically significant at 5% significance level. The coefficient of education in the regression results showed that a 
level increase in education (a one year increase on the number of years of schooling) will result in 8.6% decreases in 
the probability of a household becoming poor.  

Family size was found statistically significant at 1% significance level and have positive impact on the 
probability of a household becoming poor. As can be seen from table 33, an addition of one extra member to the 
household will result a 41% probability of a household becoming poor. This is not peculiar only in Debre Markos 
but most developing countries. Babatunde et al. (2008); Apata et al. (2010) they  argued that, poverty increases with 
increasing in family or household size because large family size tends to reduce the per capita income available to 
the household. 

The income of household has statistically significant at 1% significance level and negative impact on the 
probability of a household being poor in the study area. As a one birr increase in per adult equivalent income of the 
household decreases the probability of a household falling in to poverty by 0.025%.  However, its effect was not as 
big as expected since urban poverty is mostly determined by income earning capacity especially cash money. 

As expected, unemployment (having larger unemployed family members) in a household was found statistically 
significant at 10% significance level and has positive impact on the probability of being poor. That means, a 
household having one extra unemployed family member result in 16.3% increase in the probability of becoming 
poor in the study area. 

The regression result showed that home ownership (households who had no a residential house) was found 
statistically significant at 10% significance level and it is a positive correlate to poverty. That means, as a 
household had no his/her own house, the probability of a household becoming poor increases by 19.7%. This is in 
line with the view that, as a household own a house the cost that were to be paid as a rent will be saved and the 
house itself can be used as a productive asset.  

The household health status has found statistically significant at 1% significance level and negative impact on 
the probability of being poor. As apparent from Table 18 households with better health status or those households 
whose family member is not frequently suffered from disease decreases the probability of a household becoming 
poor by 38.2%. 
 



Economy, 2019, 6(2): 65-75 

74 
© 2019 by the authors; licensee Asian Online Journal Publishing Group 

 

 

5. Conclusions and Policy Implications  
 To reduce urban poverty and to promote urban livelihood, a useful place to start is with poverty reduction 

arrangements based on poverty analysis aimed at identifying the contributing factors of urban poverty. 
Based on the empirical findings of the study, the researcher portrays the following policy implications to 
mitigate poverty at a household level. 

 The findings of the study showed that, in the study area the proportion of the poor people is estimated to 
be 46.8% indicating that almost half of the people were unable to meet the monthly per adult consumption 
expenditure of 315 ETB. Moreover, the national HICES survey result shows that the poor urban 
households require only 7.3% at national level and 8% in the region. However, in the study area it was 
found that there is a requirement of 34.4% to escape from the poverty group. The poverty severity index 
was 37.4% in the study area while, it was 2.9% at national level and 3.2% in the region. Therefore, one of 
the policy implications of the present study is that the ongoing national and regional governments’ policy 
intervention should target 37.4% of the poorest of the poor that needs an immediate and a joint policy 
intervention by the government and other stakeholders. 

 Urban poverty alleviation is impossible unless the existing economy should generate opportunities for 
investment thereby job creation and sustainable livelihoods. In the study area households’ evaluation of the 
level of investment and its job creation is almost negative indicating that investment has no significant 
impact on poverty reduction efforts and in the process of improving residents’ life standard. Therefore, 
there is a need for policies and strategies that create an enabling environment to attract new investments 
and promoting the existing ones. 

 Though the cost of housing/rent  is one of the main contributing factors to urban poverty, the study 
showed that in the study area majority of the residents, about 68% have no their own house rather they 
were lived in private/rental houses. Therefore, the prices that were paid as a rent just increases the 
residents cost of living. The policy implication of this study is that, there is a need that governments of 
developing countries like Ethiopia should expand their housing program to support ‘non-conventional’ 
incremental social housing, that is, the production of good quality public housing that includes socially 
controlled rental accommodation that is affordable to those households in the lowest income groups. Given 
that most of the management and administration of urban housing in most developing countries are 
handled by the central governments, there should be devolution of authority in the housing sector to the 
regional and local levels so that the increasing needs of the public can be met in short waiting time.  As a 
general remark, public authorities or concerned bodies need to establish policies and strategies that address 
the problem of housing especially mechanisms(which are accessible by the poor) to address the poor 
through the provision of land for construction purpose, construction of condominium houses at affordable 
prices and other options. 

 Urban poverty reduction efforts should be made at the grass root level through the provision of basic social 
services. Inability to afford essential public services of adequate quality and quantity like water can cause 
unhygienic living conditions and ill health. In the present study the analysis showed that, there is a serious 
shortage of water and the service provision was in poor conditions. Households’ evaluations of educational 
provision in terms of quality and quantity were also negative and the provision is not equipped with the 
needs of the society. Therefore, public authorities, city water service, and other stakeholders should try to 
address these social service problems and devise mechanisms to provide adequate and quality water and 
educational services to the society.  

 The findings of the study showed that, poverty and family size were found significant positive correlates in 
the study area. Therefore, the policy implication of this study on such an issue is that so as to reduce the 
impact of larger family size on poverty, city health service and other stake holders should take a remedial 
action through extension services awareness creation about the impact of larger family size on poverty and 
family planning especially targeting on the poor. 

 Generally, efforts should be made to raise the real income of households through well-paying and steady 
job creation by the setup of micro and small scale enterprises, with the increased provision of economic and 
social infrastructure of houses, education, and better water sanitation services for poverty reduction. 
 

References 
Apata, T.G., O.M. Apata, O.A. Igbalajobi and S.M.O. Awoniyi, 2010. Determinants of rural poverty in Nigeria: Evidence from small holder 

farmers in South-Western, Nigeria. Journal of Science and Technology Education Research, 1(4): 85 – 91. 
Aredo, D., 2005. Migrant remittances, shocks and poverty in urban Ethiopia: An analysis of micro-level panel data. Addis Ababa University, 

Addis Ababa, Ethiopia. 
Babatunde, R., E. Olorunsanya and A. Adejola, 2008. Assessment of rural household poverty: Evidence from South-Western Nigeria. 

American-Eurasian Journal Agriculture and Environmental. Science, 3(6): 900-905. 
Bigsten, A., B. Kebede, A. Shimeles and M. Taddesse, 2003. Growth and poverty reduction in Ethiopia: Evidence from household panel 

surveys. World Development, 31(1): 87-106.Available at: https://doi.org/10.1016/s0305-750x(02)00175-4. 
CSA, 2010. FDRE General Country Data. 
CSA, 2013. Population Projection of Ethiopia for All Regions at Wereda  
CSA (Central Statistical Agency), 2007. Population and housing census 2007. Addis Ababa: Federal Democratic Republic of Ethiopia. 
Debre, M., 2016. Debre Markos Municipality Annual Report (Amharic Version). 
Dercon, S. and K. Pramila, 2003. Poverty in rural Ethiopia 1989–95: Evidence from household panel data in selected villages. In The New 

Poverty Strategies. London: Palgrave Macmillan. pp: 179-202. 
Dercon, S. and M. Taddesse, 1997. A comparison of poverty in rural and urban ethiopia. Mimeo, centre for the study of african economies 

and department and addis ababa university. 
Disney, R., A. Kedir and A. McKay, 2003. Price deflators and food poverty in urban Ethiopia. Mimeo, University of Nottingham. 
Esubalew, 2006. Determinants of urban poverty: The case of debre Marekos. Ethiopian Economics Association Proceeding of the Fifth 

International Conference on the Ethiopian Economy, Addis Ababa, Ethiopia, 2: 37 -60. 
Gujarati, 2004. Basic econometrics. 4th Edn., New York: Mc Graw-Hill, Inc. 
Kedir, A.M. and A. McKay, 2003. Chronic poverty in urban Ethiopia: Panel data evidence. International Planning Studies, 10(1).Available at: 

https://doi.org/10.1080/13563470500159246. 



Economy, 2019, 6(2): 65-75 

75 
© 2019 by the authors; licensee Asian Online Journal Publishing Group 

 

 

Lanjouw, P. and R. Martin, 1995. Poverty and household size. The Economic Journal, 105(433): 1415-1434. 
Ministry of Finance and Economic Development (MoFED), 2013. Development and poverty in Ethiopia 1995/96-2010/11.Ministry of 

Finance and Economic Development, Addis Ababa, Ethiopia. 
OPHDI, O.P.A.H.D.I., 2013. Ethiopia country briefing. Multidimensional poverty index data bank. Ophi, University of Oxford. Available 

from www.ophi.org.uk/multidimensional-poverty-index/mpi-country-briefings/. 
Sharma, P., S. Dwivedi and D. Singh, 2016. Global poverty, hunger, and malnutrition: A situational analysis. In Biofortification of food crops. 

New Delhi: Springer. pp: 19-30. 
Sumner, A. and T. Meera, 2015. Global poverty reduction to 2015 and beyond: What has been the impact of the MDGs and what are the 

options for a post-2015 global framework?. IDS Working Papers 2010.348 (2010): 01-31. 
United Nations Development Programme, 2014. Human development report 2014. New York: Sustaining Human Progress: Reducing 

Vulnerability and Building Resilience. 
 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Asian Online Journal Publishing Group is not responsible or answerable for any loss, damage or liability, etc. caused in relation to/arising out of the use of the content. 
Any queries should be directed to the corresponding author of the article. 

 


