Page 1
DSpace Institution
DSpace Repository http://dspace.org
Soil Science Thesis and Dissertations
2019-01-02
THE IMPACT OF MICROFINANCE
INSTITUTION ON BUILDING THE
RESILIENCE OF FEMALE HEADED
HOUSEHOLDS: IN JIGJIGA CITY
SOMALI REGIONAL STATE, ETHIOPIA
ABDURAHMAN KEDIR ALI
http://hdl.handle.net/123456789/9240
Downloaded from DSpace Repository, DSpace Institution's institutional repository
Page 2
BAHIR DAR UNIVERSITY
INSTITUTE OF DISASTER RISK MANAGEMENT AND FOOD SECURITY
STUDIES
GRADUATE PROGRAM
THE IMPACT OF MICROFINANCE INSTITUTION ON BUILDING THE RESILIENCE OF
FEMALE HEADED HOUSEHOLDS: IN JIGJIGA CITY SOMALI REGIONAL STATE,
ETHIOPIA
M.SC. THESIS RESEARCH
BY
ABDURAHMAN KEDIR ALI
January, 2018
Bahir Dar
Page 3
ii
BAHIR DAR UNIVERSITY
INSTITUTE OF DISASTER RISK MANAGEMENT AND FOOD SECURITY
STUDIES
GRADUATE PROGRAM
THE IMPACT OF MICROFINANCE INSTITUTION ON BUILDING THE RESILIENCE OF
FEMALE HEADED HOUSEHOLDS: IN JIGJIGA CITY SOMALI REGIONAL STATE,
ETHIOPIA
M.SC. THESIS RESEARCH
BY
ABDURAHMAN KEDIR ALI
SUBMITTED IN THE PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE
OF MASTER OF SCIENCE [MSC] IN DISASTER RISK MANAGEMENT AND SUSTAINABLE
DEVELOPMENT
January , 2018
Bahir Dar
Page 4
iii
THESIS APPROVAL SHEET
As member of the Board of Examiners of the Master of Sciences (M.Sc.) thesis open defense
examination, we have read and evaluated this thesis prepared by Mr. AbdurahmanKeldir
Ali entitled “The Impact of Microfinance Institution on Building the Resilience of Female
Headed Households: In Jigjiga City Somali Regional State, Ethiopia” We hereby certify that,
the thesis is accepted for fulfilling the requirements for the award of the degree of Master of
Sciences (M.Sc.) in Disaster Risk Management and Sustainable Development.
Board of Examiners
________________________________ _________________ _____________
Name of External Examiner Signature Date
________________________________ _____________ _______________
Name of Internal Examiner Signature Date
________________________________ _____________ _____________
Name of Chairman Signature Date
Page 5
iv
DECLARATION
This is to certify that this thesis entitled “The Impact of Microfinance Institution on Building
the Resilience of Female Headed Households: In Jigjiga City Somali Regional State,
Ethiopia” submitted in partial fulfillment of the requirements for the award of the degree of
Master of Science in “disaster risk management and sustainable development” to the Institute
of disaster risk management and food security studies, Bahir Dar University by Mr.
AbdurahmanDedir Ali (ID. No. BDU0805594PR) is an authentic work carried out by him under
our guidance. The matter embodied in this project work has not been submitted earlier for award
of any degree or diploma to the best of our knowledge and belief.
Name of the Student
Abdurahman Kedir Ali
Signature & date _____________________
Name of the Advisors
1) Dr. Zemen Ayalew(Major Advisor)
Signature & date_____________________
2) NajibAbdi Hassan (Co-Advisor)
Signature & date_____________________
Page 6
v
ACKNOWLEDGEMENT
It is a great honor for me to work on the assigned topic and I feel glad to accomplish my task.
Along with my sincerity and interest, there are few people, who really helped me to make this
endeavor to be a successful one.
At first, I would like to pass my appreciation, gratitude and thanks to my honorable advisor and
co-advisor, Dr. Zemen Ayalew and Najib Abdi Hasan. Their valuable suggestions and ideas in
every step of my work helped me a lot to prepare this thesis.
I am very much grateful to my friends: Omar Sharif, Abdirahim Garad, Abdirahman Maktal,
Abdikadir Hasan, my uncle Mohamed Sharif and Habib Mustafa and his wife Bishara Mohamed,
who also contributed a lot in accomplishing this piece of work to be a successful one.
I would like to express my sincere appreciation and thanks to the officials, and staff of Somali
microfinance institution, specially, Mr. Kadir Ahmed Abdide puty general manager, Mr. Ahmed
Sayid Abdirahmanlegal service department head, Aydarus Omar bade business development
department head and Nur Hussein Farah, Jigjiga branch loan officer.
Lastly, I want to say that without the commitment and support of those persons, this Study would
never be taken shape. For these reasons, I am truly thankful to those people.
Page 7
vi
DEDICATION
This thesis is dedicated to my parents Maymuna Hasan and Keldir Ali, my wife Nura Abdi, my
uncles Dr. Alawi Sharif and Mohamed Sharif and all my beloved family members and friends;
For their unconditional and unbounded love, patience and strength that helped me to complete
this work.
Page 8
vii
ACRONYMS AND ABBREVIATIONS
AFDB African Development Bank
BARC Bangladesh Agricultural Research Council
CBs Commercial Banks
CSA Central Statistics Agency
FDRE Federal Democratic Republic of Ethiopia
FGD Focus Group Discussion
FHH Female Headed Households
IFAD International Fund for Agricultural Development
ILO International Labour Organization
LZ Livestock Zone
MDIs Micro Deposit Taking Institutions
MFI Microfinance Institution
MIX Microfinance Information Exchange
NGO Nongovernmental Organization
RUFIP Rural Financial Intermediation Program
SACCOS Saving and Credit Cooperative Societies
SMFI Somali Microfinance Institution
SRBOFED Somali Region Bureau of Finance and Economic Development
UBOs Ugandan Bureau of Statistics
UNISD United Nation‟s International Strategy for Disaster Reduction
USAID United States Agency For International Development
Page 9
viii
ABSTRACT
Ethiopia is one of the least developed countries with high incidence of poverty. Poverty
Reduction has been the center of development strategy in the country. Microfinance Program on
the other hand has been accepted as instruments in fighting against poverty throughout the
world. The main objective of this study is to assess the impact of Somali Microfinance Institution
(SMFI) on building the resilience of female headed households in Jigjiga City. To realize this
objective cross-sectional survey was employed to collect data from 150 respondents, in which 75
were microfinance loan users and 75 were non-users. The study has employed both logit model
and propensity score matching method to estimate the determinant factors that affect the
participation of female heads of households in microfinance loan and the impact of
microfinance on income of the female headed households. Quantitative research approach
was used. The study revealed that preference for group lending, age of the household head,
marital status of the household, attitude towards risk taking, experience of the household in loan
use, average household income and adequacy of loan repayment were significant importance to
determine microfinance institution. The propensity score estimation technique revealed that
microfinance loan users were build better resilience than non clients of microfinance. Moreover,
old age household-heads are more responsible to increase access of microfinance to elongate
their family status. In addition educated households were less likely to access microfinance to
enhance the gain from additional income sources rather they may get income from other
sources. Hence, propensity score matching is capable of extracting comparable pair of
treatment-comparison households. Therefore, to improve the microfinance accessibility the
institution has to work in those identified factors. In fact, the Somali Microfinance Institution
plays a significant role for the resilience of female headed households.
Key words: Impact, Microfinance Institution, Resilience, Propensity Score Matching,
Jigjiga
Page 10
ix
TABLE OF CONTENTS
APPROVAL SHEET………………………………………………………………………… III
DECLARATION………………...……………………………………………………..….….IV
ACKNOWLEDGEMENT………………………………………………………….…………. V
DEDICATION……………………………………………………………………..………… VI
ABSTRACT………………….………………………………………………….…………….VII
ACRONYMS AND ABBREVIATION…………………………………………..……………..x
LIST OF TABLES…….…………………………………………………………………..……XI
LIST OF FIGURES... ……………………………………………………………...……..……..XI
LISTOF APPENDICES..............................................................................................................XII
1. INTRODUCTION
1.1. Background of the Study .................................................................................................... 13
1.2. Statement of the Problem ................................................................................................... 16
1.3.1. General Objective ........................................................................................................ 18
1.3.2. Specific Objectives ...................................................................................................... 18
1.4. Research Questions ............................................................................................................ 19
2. LITERATURE REVIEW
2.1. Definitions and Concepts ................................................................................................... 20
2.2. Characteristics of Microfinance ......................................................................................... 21
2.3. The Microfinance Institutions Service to Clients ............................................................... 22
2.4. Definitions and Concepts about the Resilience .................................................................. 23
2.5. Household Disaster Resilience ........................................................................................... 25
2.6. Gender Inclusion for Disaster Resilience ........................................................................... 25
2.6.1. Female Headed Households ........................................................................................ 27
2.7. Microfinance in Ethiopia .................................................................................................... 30
2.7.1. Somali Microfinance Institution .................................................................................. 31
2.7.2. Challenges of Ethiopian Microfinance Institutions ..................................................... 32
2.8. Empirical Evidence of Microfinance Institutions .............................................................. 33
2.9. conceptual framework………………………………………………………...………..35
3.THE RESEARCH CONTEXT AND METHODOLOGY………………………..…………...36
3.1. Description of the Study Area ............................................................................................ 38
Page 11
x
3.2. Methods of Data Collection ............................................................................................... 39
3.3.1. Sample size determination……………………………………………………….………..28
3.3.2 Methods of Data Analyze ................................................................................................. 41
3.3.3. Descriptive statistics ……………………………………………………………….……..30
3.3.4. propensity score matching...................................................................................................30
3.3.4.1. procedures of propensity score estimators……………………………………….……..32
3.3.4.2. Matching estimators……………………………………………………………….……35
3.3.4.3.Testing the matching quality……………………………………………………….……37
3.5.Definition of variables………………………………………………………………….……37
4. RESULT AND DISCUSIONS……………………………………………………………41
4.1. Demographic characteristics of the study………………………………………………….41
4.2. Social characteristics of the study…………………………………………………………41
4.3.Risk factors…………………………………………………………………………………44
4.4.Economic factors…………………………………………………………………………….46
4.5. Outcome variable: monthly income…………………………………………………………47
4.6. Estimation results………………………………………………………………………........48
4.6.1.Propensity score……………………………………………………………………………48
4.6.2. Matching programe and non programe households………………………………….……52
4.6.3.Choice of matching algorithm………………………………………………………..…...54
4.6.4.Testing the balance propensity score and covariates……………………..………………55
4.6.5.Treatment effect on the treated……………………………………………..…………….56
4.6.6.Factors influencing treatment effect on the treated………………………………………57
4.6.7.The sensitivity of the evaluation results……………………………………………..……59
5. CONCLUSION AND RECOMENDATION
5.1.Conclusion……………………………………………………………………..…………..61
5.2.Recomendations…………………………………………………………………………….62
REFERENCES…………………………………………………………………………….……64
Page 12
xi
LIST OF TABLES
Table1: Sample size by kebele…………………………………………………..…….…….42
Table2: Age of sample respondents……………………………………….……………...…55
Table3: marital status of sample households………………………………….………….…55
Table4: Social characteristics of sample households…………………………….………….57
Table5: Attitude towards risk taking of sample households………………………..…..…..58
Table6: Number of years of experience of loan taking………………………………….…..59
Table7: Adequacy of loan repayment period………………………………………….…….60
Table8: Average monthly income of the respondents………………………………………61
Table9: Breusch-Pagan / Cook-Weisberg test for heteroskedasticity………………………62
Table10: multi colinearity test among explanatory variables:.……………………..……..62
Table11: Results of logistic regression model…………………………………………..;…63
Table12: Distribution of sample households by estimated propensity scores……………….66
Table13: Performance of different matching estimators…………………………………….68
Table14: Results of the Balancing tests of Covariates Using Kernel band width0.25Estimator...69
Table15: Average treatment effect on the treated for monthly income……………………..….69
Table16: Variance inflation factor for all explanatory variables……………………….……….70
Table17: Results of the Multiple Linear Regression Model for annual income Variables...……71
Table18: Sensitivity analysis result……………………………………………………………..72
Page 13
xii
LIST OF FIGURES
Figure1: conceptual framework………………………………………….38
Figure2: map of the study Area……………………………………………40
Figure3: kernel density of propensity scores……………………………….65
Figure4: kernel density of propensity scores of program households……..66
Figure5: kernel density of propensity scores of non program households…..67
Page 14
13
CHAPTER ONE
INTRODUCTION
1.1. Background of the Study
Wide-spread poverty, with all the problems that comes with it, is the greatest challenge of our
time. One of the identified constraints facing the poor is lack of access to formal sector funds to
enable them to take advantage of economic opportunities to increase their output, thereby move
out of poverty (Sumner, 2007). The microfinance revolution has changed attitudes towards
helping the poor in many countries and in some has provided substantial flow of finance, often to
very low-income groups or households, who would normally be excluded by conventional
financial institutions (Kurmanalieva et al., 2003).
Microfinance has proven to be an effective and powerful tool for poverty reduction (Morduch
and Haley, 2001).As a result, in recent years, microfinance has been considered as an integral
component of poverty reduction strategy by many governments, international organizations and
donors. Improved access and efficient provision of savings, credit, and insurance facilities in
particular can enable the poor to smooth their consumption, manage their risks better, gradually
build their asset base, develop their micro enterprises, enhance their income earning capacity,
and enjoy an improved quality of life. Like many other development tools, however,
microfinance has insufficiently penetrated the poorer strata of the society. The poorest still form
the vast majority of those without access to primary health care and basic education; similarly,
they are the majority of those without access to microfinance (Irobi, 2008).
Ethiopia has an estimated population of more than 80 million with about 85% of the country‟s
population living in rural area. The country‟s dependence on subsistence agriculture, making up
55% of GDP and 85% of total employment, left the economy to shocks and unable to feed its
citizens (Wiss, 2005). Consequently, widespread poverty has become the country‟s main feature
both in the rural and urban areas (Tsehay&Mengistu, 2002). Poor economic growth, low
technological base, periodic drought and famine, and internal conflicts and displacement have
Page 15
14
continued to exacerbate poverty in the country (Yohannes, 2007).These and other complex
factors have caused slowdown in the pace of economic growth of the country and deterioration
in the living condition of its people.
To minimize the problems, the Ethiopian government implemented policy measures such as
Agricultural Development Led Industrialization (ADLI), Food Security Strategy, Poverty
Reduction Strategy Paper (PRSP) and Plan for Accelerated and Sustainable Development
Program (PASDEP) to increase productivity and reduce poverty (Alemayehu, 2008).
According to World Bank report (2015) the economic growth of Ethiopia brought with positive
trends in poverty reduction, in both urban and rural areas. While 55.3% of Ethiopians lived in
extreme poverty in 2000, by 2011, this figure was reduced to 33.5% as measured by the
international poverty line, of less than $1.90 per day.
The government is currently implementing the second phase of its Growth and Transformation
Plan (GTP II) which will run from 2015/16 to 2019/20, aims to continue improvements in
physical infrastructure through public investment projects and transform the country into a
manufacturing hub. The overarching goal is to turn Ethiopia into a lower-middle-income country
by 2025. Growth targets are comparable to those under the previous plan with annual average
GDP growth of 11%.In line with the manufacturing strategy, the industrial sector is slated to
grow by 20% on average.
The government identified also a number of priority areas of actions as part of the government's
poverty reduction and development programs. One of the priority areas acknowledged is the
provision of support to microfinance institutions. In this regard the government is working hard
to solicit funds from international donors for supporting the microfinance sector; hence, the
International Fund for Agricultural Development (IFAD) and African Development Bank
(AFDB) with the support of Rural Financial Intermediation Program (RUFIP) and the European
Union supported Micro and Small enterprise Development program (Meklitet al., 2005).
Page 16
15
Delivery of microfinance services has also been considered as one of the policy instruments of
the government and non-governmental organizations to induce the adoption of new technologies
and enable poor households increase their productivity and income, and reduce poverty. The
establishments of sustainable and profitable microfinance institutions that reach a large number
of poor households who are not served by the conventional banks, such as commercial and
development banks, due to their institutional and structural problems, have been a prime
component of the new development strategy of Ethiopia, i.e., poverty reduction (Wolday, 2000).
Somali microfinance institution which was registered by the national bank of Ethiopia in the
microfinance proclamation of 626/2009 on January 31st 2011isone of the regional government‟s
instruments for poverty alleviation in the region. The institution is the first MFI in the Somali
region (SMFI, 2016).According to this Annual Achievement Report of the 2015/2016 plan of the
institution: In the 2015/16 loan amount of Birr 119,160,000.00accessed to 7,100 solidarity group
clients, the total SMFI active clients in 2015/2016 were 17,321, out of which 14,106 clients were
female while the remaining 3,215 clients were male. At the moment the institution have 23
branches in the region.
Moreover, women empowerment is a key objective of MF interventions. Women need
empowerment as they are constrained by the norms, beliefs, customs and values through which
societies differentiate between women and men. MFI cannot empower women directly, but, can
help them through training and awareness rising to challenge the existing norms, cultures and
values that place them at a disadvantage in relation to men and to help them have greater control
over resources and their lives (Kabeer, quoted in (Mosedale, 2003). Littlefield (2003) stated that
access to MFI can empower women to become more confident, more assertive, more likely to
take part in family and community decisions and better able to confront gender inequities.
Hulme& Mosley (1996) also made this point when they referred to the naivety of the belief that
every loan made to a woman contributes to the strengthening of the economic and social
positionof women.
Page 17
16
MF projects can reduce the isolation of women as when they come together in groups they have
an opportunity to share information and discuss ideas and develop a bond that was not there
previously.
Taking these considerations into account, this study try to assess the impact of Somali
Microfinance Institution on Building the Resilience of Female Headed Households in Jigjiga
City, Fafan Zone of Somali Regional State of Ethiopia.
1.2. Statement of the Problem
There are more than 1.2 billion people around the world living in extreme poverty or less than
$1.25 a day (World Bank data, 2010). Fighting against poverty has thus become an urgent task
for every nation, particularly developing ones. One cause of poverty observed in developing
countries is the credit constraint imposed on the poor. Coleman (1999) described this situation as
a poverty trap: “The poor often find themselves in a vicious circle: producing at a subsistence
level makes it difficult to accumulate savings or other assets, thus making it difficult either to
invest in productive resources or to gain access to credit in formal capital markets, resulting in
low productivity and continued poverty”. Therefore, providing cheap credit to the poor or
microfinance has been considered as a tool for economic development and poverty reduction.
A great majority of women in the world today, mostly in developing countries, live in poverty.
As it has frequently been asserted, women constituted about 70% of the world's poverty stricken
population (Quisumbing et al., 2001). Many researches on women and development have also
consistently shown that women in general and female-headed households in particular are poor.
Although most poor women can also be found in households headed by a man, the poorest
women are in female-headed households (UNFPA, 2002).
In order to alleviate poverty crisis, providing the poor with loan and using local entrepreneurship
have been suggested as a solution. As a result, governmental and nongovernmental organizations
have started providing the poor with capital since 1970s. Microfinance institutions assist in
building the capacity of the poor and graduating them to sustainable self-employment activities
by providing them financial services like credit, saving and insurance among other things. To
Page 18
17
provide microfinance and other support services, MFIs should be able to sustain themselves for a
long period (Elsabeth and Little, 2003).
MFIs in Ethiopia are established with the objective of alleviating poverty through provision of
financial services to the economically inactive people. Those institutions provide an access of
financial resource to the poor in ways that enable them to identify their own livelihood projects
create source of income and provide self employment and mobilize underutilized local resources
(wolday, 2007). So, the eligible persons who access the microfinance are poor peoples.
Many MFIs and NGOs operate in Ethiopian Somali region for the goals of poverty eradication.
Somali microfinance institution (SMFI) is one of the institutions that have the most outreach in
rural areas of the region. So, it is supposing to cover less in terms of outreach, and number of
clients in the region that provides loan for poor in general and for poor women in particular to
make them the beneficiaries of the services since 2011 (SMI, 2016).
Some study findings informed that microfinance alone is not the best intervention for the poor
(Hossian, 2002; Elsabeth and Little, 2003). Concerning this Simanowit and Walter (2000) argued
that the poor cannot be reached. This indicated that better design in the provision of the service so as
to impact the target group is advisable. In fact, there are no any concrete evidential studies which
indicated whether the Somali Microfinance institution support, especially for female headed
households, to resilience with the usual drought occurrence in the study areas. In this regard this
study tried to examine the impact of Somali Microfinance Institution on Building the Resilience
of Female Headed Households in Jigjiga City, Fafan Zone of Somali Regional State of Ethiopia.
There are many studies about the microfinance in global and national levels, but very limited in
Somali region level and Jigjiga specifically. In national level there are many studies conducted
by different researchers, such as study conducted by Alemayehu (2008) on “The performance of
Micro Finance Institutions in Ethiopia”, other study by Sara (2014) on “Determinants of
microfinance institutions loan portfolios,” the study conducted by Martha (2014) on “The
contribution of microfinance institutions to the livelihood of micro credit beneficiaries,” Aziza
(2013) also conducted a research on “The role of microfinance in poverty reduction. Bereket
Page 19
18
(2008) is also conducted a study about “The impact of microfinance services on the living
conditions of the households with low income earning,” and Meron (2003) conducted study on
“Female headed households and poverty in urban Ethiopia”.
Moreover, some researchers did a study in Somali region. Some of them are study done by
Abdkadir (2016) about “The impact of microfinance institution on performance of micro and
small scale enterprises in Jigjiga city”, and other research done by Mubarak (2014) about “The
Contribution of Micro Finance Institutions towards Poverty Reduction”. That is, most of the
researches in this regard focused on much broad in microfinance institution for poverty
reduction. As far as the researcher concern there is no ground study associated with this topic.
That means, no particular study has been done about the impact of Somali Microfinance
Institution on Building the Resilience of Female Headed households. So, it is difficult to make
insightful conclusions without specific researches for the issue of impact of microfinance
institution on building the resilience of female headed.
1.3. Objectives of the Study
1.3.1. General Objective
The main objective of this study is to assess the impact of Somali Microfinance Institution
(SMFI) on building the resilience of female headed household (FHH) in JigjigaCity.
1.3.2. Specific Objectives
1. To identify factors that affect female headed household participation on Somali
Microfinance Institution (SMFI);
2. To assess the impact of Somali Microfinance Institution (SMFI) on female headed
households‟ income;
3. To examine factors influencing the impact of SMFI on building the resilience of female
headed household.
Page 20
19
1.4. Research Questions
The study tried to answer the following research questions:
1. What factors are affecting female household participation on Somali
Microfinance Institution?
2. What are the impacts of Somali Microfinance Institution (SMFI)on female
headed households‟ income?
3. What are the factors influencing the impact of SMFI on building the resilience
of female headed households?
Page 21
20
CHAPTER TWO
LITERATURE REVIEW
2.1. Definitions and Concepts
The definitions of microfinance institution proposed by some authors and organizations are
seemingly different from one another; however, the essences of the definitions are usually same.
It refers to the provision of financial services primarily saving and credit to the poor and low
income individuals and households those don‟t have access to formal financial markets. While
the word microcredit and microfinance are often used interchangeably. Microcredit was coined
initially to refer to institutions like the Grameen Bank that were focusing on getting loans to the
very poor. The focus was explicitly on poverty reduction and social change, and the key players
were NGOs. The push to “microfinance” came with recognition that households can benefit from
access to financial services more broadly defined, at first the focus was mainly on savings, and
not just credit for microenterprises (Beatriz et al.,2010). Microfinance is a form of financial
development that has primarily focused on alleviating poverty through providing financial
services to the poor. Most people think of microfinance, if at all, as being about micro-credit, i.e.,
lending small amounts of money to the poor. Microfinance is not only this, but it also has a
broader perspective which also includes insurance, transactional services, and importantly,
savings (Khan et al.,2007). Definition of microfinance as provided by Robinson (1998) is a
small-scale financial services for both credits and deposits that are provided to people who farm
or fish or herd; operate small or microenterprises where goods are produced, recycled, repaired,
or traded; provide services; work for wages or commissions; gain income from renting out small
amounts of land, vehicles, draft animals, or machinery and tools; and to other individuals and
local groups in developing countries, in both rural and urban areas.
According to Sara Adugna(2014) the definition given by the Microfinance Information
Exchange(MIX) is more appealing than the rest provided in the above paragraph. The MIX is the
microfinance institutions as a variety of financial services that target low-income clients,
particularly women. Since the clients of microfinance institutions have lower incomes and often
have limited access to other financial services, microfinance products tend to be for smaller
Page 22
21
monetary amounts than traditional financial services. These services include loans, savings,
insurance, and remittances. Micro-loans are given for a variety of purposes, frequently for micro-
enterprise development. Historically, the poor have lacked access to reliable and less expensive
financial services. This has been found to act as important constraints for the poor in taking
advantage of opportunities, smoothing consumption, and protecting themselves against different
types of vulnerabilities (Rutherford, 1999).
The current use of microfinance, however, can be traced back to the 1970s when Mohammad
Yunus established the Grameen project in Bangladesh. Yunus established his first micro lending
program in 1976 with the objective of providing a „hand-up‟ rather than a „hand-out‟ to the
impoverished masses of Bangladesh by pioneering the so-called „solidarity circles‟ methodology;
wherein, joint guarantees by groups of borrowers encouraged very high repayment rates on
microloans as mechanism to sustainably provide hundreds of thousands of microloans to the very
poorest (Yunus&Jolis, 2003).
While the primary goal of most microfinance institutions (MFIs) is improving the economic
status of poorer segments of the population, most service providers aim for a broader impact of
enhanced well-being. Because, households function as social and economic units,
microenterprise programs have a unique opportunity to impact the economic, social, and general
well-being of households.
In sum up, microfinance is typically viewed as an economic development strategy, and it is a
particularly relevant approach in countries where disadvantaged groups tend not to benefit from
involvement in the formal economy.
2.2. Theoretical findings
Microfinance came into being from the appreciation that micro-entrepreneurs and some poorer
clients can be „bankable‟, that is, they can repay, both the principal and interest, on time and also
make savings, provided financial services are tailored to suit their needs. Microfinance, as a
discipline has created financial products and services that together have enabled low-income
Page 23
22
people to become clients of a banking intermediary. The characteristics of microfinance products
include:
Little amounts of loans and savings.
Short- terms loan (usually up to the term of one year).
Payment schedules attribute frequent installments (or frequent deposits).
Installments made up from both principal and interest, which amortized in course of time.
Higher interest rates on credit (higher than commercial bank rates but lower than loan-
shark rates), which reflect the labor-intensive work associated with making small loans
and allowing the microfinance intermediary to become sustainable over time.
Easy entrance to the microfinance intermediary saves the time and money of the client
and permits the intermediary to have a better idea about the clients‟ financial and social
status.
No collateral is required contrary to formal banking practices. Instead of collateral,
microfinance intermediaries use alternative methods, like, the assessments of clients‟
repayment potential by running cash flow analyses, which is based on the stream of cash
flows, generated by the activities for which loans are taken (Murray &Boros, 2002).
2.2.1. Service of Microfinance Institutions
Since the 1970‟s, microfinance has much expanded and now includes a wide range of financial
products and services. Similar speaking, Ledgerwood (1999) have stated that there are four broad
categories of products/ services that may be provided to microfinance clients namely:
I. Financial intermediation or the provision of financial products and services such as
savings, credit, insurance, credit cards and payment services,
II. Social intermediation or the process of building the human and social capital required by
sustainable financial intermediation for the poor,
III. Enterprise development services, non-financial services that assist micro entrepreneurs
include business training, marketing and technology services, skills development and
subsector analysis;
IV. Social services or non-financial services that focus on improving the wellbeing of micro
entrepreneurs include health, nutrition, education and literacy training.
Page 24
23
However, the degree to which MFI provides each of these services depends on whether it takes a
minimalist or integrated approach. Many MFIs provide savings and credit services without
getting involved in related development activities. However, many scholars argues that
integrating financial with non-financial services is usually seen as essential in addressing the
causes of poverty identified in a particular area or by a particular group of people; it is rarely the
case that savings and credit activities alone will reduce poverty (Harper 2003; Johnson and
Rogaly 1997; Ledgerwood 1999).
2.2.2. Resilience
According to the UNDP definition building resilience imply a transformative process of
strengthening the capacity of men, women, communities, institutions, and countries to anticipate,
prevent, recover from and transform in the aftermath of shocks, stresses and change. According
to scientists in the resilience alliance, a resilient system is therefore one capable of anticipating,
adapting and coping with uncertainties and unexpected extreme events without losing its
stability, performance and regenerative ability (Ostromet al., 1998).
In the definition of disaster vulnerability proposed by Cardona (2004) comprises three elements:
physical fragility or exposure, socio-economic fragility or sensitivity, and lack of resilience. It
becomes clear that “resilience is the flip side of vulnerability. Its system or population is not
sensitive to natural hazards, climate variability and change and has the capacity to adapt”
(Cardona, 2003; Thywissen, 2006: 23). More precisely, resilience is the capacity of a system,
community or society potentially exposed to hazards to adapt by resisting or changing in order to
reach and maintain an acceptable level of functioning and structure. This is determined by the
degree to which the social system is capable of organizing itself to increase its capacity for
learning from past disasters for better future protection and to improve risk reduction measures
(Thywissen, 2006: 23). So, microfinance institution plays a significant role to being disastrous
community resilience earlier by improving the community financial capacity.
When the community financial power escalated, even the disaster occur they can cope up or
rehabilitate early. Farther more, the community would have thought beyond the delay food and
Page 25
24
how to make the environment better as well how to protect if as the same riskiest problem
happened.
Resilience is the ability of people, households, communities, countries, and systems to mitigate,
adapt to and recover from natural or manmade shocks, such as drought, land slide, earthquake,
fire, fighting of rebellion groups etc and stresses in a manner that reduces chronic vulnerability
and facilitates inclusive growth (USAID, 2012b). In the study area drought is a common disaster.
Currently, this shock affects most of the households those lived in the study areas. According to
the United Nations' Hyogo Framework for Action 2005-2015, which was adopted by 168
countries in 2005, resilience is a key aspect of improved disaster prevention and response,
mentioned in its five major areas for action. The Framework mentions that "a gender perspective
should be integrated into all disaster risk management policies, plans and decision-making
processes, including those related to risk assessment” (UNISDR, 2005). In fact there, a lot of risk
management procedures are using to improve disasters may occur in a certain place. One of the
methods is financially capacitating the community. In other term, if the income of the
community improves, the resilience capacity will accelerate.
The concept of resilience originated in the field of ecology, but it has been used within a wide
diversity of disciplines from psychology, geography, social science to engineering and systems
science (Klein et al. 2003). According to Timmerman (1998), resilience roughly comprised of
two common features for disaster:
1. The ability to resist and absorb disturbances;
2. The ability to reorganize and recover reasonably quickly (retain the same basic structure
and ways of functioning) (Mayunga, 2009).
Long-lasting concerns from the research community focus on disagreements as to the definition
of resilience, whether resilience is an outcome or a process, what type of resilience is being
addressed (economic, infrastructure, ecological, or community systems), and which policy realm
(counterterrorism, climate change, emergency management, long-term disaster recovery,
environmental restoration) should target (Cutter et al., 2010).
Page 26
25
2.5. Household Disaster Resilience
Household is a group of persons who normally live and eat together or share living arrangement
(UBOS, 2014; Arbonet al., 2013). In this study, a household is understood as a group of people
living together or even far away from each other.
Household disaster resilience is the capacity of a person or people sharing a living arrangement
to:
sustain their household even under stress;
adapt to changes in the physical, social and economic environment;
be self-reliant if external resources are limited or cut off and
Learn from the experience to be more prepared for next time (Arbonet al., 2013).
It is clear that resilience is not a state to be attained, so that, attention can then be paid to other
issues. It is an ongoing process that requires consistent and repeated reinforcement to be at a
suitably high level should disaster strike. It is the individuals‟ or households‟ resources and
preparedness, which is bolstered through their active networks, which work together, especially
in times of need, to assist individuals or households to adapt, learn and recover from emergency
events or disasters. Because, preparedness actions take time to implement and because
emergency events are frequently of sudden onset and unexpected, household resilience building
must be an everyday activity. The resilience of households will depend on a range of relatively
small actions and activities that build resources, preparedness and resilience networks.
2.6. Gender Inclusion for Disaster Resilience
While gender equality matters in its own right, particular gender dimensions related to disaster
and climate risk management need to be addressed. Women are disproportionately at risk to the
effects of natural hazards and climate change. Women typically outnumber men among those
dying from natural disasters, often because of cultural and behavioral restrictions on women's
mobility, and socially ascribed roles and responsibilities. However, this gap in vulnerability is
not inevitable. In Bangladesh, when Cyclone Gorky hit in 1991, women outnumbered men by
14:1 among those dying as a result of cyclone-induced flooding. When Cyclone Sidr hit in 2007,
Page 27
26
the gender gap in mortality rates had shrunk to 5:1 by specifically addressing the cultural reasons
why women were reluctant to use cyclone shelters (World Bank, 2011).
Post-disaster relief and recovery efforts can reinforce or increase existing inequalities. It is
critical to assess and understand the different needs of women, girls, boys and men for recovery,
including the indirect economic impacts women typically suffer from being in the informal
economy.
Aid agencies can integrate practical steps to promote gender equality easily and speedily in the
recovery process. These include issuing deeds for newly constructed houses in both the woman‟s
and man‟s names, including women in housing design as well as construction, and promoting
land rights for women. Other steps include building non-traditional skills through income-
generation projects, distributing relief through women, and funding women‟s groups to monitor
disaster recovery projects.
A recent World Bank study conducted in Bolivia revealed that women have adaptation strategies
that employ a more efficient use of existing resources than male community members (Ashwillet
al., 2011).
The World Bank is committed to engaging women and community leaders as active agents of
resilience building rather than passive recipients of adaptation support, and suggests several key
areas of focus such as:
Post-disaster challenges and opportunities;
Earmarking funds to support grassroots women‟s organizations as DRM/resilience
champions;
Building in country institutional capacity at central and local level to address gender
dimensions and formalize role of women leaders; and
Promoting gender-based participation in stakeholder discussion at all levels on DRM
policies, programs, climate change finance, etc.
Page 28
27
2.6.1. Female Headed Households
According to this study female-headed household, is the household where either no adult males
are present, owing to divorce, separation, migration, non-marriage or widowhood, or where men,
although present, do not contribute to the household income (ILO, 2005)
Chant (2003) states that there are two major categories of female-headed households:
I. The de jure category, which includes single, widowed, divorced or separated women;
II. The de facto category, which comprises of wives of male migrants, or women who play
the dominant role even with the presence of a male partner in their lives.
Buvinic and Gupta (1997) say that the concept of female headship seems problematic,
transitional and not neutral. This is because headship is backed by They argue that the other
factors that make one a household head include one‟s economic status.
There are many causes of female headship of households; Mullings (1995) says that wars,
migration and increasing unemployment highly accelerated the phenomenon of FHHs and
women raising children by themselves. FHHs were evidenced in both industrialized and
developing countries such as Iraq and South Africa. The scholar further states that while female
headship of households is a global phenomenon, different groups of people from different parts
of the world and / or with different ethnic backgrounds has different experiences in relation to
FHHs among them than others.
Likewise, a study by Lokshin, et al. (2000) found that the growing incidence of single mothers in
Russia was mostly as a result of the high rate of divorces in the country.
Chant (2007) found that domestic violence is one of the factors causing FHHs in Countries such
as Costa Rica. In order for women to protect them and their children from abusive men, they turn
to single motherhood and run their households.
Page 29
28
Ruwanpura (2003) asserts that FHHs are a result of conflicts that cause death of husbands.
Extramarital affairs by men also make wives leave husbands and end up as heads of households.
And the girls born in FHHs are more predisposed to heading their own households as adults.
According to Chant (2003), FHHs are assumed to be the poorest households. Women have been
marginalized and their access to resources such as land is limited. Their low levels of
employment and heavy work burden with low salaries also contribute to why they may be
assumed as poor. Single mothers and women in FHHs in some cultures have actually been
termed as the “new poverty paradigm.” He further argues that unlike developed countries, some
developing countries have not yet established schemes that can help support FHHs such as
giving benefits from the state. With such a lack of support, FHHs are challenged.
Buvinic and Gupta (1997) say that FHHs seem poor and challenged in their livelihoods because
they have low incomes with many dependents. This makes the FHHs vulnerable and targets for
anti-poverty schemes. FHHs are faced with the burden of domestic work and discrimination in
the employment sector due to their low levels of education, which may lead to the existence of
poverty among their children and future generations.
2.6.2. Women and Microfinance
A woman‟s role in the economy is an important determinant of her ability to provide health care
services, education and safe housing for herself and her family.
It also has an impact on her decision-making power, as well as her ability to speak and act
against inequalities, injustice, and violence in her home as well as in the community (Mayoux,
2002).
The ownership of working capital is a means to building a woman‟s confidence, self-respect,
and the capacity to use her voice to shape her life and the lives of her family members (U.N.,
1995).
Microfinance has over the years been seen to prove successful in targeting women when it comes
to providing working capital for them. The premises behind such targeting are twofold: that
microfinance is an effective tool in improving women‟s status, and that overall household
welfare is likely to be higher when microfinance is provided to women rather than men.
Page 30
29
Women‟s status, household welfare, and microfinance interact in the following ways: A
woman‟s status in a household is linked to how well she can enforce command over available
resources. Increased ability to tap financial resources independently enhances her control, and,
therefore, her influence in household decision-making processes (Cheston&Khun, 2001).
Microenterprises financed by MFIs open up an important social platform for women to interact
with markets and other social institutions outside the household, enabling them to gain useful
knowledge and social capital. Many microfinance programs organize women into groups, not
just to reduce transactions costs in credit delivery, but also to assist women in building and
making effective use of these opportunities. cheston and Kuhn (2002)
Women‟s preferences regarding household business management and household consumption
goals differ from men‟s, particularly in societies with severe gender bias. In such situations,
placing additional resources in the hands of women is not a mere equalizer. It also materially
affects both the quality of investments financed by the microfinance programs and how extra
income is spent. International Food Policy Research Institute (IFPRI, 1997) study on women and
their control of resources have underlined the importance of women‟s control of resources in
achieving better welfare outcomes in food, nutrition, education, and other health statuses of
children and their families. Women are seen as better borrowers than men, timely repayment of
loans is more likely to take place when women borrow.
IFPRI (1997) showed that in Bangladeshi, groups with a higher proportion of women had
significantly better repayment rates. Loans are not simple handouts. If microfinance programs
are designed to cover all costs, a potential win-win situation emerges. Development goals related
to women‟s empowerment and improved household welfare are self-financing and no subsidies
are required.
Unfortunately, positive empowerment effects cannot be unconditionally guaranteed. In some
male-dominated societies, men may use the agency of the woman to gain access to microfinance
funds, diminishing women‟s role to being mere conduits of cash. Even if women can maintain
autonomy in how they access and use microfinance services, their management of newly
Page 31
30
financed enterprises and shouldering of all attendant risks may alter inter household dynamics.
Since loans have to be repaid even if the project fails, new activities may increase exposure to
financial risks and may impose additional pressures on the already overburdened woman.
Finally, in societies following the practice of female seclusion, the new pressures to interact in
the marketplace may initially involve a difficult learning period and trigger negative responses.
Project failures may lead to serious reprimand and additional negative sanctions against the
woman, especially if household resources have to be diverted to repay outstanding debt.
2.7. Microfinance in Ethiopia
The development of microfinance institutions in Ethiopia is a recent phenomenon. The
proclamation, which provides for the establishment of microfinance institutions, was issued in
July 1996. Since then, various microfinance institutions have legally been registered and started
delivering microfinance services (WoldayAmha, 2000). In particular, the Licensing and
Supervision of Microfinance Institution Proclamation of the government encouraged the spread
of Microfinance Institutions (MFIs) in both rural and urban areas as it authorized them among
other things, to legally accept deposits from the general public, hence, diversify sources of funds
to draw and accept drafts, and to manage funds for the micro financing business
(GetanehGobezie, 2005).
The legal foundation for the microfinance industry was laid in the country with the issuance of
Proclamation No.40/1996 on licensing and supervision of micro financing institutions in 1996.
MFIs established in accordance with the proclamation can provide a loan amount of not more
than five thousand Birron the basis of group guarantee and to borrowers who have joined a
membership arrangement as well as lend limited scale to non-members on the basis of physical
or other collateral (NBE, 2002). The major objectives of microfinance institutions in Ethiopia are
the users as policy instruments to enable rural and urban poor to increase the output and
productivity, induce technology adoption, improve input supply, increase income, reduce
poverty, and attain food security (WoldayAmha, 2001).
In recent years, the state and regional governments have made a major push to increase financial
services for agriculture, micro and small enterprises and low-income households (IFAD, 2009).
Page 32
31
The Ethiopian microfinance sector is characterized by its rapid growth, an aggressive drive to
achieve scale, a broad geographic coverage, a dominance of government backed MFIs, an
emphasis on rural households, the promotion of both credit and savings products, a strong focus
on sustainability and by the fact that the sector is Ethiopian owned and driven. The industry has a
strong focus on loans to the very poor, as indicated by the relatively small loans when compared
to neighboring countries. Sector outreach is impressive and the financial performance of the
sector is considered good, although the operational margins and profitability are low. MFIs have
also mobilized a significant amount of savings, thereby improving financial as well as
operational sustainability (MFT, 2011).
The MFIs are motivated to extend the frontier of financial intermediation to those traditionally
excluded from conventional financial markets, the poor. Previous studies on the efficiency of
financial intermediaries consider MFIs and CBs as completely different sectors. The two
industries are conventionally left separate in efficiency analysis of financial firms; even though,
MFIs are motivated merely to extend financial services to those who were not able to access the
conventional banking services (Hundanol&Berhanuwassie, 2012). The Ethiopian government
identified a number of priority areas of actions as part of the government's poverty reduction and
development programs. One of the priority areas acknowledged is the provision of support to
microfinance institutions. In this regard the government is working hard to solicit funds from
international donors for supporting the microfinance sector, hence, the IFAD and AFDB
supported Rural Financial Intermediation Program (RUFIP) and the European Union supported
Micro and Small enterprise Development program (EbisaDeribieet al., 2013).
2.7.1. Somali Microfinance Institution (SMFI)
SMFI is formed by Somali regional state and five private business persons and it is registered by
the National Bank of Ethiopia according to proclamation No. 626/2009. This operational plan is
intended to provide operational policies and procedures related to sustainable microfinance
service provider in Ethiopia particularly in Somali regional state. This proclamation addressed
the financial service needs of the urban and rural low income people. SMFI believes in a credit
plus or integrated approaches as financial interventions alone cannot bring all round and
sustainable changes on the lives of clients. The prevailing poor infrastructure, illiteracy,
Page 33
32
backward technologies at the disposal of potential clients etc. necessitates an integrated approach
to meet the various needs of the clients SMFI annual report (2016).
The interventions or services of Somali Microfinance can be classified as financial services like
provision of credit, financing and investment, mobilization of savings, provision of insurance
services, pension management, fund management, local money transfer, mobile and agent
banking serves etc. and non financial services such as business development management
services, counseling, marketing and related support service ,referral services to appropriate
organizations for accessing working premises, market places etc. SMFI annual report(2016)
The needs of clients depend on the actual situation and prevailing problems faced in the area
clients operate their businesses. Issues of health, appropriate technology to improve productivity,
family planning, skill trainings, numeracy and literacy programs etc are some of the areas that
clients need support of partner organizations specialized in each of these fields (SMI, 2010).
Especially, the Somali MFI has now 23 branches with in different Woredas and One sub
branches these branches divided into the nine zones of the Ethiopia Somali regional state
(SRBOFED, 2015).
2.7.2. Challenges of Ethiopian Microfinance Institutions
One of the key bottlenecks in Africa is the shortage of strong institutions and managers. Skills
and systems need to be built at all levels:
I. Microfinance institutions
II. External services,
III. Central banks and other government agencies.
In spite of the rapid strides and strength of microfinance institutions in Ethiopia, there are several
challenges that are threats for the development of the industry. The main challenges could be:
Shortage of loan capital,
Policy, legal and regulatory constraints,
High risk,
Inflexible financial products,
Page 34
33
Limited capacity, and
External factors (Wolday amha,2008)
2.8. Empirical Evidence of Microfinance Institutions
Importance of microfinance has been assessed at different levels throughout the world. Usually
an impact assessment is made at the household, individual and enterprise level. But, sometimes it
is also assessed at a community level. In the global arena there is already the impression that
microfinance is successful in reducing poverty. Many policy makers are therefore engaged on
how to make microfinance sustainable and available to many poor households in the future.
Many Stake holders in the microfinance industry especially donors and investors argue that,
microfinance can pay for itself, and must do so if it is to reach very large numbers of poor
households.
According to Alexander (2011) finding, in study title of the impact of microfinance on the
livelihoods of women in rural communities: a case study of Jaman south district, Ghana, about
20.1 percent of the respondents were single whiles 67.5 percent were married. Those that were
divorced or separated formed 12.4 percent. The marital status of respondents is directly linked to
their living conditions as the responsibility to perform certain duties, like housekeeping,
children‟s education and provision of good health for the family is associated with ones marital
status. Divorced/Separated parents mostly performed their roles as single parents which normally
affect their living conditions. The single women are also would-be married people and therefore
need to be financially independent so as to be useful to themselves and their would-be family.
A study of 16 different MFIs from all over the world pointed out that having access to MF
services have led to an enhancement in the quality of life of clients, had increased their self
confidence, and had helped them diversify their livelihood security strategies and thereby
increase their income (Robinson, 2001). Health care and education are two key areas of non-
financial impact of MF at a household level. Wright (2000) stated that from the little research
that has been conducted on the impact of MF interventions on health and education, nutritional
indicators seem to improve where MFIs have been working. MF interventions have been shown
to have a positive impact on the education of clients‟ children because one of the first things that
Page 35
34
poor people do with new income from micro-enterprise activities are to invest in their children‟s
education (Littlefield et al., 2003).
Some households perceived that group lending is difficult to access finance since every
individual in a group is responsible to repay the loan, if loan default occurs in one of the
individuals (Getaneh, 2005). Others perceive group lending is the best solution for those who
have no other alternative to get loan from any source individually (Mekonnen, 2008).Therefore,
it has been expected that group lending influence access to finance both positively and
negatively.
The other expected factor which may affect the female household heads to access microfinance
is attitude towards risk. According to Yirga Workneh (2012) finding, attitude towards risk taking
was significant at 10 percent probability level and negatively related with the state of access to
credit as expected. The odds favoring access to credit decreases by a factor of 2.305 for farmers
who are risk averse. This is consistent with the prior expectation. The possible explanation is that
farmers fear to take credit because it is difficult to repay the loan if the risk appears and they
might be penalized based on the previous agreement. This study is also consistent with the
finding of Bigstenet al., (2003).
Ernest (2008) argues that poorly educated people face challenge in accessing financial services.
Because, it is difficult for them to analysis credit risks and the profitability of a loan or savings
scheme, to provide all documents and information (such as a business plan) required to apply for
a loan, and to understand conditions and contracts. Some institutions fail to communicate interest
rates and commissions in a transparent manner, and small prints in contracts can contain
additional costs for borrowers. Schlaufer (2008) also added, besides the challenges financial
institutions face, not considering the needs of rural households and small entrepreneurs related to
loan size, loan distribution time and the repayment period are the other weakness of both formal
and informal financial institutions, which is, simply not tailored to the needs of rural clients.
Financial institution and its policy will often determine credit access. Syedaet al. (2008) reported
that loan duration, terms of payment, required security of credit had negative influence on access
to credit. Hoque and Itohara (2009) indicated that the provisions of supplementary services and
Page 36
35
interest rate do not fit the needs of the target groups, potential borrowers will not apply for credit
even where it exists and when they do, they will be denied access.
The duration at which the borrower should repay the loan is the other factors which expect to
affect the women to access microfinance institution. According to Yirga Workneh (2012) study
finding on the title of the determinants of access to credit by smallholder farmers: the case of
Gozamen District, East Gojjam Zone, Ethiopia; the repayment period or loan duration is varying
from institution to institution. His survey result revealed that, of the total sample respondents
more than 63 percent of the respondents believed that the loan duration were inadequate. It
shows that with this short duration of loan, farmers cannot perform profitable activity; as a result
they retreat applying for credit.
The study result reviled by Mekonnen (2010) as sited in Anwar Ahmed (2015) indicated
household income determines the household„s ability to secure food. It is an important variable
which explains the characteristics of food secured and food insecure households and treated as
continues variable. Income earned from any source improves the food security status of the
household. High-income families are less likely to be food insecure.
In general, a study in various micro credit programs recalled that there is a different in impact
throughout the world. Some programs exhibits higher impact at household level compared to
others types of impacts. Some other countries also experienced quite the opposite. Impact survey
on three countries namely Zimbabwe, India and Peru indicated that client in India showed
significantly positive impact at enterprise level where as at household level it was found
insignificant. And for Zimbabwe, the extent of impact was found in between of the two
conditions (Snodgrass, 2002).
Page 37
36
2.9. Conceptual framework
It is clear that several factors may help to explain the impact of microfinance institution on
building the resilience of female headed households. However, due to the difficulty of getting
adequate fund and sufficient time to collect all the required data, this research considers the
important variables in the study area. In addition, the relationship generally exists among number
of factors, precluding their inclusion in the analysis efforts. Considering this limitation, therefore,
those factors considered and defined to exert the largest impact on access to microfinance. Based
on the objective of the study, the independent variables selected to achieve the ultimate objective
of the study are broadly categorized in to social, economic, demographic and risk factors related
variables. The relationship between dependent and independent variables of the study are
described in the figure below.
In fact, this pandemic indicator was emphasized on the relationship between the explanatory
variables with the dependent variables. However, the relationships of the explanatory variables
with themselves were not shown in the diagram. This does not mean that there is no relationship
between explanatory variables, but simply to concentrate on their relationship with the dependent
variables rather than relationship among themselves. The conceptual framework described below
incorporates these factors which have direct contribution for the impact of microfinance
institution towards the building of resilience of FHH.
Page 38
37
Figure1: Conceptual framework
Source: From difference literature perspective and modified by the researcher.
Societal factors
- Preference for group lending
Access to
Microfinance
Risk factors
- Attitude towards risk taking
- Experience of the HH in loan use
Demographic factors
- Age of the HH - Marital status of
the HH - Educational statues
of the HH -
Economic factors
- Household income - Adequacy of loan
repayment
Page 39
38
CHAPTER THREE
MATERIALS AND METHODS
In this chapter the research methods that the researcher was used discuss in a better manner. The
chapter included description of the study area, materials used, methods of data collection and
methods of data analyses.
3.1. Description of the Study Area
The study was conducted in Jigjiga City, Somali Regional State of Ethiopia situated at an altitude
of 1600 meter above sea level. It is 630 km from Addis Ababa. Climate is arid in most parts of
the region and weather is therefore hot in most parts of the year, with mean temperatures ranging
from 18 to 45oC. Annual rainfall ranges from 150mm in the low lying areas of the region to
660mm received in high altitude areas. Moreover, 85% of the region‟s population are pastoralists
(Abdikadir Hasan, 2016) Unlike other parts of the region, the rainy season in Jigjiga has three
sub seasons locally known as; Dira‟ (April - May), Hagaa(June ‐ July), and Karan (August ‐
September) which are highly important for crop cultivation and pasture availability for livestock.
Furthermore, the dry season, Jilaal(October ‐ March) is divided in to two sub seasons:
Deyr(October ‐ November) and Kalil(December-March) (Devereux, 2006; SC UK, 2007).
Jigjiga market is a source of crop production and all species of livestock (both export and local
qualities). Even in a bad year Jigjiga market is fairly stable in relation to supply and prices.
However, seasonal fluctuations can be expected, AbdikadirHasan(2016).
The latest information of Jigjiga population indicates that there are 197,438 inhabitants in
Jigjiga, 111,091 of them are female while the remaining 86,347 are women (ESBOFED, 2016).
There are many banks in Jigjiga, but, the Somali microfinance institution is the only
microfinance institution in the city.
Page 40
39
Figure 2: Map of the study area
Source: Jigjiga sedentary farming livestock (UK, 2000)
3.2. Materials used
To collect the data, the study used structured questionnaire. In fact, the structure questionnaires
are one of the most commonly used data collection technique and tools within the survey
strategy. Moreover, questionnaires are the most convenient methods for the respondents to
answer the questions, because it gives them an opportunity to choose a possible alternatives
which is given in a structured inquires, hence on the part of the SMFI customers, structured
questioners were distributed through the selected samples.
Page 41
40
3.3. Methods of Data Collection
In order to gather the data from relevant sources, both primary and secondary data collection
instruments had been used. The primary data were collected in the form of personal interviews
both SMFI loan receivers and non-received women by using a structured questionnaire which
was designed for both of the two groups.
Secondary relevant information was extracted from relevant textbooks, newspapers, reports
articles, journals, bulletins and documents presented by corporate financial analysts and policy
issues relating to the operations of SMFI and other stakeholders like Somali Regional Bureau of
finance and economic development (ESBOFED), other financial institutions, and National Bank
Publications.
3.3.2. Sample size determination
Jigjiga City will be selected purposively because more Micro-finance service beneficiaries are
found in there. From the city the researcher has selected five Kebeles purposively, because those
Kebeles are mostly populated by Somali microfinance institution‟s clients those engaged in
various activities for at least two years, in fact the impact assessment required long time but the
number of client start to increased before two years. If the researcher use the data above five
years the target population was not sufficient. So, the researcher choose clients benefited above
two years; because, they are well informed and know much about the pros and cons about
microfinance, so they can reflect better to our questionnaire.
For a comparison purpose the sample respondents were both loan receivers and non-receivers
using multistage proportional sampling followed by simple random sampling technique.
Information about the loan receivers had been taken from Somali microfinance institution. The
data were appeared in English language but during interview respondents had been asked in local
language which is Somali and entered and analyzed in the English version.
Page 42
41
There are several approaches to determine the sample size. These include using a census for
small populations, duplicating a sample size of similar studies, using published tables, and
applying formulas to calculate a sample size.
This study was applied a simplified formula provided by Yamane (1967).
21 eN
Nn
............................................................................ (1)
Where n is the sample size,
N is the population size (total household size), and
E is the level of precision.
Based on the data received from Jigjiga City Administration, the total number of female headed
households were 4,040.Therefore, to determine the required sample size with 92% confidence
interval and8%,level of precision the sample size have been determined.
150
08.040401
40402
n
And then by proportionate random samples of 75 households of SMFI Clients and 75 households
from the non-clients were selected. The selected participants by proportional sampling
techniques are indicated in the table below (Table 1).
Page 43
42
Table 1: sample size by Kebeles
S/N Sampled
Kebeles
Female Headed
HH
Clients Sample
from
clients
Non-clients Sample
from non-
clients
1 Kebele 01 111 54 13 57 10
2 Kebele 02 102 44 11 58 11
3 Kebele 03 114 58 14 56 10
4 Kebele 04 253 109 26 144 27
5 Kebele 18 142 48 11 94 17
Total 722 313 75 409 75
Source: Jigjiga city council administration, 2017
In here any one should remembered that the client and the non client size purposively divided
into equal part for comparison. So, due to this improper size determination selection bias would
be expected to determine for the population. That is, in the case of this reason it may in turn limit
the ability to make broader generalization from the study undergone.
3.3. Methods of Data Analyze
The empirical data were analyzed using both descriptive and inferential statistical tools followed
by propensity score matching (PSM). In what follows, these tools are outlined and discussed.
3.3.1. Descriptive statistics
Descriptive statistics which were applied for analyzing data for this study include mean, standard
deviation, percentages, graphs and tables. For the case of inferential statistics the researcher used
modeling and Propensity Score Matching (PSM) and various statistical tests which are conducted
in the process of data summarizing.
Page 44
43
3.3.2. Propensity score matching (PSM) method
The impact of SMFI on building the resilience of FHHs is the difference in households‟ mean
income change with the institution and without the institution. However, households
participating in the institution cannot be simultaneously observed in two states. A household can
either be in the institution or outside the institution. Thus, the fundamental problem of such an
impact evaluation is a missing data problem. In other words, anybody expected to answering the
research question which stated as “what would have been the resilience of participating
households be if SMFI was not in place?” Hence, this study applies a propensity score matching
technique, which is a widely applied impact evaluation instrument in the absence of baseline
survey data for impact evaluation.
Rosenbaum and Rubin (1983) were the first to develop the PSM statistical tool. The technique
has attracted attention of social program evaluators since the last fifteen years (Jalan &
Ravallion, 2003; Dehejia&Wahba, 1999). The present study also used a PSM technique to
address its main objectives. The PSM technique enables for the user to extract from the sample
of non-participating households, which means a person who do not access microfinance
institution, a set of matching households that look like the participating households those access
MFI in all relevant pre-intervention characteristics. In other words, each participant household
with a non-participant household has (almost) the same likelihood of participating into the
institution.
PSM is preferred to the traditional regression method in several ways. Among others, PSM
compares outcome for observations, who share similar observable characteristics. Moreover,
PSM only compares households lay in the common support and excluded others from the
analysis. This study attempts to estimate the average impact of treatment on treated (ATT). In
this thesis “treatment” implies participation in the Somali microfinance Institution, and “impact”
meant for the change of resilience using income as an outcome indicator. On the other hand,
“control” stands for non-participant/non-treated households used for comparison.
Page 45
44
According to Caliendo and Kopeinig (2005), there are steps in implementing PSM. These are
estimation of the propensity scores, choosing a matching algorism, checking on common support
condition and testing the matching quality.
3.3.2.1. Procedures of propensity score estimation
The first step in PSM method is to estimate the propensity scores. As described by Rosenbaum
and Rubin (1983), matching can be performed conditioning on P(X) alone rather than on X,
where P(X) = Prob(D=1|X) is the probability of participating in the program conditional on X. If
outcomes without the intervention are independent of participation givenX, then they are also
independent of participation given P(X). This reduces a multidimensional matching problem to a
single dimensional problem.
A logit model was used to estimate propensity scores using a composite of pre-intervention
Characteristics In estimating the logit model, the dependent variable was participation, which
takes the value of 1 if a household participated in the program and 0 otherwise. The
mathematical formulation of logit model is as follows:
i
i
z
z
ie
e
1 ----------------------------------------------------------- (1)
Where, Pi is the probability of participation,
)2.........(......................................................................1
0
n
i
iii UXaaZ i
Where,
n --,- 3, 2, ,1i
0aIntercept.
iaRegression coefficients to be estimated
Page 46
45
iUA disturbance term, and
iXPre-intervention characteristics or the independent variables
The probability that a household belongs to non-participant is:
ziie
P
1
11
-------------------------------------------------------------- (3)
Then the odds ratio
)4.......(..................................................
11
1
1i
i
i
i
z
z
z
z
i
i e
e
ee
p
p
The left hand side of equation (4)
i
i
p
p
1is simply the odds ratio in favor of participating in
SMFI.
It is the ratio of the probability that the household would participate in the SMFI to the probability
that he/she would not participate in the SMF. Finally, by taking the natural log of
Equation (4) the log of odds ratio can be written as:
)5(........................................ln1
ln1
0
.1
0
n
i
ii
UiaiXia
i
i
i UXaaep
pz i
n
i
Where, iz is log of the odds ratio in favor of participation in the SMFI, which is not only
linear in iX but also linear in the parameters?
The effect of household‟s participation in the SMFI on a given outcome (Y) is specified as:
)6..(................................................................................01 iiiii DYDYT
Where
Page 47
46
iT Is treatment effect (effect due to participation in SMFI),
iY Is the outcome on household I,
iD Is whether household i has got the treatment or not (i.e., whether a household
participated in the SMFI or not).
Nonetheless, since 1ii DY and 0ii DY cannot be observed for the same household
simultaneously, estimating individual treatment effect iT is impossible and one has to shift to
estimating the average treatment effects of the population than the individual one. The most
commonly used average treatment effect estimation is the „average treatment effect on the treated
( ATTT ) which is specified as:
)7.......(..............................1
)0(1
)1(1
DY
ED
YE
DTETATT
Since the counterfactual mean for those being treated,
1)0(D
YE is not observed, there is a
need to choose a proper substitute for it to estimate ATT. Though it might be thought that using
the mean outcome of the untreated individuals,
0)0(
DY
E as a substitute to the counterfactual
mean for those being treated,
1)0(D
YE is possible, it is not a good idea especially in non-
experimental studies. This is because it is likely that components which determine the treatment
decision also determine the outcome variable of interest.
In our particular case, variables that determine household‟s participation in the SMFI could also
affect household‟s resilience and income. Therefore, the outcomes of individuals from treatment
and comparison group would differ even in the absence of treatment leading to a self-selection
bias. However, by rearranging and subtracting
0)0(
DY
E From both sides of equation 7, ATT can be specified as
Page 48
47
)8......(....................1
)1(.
1)0(
.0
)0(1
)1(
DY
ED
YET
DY
ED
YE ATT In
equation (8), both terms in the left hand side are observables and ATT can be identified if no self-
selection bias. That is, if and only if 01
)1(.
1)0(
DY
ED
YE
However, this condition can be ensured only in a randomized experiments (i.e., when there is no
self-selection bias). Therefore, some identified assumptions must be introduced for non
experimental studies to solve the selection problem. Basically there are two strong assumptions to
solve the selection problem. These are: Conditional independence assumption and common
support.
3.3.2.2. Matching estimators
After estimation of the propensity scores, seeking an appropriate matching estimator is the major
task of a program evaluator. There are different matching estimators in theory. Below, only the
most commonly applied matching estimators are described.
Nearest Neighbor (NN) Matching: it is the most straightforward matching estimator. In NN
matching, an individual from a comparison group is chosen as a matching partner for a treated
individual that is closest in terms of propensity score (Caliendo&Kopeinig, 2005).NN matching
can be done with or without replacement options. In the case of the NN matching with
replacement, a comparison individual can be matched to more than one treatment individuals,
which would result in increased quality of matches and decreased precision of estimates. On the
other hand, in the case of NN matching without replacement, a comparison individual can be
used only once. Matching without replacement increases bias but it could improve the precision
of the estimates. In cases where the treatment and comparison units are very different, finding a
satisfactory match by matching without replacement can be very problematic (Dehejia&Wahba,
2002). It means that by matching without replacement, when there are few comparison units
similar to the treated units, the users may be forced to match treated units to comparison units
that are quite different in terms of the estimated propensity score.
Page 49
48
Caliper Matching: The above discussion tells that NN matching faces the risk of Bad matches,
if the closest neighbor is far away. To overcome these problem researchers use the second
alternative matching algorism called caliper matching. Caliper matching means that an individual
from the comparison group is chosen as a matching partner for a treated individual that lies
within a given caliper (propensity score range) and is closest in terms of propensity score
(Caliendo&Kopeinig, 2005).If the dimension of the neighborhood is set to be very small, it is
possible that some treated units are not matched because the neighborhood does not contain a
control unit. On the other hand, the smaller the size of the neighborhood the better is the quality
of the matches (Becker &Ichino, 2002). One problem in caliper matching is that it is difficult to
know a priori what choice for the tolerance level is reasonable.
Kernel Matching: this is another matching method whereby all treated units are Matched with a
weighted average of all controls with weights which are inversely proportional to the distance
between the propensity scores of treated and controls (Becker &Ichino 2002Venetoklis,2004).
Kernel weights contribution of each comparison group member so that more importance is
attached to those comparators providing a better match. The difference from caliper matching,
however, is that those who are included are weighted according to their proximity with respect to
the propensity score. The most common approach is to use the normal distribution (with a mean
of zero) as a kernel, where the weight attached to a particular comparator is proportional to the
frequency of the distribution for the difference inscores observed (Brysonet al., 2002).According
to Caliendo and Kopeinig (2005) a drawback of this method is that possibly bad matches are
used as the estimator includes comparator observations for all treatment observation. Hence, the
proper imposition of the common support condition is of major importance for kernel matching
method. A practical objection to its use is that it will often not be obvious how to set the
tolerance. However, according to Mendola (2007) kernel matching with 0.25 band width is most
commonly used.
The question remains on how and which method to select. Clearly, there is no single answer to
this question. The choice of a given matching estimator depends on the nature of the data set
(Brysonet al., 2002). In other words, it should be clear that there is no `winner' for all situations
Page 50
49
and that the choice of a matching estimator crucially depends on the situation at hand. The
choice of a specific method depends on the data in question, and in particular on the degree of
overlap between the treatment and comparison groups in terms of the propensity score. When
there is substantial overlap in the distribution of the propensity score between the comparison
and treatment groups, most of the matching algorithms will yield similar results
(Dehejia&Wahba, 2002). To give an example, if there are only a few control observations, it
makes no sense to match without replacement. On the other hand, if there are a lot of comparable
untreated individuals it might be worth using more than one nearest neighbor to gain more
precision in estimates (Caliendo and Kopeinig, 2005).
3.3.2.3. Testing the matching quality
One important concern that should be taken care of while doing PSM is balancing test. While
Differences in covariates are expected before matching, these should be avoided after.
The primary purpose of the PSM is that it serves as a balancing method for covariates between
the two groups. Consequently, the idea behind balancing tests is to check whether the propensity
score is adequately balanced. In other words, a balancing test seeks to examine if at each value of
the propensity score, a given characteristic has the same distribution for the treatment and
comparison groups. The propensity scores them serve only as devices to balance the observed
distribution of covariates between the treated and comparison groups. The success of propensity
score estimation is therefore assessed by the resultant balance rather than by the fit of the models
used to create the estimated propensity scores (Lee, 2006). Finally, using predicted probabilities
of participation in the program, i.e. propensity score match pairs are constructed using alternative
methods of matching estimators.
Then the impact estimation is the difference between simple mean of outcome variable of
interest for participant and non participant households (Lee, 2006).
3.4 Definition of Variables
Page 51
50
A combination of demographic, social, economic and risk factors will be used to explain
household access to Somali microfinance institution as well as the resulting outcomes in terms of
resilient building.
The dependent variable of the model:
Access to micro finance: is defined as the act of borrowing money from micro finance
institution by female headed households, which will be paid back later. This variable is a
dichotomous; taking two values that is“1”if the woman has borrowed money and “0”if the
women did not.
The independent variables of the model: the independent variables which are expected to have
association with the access to MFI are presented below:
Age of Household Head (AGE): It is a continuous variable which is measured in number of
years completed by the women. Age has vital role in the access to finance. As child bearing is
highly desired and expected culturally after marriage; a need for children couples have is very
high. Hence, it is expected that women with higher age will have more exposure and access to
finance because of increase in number of children will increase the burdens and responsibility of
household head, i.e., age may have positive relationship with access to finance.
Educational status of household head (EHH): Education status of women is believed to
increase women‟s capacity and empower them to claim and use social and economic services
including finance and other services than illiterate women. So, educational status affects
positively women‟s attitude towards finance prospects. Hence, education level of the respondent
is expected to influence the access to finance positively.
The variable is measured as a dummy variable: takes the value of 1, if the woman is literate and
0, otherwise.
Household Family size (HHFS) refers to the total number of household members who lived and
eat with same pot at least for six months. It is an important variable which determines the state of
household micro-finance access. Thus, it is hypothesized that the family with relatively large
Page 52
51
number of family members may positively affects household access to microfinance and more
likely to be participant, i.e., positive relationship with access to MFI.
Marital status of household head (MAS): it is dummy variable and it takes the value of 1 if
she is married and 0 otherwise. For the purpose of the study divorced, separated, and widowed
women are considered as single because the purpose of the study is to see the difference
regarding access to finance between those who have husband and those who don‟t have.
Hence, access to finance for not married woman may be more as compared to married woman.
Hence, it is expected that unmarried woman has higher access to finance than those
counterparts with their husbands.
Attitude towards risk (A. TORIKE): The other factor, which influences the household‟s
Access to finance, is their attitude towards risk. Many households, as can be expected, are very
risk averse that even when finance is available, they do not like to venture into activities. This is
due to risk of repaying loan, it would be measured based on their positive or negative perceptions
towards risk. This is a dummy variable which takes “1” if the respondents are risk avert to take
loans and “0” otherwise. Therefore, it is expected that households who are risk averse would not
demand loan and it affects access to microfinance negatively.
Experience of the household head in loan use (EX-LUSE): It is a continuous variable. It is the
total number of years of experience that the household head has obtained in use of loan from
different sources. Households who have experience in use of loan and who lived to the best
expectations of the lenders would develop reputation, and they might have demonstrated their
loan worthiness and become trustworthy (Atieno, 2001). Similarly, households who had
experience in loan use have developed confidence and reputation in loan acquisition and
repayment (Belay, 1998). Therefore, it is hypothesized in the present study is that, experienced
households may have better access to micro-finance than less experience households.
Preference for group lending (P-GROUPLE): Different lending institutions have their own
lending arrangements some follow individual and others use group method that can serve as
collateral. It is a dummy variable which takes a value “1” for those who prefers group lending
“0” otherwise.
Page 53
52
Adequacy of loan repayment period (AD-LRE): This refers to the time period at which the
borrower should repay the loan. Different financial institutions have their own rules and
regulations that limit the time at which the borrower should repay the loan. If household fail to
repay on time they may be liable to some measures based on previous obligation made with the
lender (Syedaet al., 2008). Due to these reason households fear taking loans from lending
institutions. Adequacy of loan repayment period, therefore, has been hypothesized to influence
access to credit positively. This is a dummy variable which takes a value “1” for those who
perceive it as adequate and “0” otherwise.
Page 54
53
CHAPTER FOUR
RESULT AND DISCUSSION
This chapter contains result of data findings together with their discussion. The first section of
this chapter presents the descriptive and inferential analysis results of the study. Results of the
descriptive and inferential analysis are presented in the form of mean, standard deviation,
percentages, independent sample t-Test for qualitative data and chi-square test for quantitative
data as discussed below. This is followed by the discussion of the econometric model results.
Propensity Score Matching (PSM) method was deployed to estimate the impact of MFI on
building the resilience of FHH community those is living in Jigjiga City. The findings intended
to answer the research questions stated in the introduction part of this study.
4.1. Demographic characteristics of the study
According to the result presented in the table below, the average age of the sampled households
was 39 years. However, the mean age of the client and non-client of the micro finance institution
household was 42 and 36 years respectively. The age difference between client and non- client
was found to be significant (p=0.000) with the t-value (t=141.67) which is different with the
hypothesized relationship with participation. Likewise, Bereket Gebremedhin (2008) reported
that age of the household head has significant effect on the living condition of the households
with low income earning. In fact as the result illustrated more on that client of the microfinance
institution are older than the non-client households which mean older age groups were targeted
to the program.
Age is one of the factors useful to describe the impact of microfinance institution on building the
resilience of female headed households. This variable also helps to provide clue about the
relationship with the client and non-client of the microfinance institution on the building of
resilience in the study area. Moreover, from the total participants about 50% of them come from
the client category those are old aged people who may aware of the microfinance importance.
Page 55
54
Table2: Age of Sample Respondents interviewed in Jigjiga City, Fafan Zone of Somali Regional
State of Ethiopia
Variables Client Non client Total t-test p-value
Mean SD Mean SD Mean SD
Age of the respondents 42.17 8.31 36.25 6.70 39.21 8.09 141.67 0.000**
Source: own survey, 2017
The sample was composed of both married and single female household heads. Of the total
sample households 50 (33.3%) and 100 (66.7%) were married and single female households,
respectively. Married female households represent about 32 (42.7%) and 18(24.0%) from
microfinance client and microfinance non-client groups respectively. As the table indicated
below, single female households had higher percentage of non-client microfinance institution
headed households 57(76.0%) as compared to married women participating households. In line
with this, from the total respondents, 75 (50.0%) of the participants comes from client group,
whilst, the remaining 75 (50.0%) respondents comes from non-client households. Therefore,
marital status of the household is statistically significant (p=0.015) between clients and non
clients.
Moreover, from the total married female headed HHs living in the area, the institution is able to
target them to improve their life better. The result of this study is support the result obtained by
Alexander (2011) and reported as marital status has relation to the building the resilience of
female headed households.
Table3: Marital status of Sample Households interviewed in Jigjiga City, Fafan Zone of Somali
Regional State of Ethiopia
Variables Client Non client Total χ2-test p-value
N % N % N %
Marital
status of the
HH
Married 32 42.7 18 24.0 50 33.3 5.88 0.015**
Single 43 57.3 57 76.0 100 66.7
Total 75 100.0 75 100.0 150 100.0
Source: own survey, 2017
Page 56
55
4.2. Social characteristics of the study
Women household heads in the study area choose the group and individual lending system.
According to the results of the finding presented in the table below, majority of the participants,
102 (68.0%), were preferred personal lending system, whereas, the remaining 48 (32.0%)
respondent prefer to lend with group. The result also shows that the perception level of the study
participants of group lending is difficult to access finance have a-significant difference between
the microfinance client and non-client households (p=0.000). The majority of targeted FHHs,
41(54.7%), those are found in the microfinance institution had preferred personal loan system.
Likewise, about 61 (81.3%) of the non-client FHHs were the same system; whereas, the
remaining 14(18.7%) non-clients preferred group loan.
In fact, group lending is not difficult to access finance if and only if the group members are
jointly accountable for the repayment in the event of default. But, if one of the group members is
absent, unless it is covered by the rest of the members‟problem of repaying the loan may face a
difficulty. The finding of Mekonnen (2008) opposed the finding of this study. Preferring of
group lending is hypothesized as positive correlated to being clients of microfinance institution
beneficiaries.
Concerning with educational status of the respondents of the result presented in the table below
shows that 111 (74.0%) of the participant women household heads were illiterate. From them 58
(77.3%) of the participants were not the microfinance beneficial; i.e., less number of literate
people 17 (22.7%) were non-client study participants. The statistical analysis revealed that there
is no significant difference between client and non-client microfinance female households in
terms of education status of the household heads (p=0.352). The finding of Robinson (2001) and
Wright (2000) results are against the finding with this study.
It is assumed that a literate household head is often tends to adopt new skills, ideas and which in
turn have positive attitude on the impact of microfinance institution on the building resilience of
FHHs. However, according to this study finding education level does not have a significant
Page 57
56
effect on the two groups. Littlefield et al., (2003) findings which are stated that MF interventions
indicated a contradict result with this findings.
Table4: Social Characteristics of Sample Households interviewed in the study areas
Variables Client Non client Total χ2-test p-value
N % N % N %
Preference
for group
lending
Not- preferred 41 54.7 61 81.3 102 68.0 12.26 0.000**
Preferred 34 45.3 14 18.7 48 32.0
Total 75 100.0 75 100.0 150 100.0
Educational
status of the
HH
Illiterate 53 70.7 58 77.3 111 74.0 0.866 0.352
Literate 22 29.3 17 22.7 39 26.0
Total 75 100.0 75 100.0 150 100.0
Source: own survey, 2017
4.3. Risk factors
Some women are risk taker, whereas, some are risk averter. This study tried to identify how
many of the study participants have an attitude towards taking a risk for loan. The result
indicates, risk taker female headed households represent about 62 (82.7%) and 10 (13.3%) from
clients and non-client of microfinance institution respectively. According to the result indicated
in the table below, the non-client households had significantly higher in number of risk averter
headed households 65 (86.7%) as compared to risk taker participating households. In line with
this, of the total 150 (100.0%) sample households, 72 (48.0%) and 78 (52.0%) were risk takers
and risk averter households respectively. Therefore, attitude towards risk taking is statistically
significant (p=0.00) and positive relationship with the participation in microfinance institution
with χ2=77.2. This implies that, risk taker female households had a capability to access
microfinance institution to have better exposure on the building resilience to their households.
But from the total risk averter study participants involved in the area the majority, 65 (86.7%) of
them were non-clients, whilst, the reaming 13 (17.3%) women participants are clients.
By nature, most of the poorest societies are risk averse. Risks associated with the inflexible
repayment period of lending institutions influence women attitude towards credit use and make it
difficult to repay the loan if the risk appears. The result of this study is the same as the result
Page 58
57
obtained by Ernest (2008) and Schlaufer (2008) reported as attitude towards risk taking has a
relation to access microfinance.
Table5: Attitude towards risk taking of the sample households interviewed in the study areas
Variables Client Non client Total χ2-test p-value
N % N % N %
Attitude
towards risk
taking
Risk taking 62 82.7 10 13.3 72 48.0 72.2 0.000**
Risk averter 13 17.3 65 86.7 78 52.0
Total 75 100.0 75 100.0 150 100.0
Source: own survey, 2017
The study result revealed that the client women had more experience in loan taking. The average
number of years of experience that the household head has obtained in use of loan from different
sources was 1.75 year.
According to the result presented in the table below clients‟ who access microfinance had an
experience of 2.67 years; while, the non-client female household heads were 1.75 years of
experience in loan taking. Moreover, the average year deference between clients and non-clients
female household heads was 1.745.
Likewise, the below table elaborated that, there is significance mean difference in years of
experience in loan taking between clients those accessed microfinance institution and non- client
female households (p=0.000) which is the same as the hypothesized relationship with
participation. For instance, this result is the same as Syedaet al. (2008) and Hoque and Itohara
(2009) findings. According to their terminology, if the loan experience of the respondents
escalating, the practice on accessing a credit will increase.
Table6: Number of years of experience in loan taking from sample respondents interviewed in
Jigjiga City, Fafan Zone of Somali Regional State of Ethiopia
Variables Client Non client Total t-test p-value
Mean SD Mean SD Mean SD
Number of years of respondents
experience in loan taking 2.67 1.92 1.92 0.64 1.75 0.29 5.99 0.000**
Source: own survey, 2017
Page 59
58
4.4. Economic factors
As the repayment period or loan duration of the study result presented in the table below,
according to the client perspective the average loan period of Somali Microfinance Institution is
0.85 year. Contrarily, the non-client respondents, those borrowed money from other than Somali
Microfinance Institution, enforced to repay the adequacy loan on the average 0.33 years. That is
the non-client participant repayment duration shorter than the clients. In other words, study
participants those are not the client of SMF but received loan from other bodies enforced to pay
their loan earlier.
The finding of the result also showed a statistically significant difference in average repayment
period of clients and non-clients (p=0.000) which is the same as the hypothesized relationship
with other participation. For instance the result of Yirga Workneh (2012) is the same as the
finding of this study. Clients of the program have more loan duration than the non-client
households.
The mean difference between the clients and non-clients study participants‟ repayment period is
0.52. In other word the loan duration of the client is a bit higher than the non-client of Somali
Microfinance Institution. When the loan period is higher the clients have more time to pay and
use the loan to replicate more. In other word the clients more profitable than the non-clients.
Table7: Adequacy of loan repayment duration from sample respondents interviewed in Jigjiga
City, Fafan Zone of Somali Regional State of Ethiopia
Variables Client Non client Total t-test p-value
Mean SD Mean SD Mean SD
Adequacy of loan repayment 0.85 0.36 0.33 0.48 0.52 0.07 7.59 0.000**
Source: own survey, 2017
Page 60
59
4.5. Monthly income
The HHs monthly income is one of the important factors that help HHs to create assets in such a
way that it contributes to the whole household consumption.
The average monthly income of sample households is 4,677.93 Birr (SD±1,934.303 Birr), i.e.,
the monthly income of the households is ranging from 6,612 to 2,744 Birr.
As the below table indicated, the mean monthly income obtained from the sampled client
households was 5,421.20; whereas, the non-client households had obtained 3,934.67 Birr. This
indicated that the average monthly income of the household those accessed microfinance had
better income compared with the non-client correspondence.
Moreover, the result revealed the mean monthly income of the household using t-test had
indicated that there was a significant mean difference among the client and non-client categories
(p=0.000).This implies that the average monthly income of the accessed microfinance institution
household is different with the non-client participants. The mean monthly income difference
between the client and non-client is 1,486.53 Birr. In other term clients get 1,486.53 Birr higher
income than the non-clients. This same finding was obtained with that of Mekonnen (2010) as
sited in Anwar Ahmed (2015) findings.
Table8: Average monthly Income equivalent of Sample Households interviewed
Variables Client Non client Total t-test p-value
Mean SD Mean SD Mean SD
Average
monthly income
5,421.20 1,951.9 3,934.67 1,611.2 1,486.53 292.43 5.08 0.000**
Source: own survey, 2017
4.6 Estimation Results
This section describes the steps followed to measure the impact of Somali Microfinance
Institution on Building the Resilience of Female Headed Households more precisely. It presents
estimation of the propensity scores matching methods used, common support region and
balancing test. It also explains the program across the participating households.
Page 61
60
4.6.1. Propensity score
This part presents the results of the logistic regression model employed to estimate propensity
scores for matching treatment household (the household who access microfinance) with control
households (the participants those do not access the microfinance). As specified earlier, the
dependent variable in this model is binary indicating whether the household had access the
microfinance which takes a value of 1 or 0 otherwise. The model is estimated with STATA 12
computing software using the propensity matching algorithm calledpsmatch2 was used for the
estimation purpose.
Heteroscedasticity test was conducted using Breusch-Pagen/Cook-Weisberg test before
proceeding to impact estimation. As a result, illustrated in the below table, reject the hypothesis
which is stated that the model is suffered by Heteroscedastic problem because the P-value
(0.513) is higher than the significant value (0.05). So, there was no need to make the standard
error robust.
Table9: Breusch-Pagan / Cook-Weisberg test for heteroskedasticity
Ho: Constant variance
Variables: fitted values of Access of microfinance
chi2(1) = 0.43
Prob> chi2 = 0.5130
Source: own survey, 2017
Similarly, Variance Inflation Factor (VIF) was applied to test for the presence of strong multi co
linearity problem among the independent variables. As the result indicated in the below table,
here was no explanatory variable dropped from the estimation model since no serious problem of
multi-co linearity was detected from the VIF results. Because, the calculated VIF results of all
variables are below 10, this indicated that there is no a serial correlation among the explanatory
variables.
Page 62
61
Table10: Multi co linearity Test among Explanatory Variables
Variable VIF 1/VIF
MONTHLYINC~E 1.48 0.677643
AGE 1.41 0.709965
RISKTAKING 1.35 0.740894
ADEQUACY 1.25 0.797073
LOANEXPRNCE 1.25 0.800172
EDUCATION 1.17 0.853915
GROUPLENDING 1.16 0.860601
MARITAL 1.05 0.952535
Mean VIF 1.27
Source: own survey, 2017
The table below shows the estimation results of the logit model. The estimated model appears to
perform well for the intended matching exercise, because, the p-value of the χ2test of the
goodness of fit tests (0.000) is lesser than the significant value (0.05), so, the hypotheses which
stated that the model is good accepted. The pseudo-R2 value is 0.60 which is lower in magnitude.
Therefore, a low R2 value means that client households do not have much distinct characteristics
overall and as such finding a good match between client and non-client households becomes
easier. Moreover, the pseudo- R2 indicates how well the regressors explain the participation
probability.
In addition, the result revealed, out of the eight explanatory variables selected, the program
participation is significantly influenced by seven explanatory variables such as preference for
group lending (GROUPLENDING), age of the household head (AGE), marital status of the
household (MARITAL), attitude towards risk taking (RISKTAKING), experience of the
household in loan use (LOANEXPERNC), average household income (MONTHLYINC~E) and
adequacy of loan repayment (ADEQUACY). Preference for group lending, age of the household
head, marital states of the household experience of the household in loan use, average household
income and adequacy of loan repayment had positive and significant effect on the access of
microfinance. What this means is that households those prefer to lend in the group get higher
probability to be the client of microfinance.
Page 63
62
Likewise, Household with older age and married respondents had a better preference to being
the client of the microfinance institution. The households those had better loan experiences are
more likely to access the institution, similarly, the time period at which the borrower should
repay the loan, i.e., the adequacy of loan repayment period has been influence access to credit
positively. When the monthly income of the household is escalating the interest of the lender
goes up to improve the household status better than before.
In the contrary, there is a negative relationship between attitude towards risk taking and access to
microcredit. Client households who have more risk averter were not fear to access the
microfinance institution.
Table11: Results of the Logistic Regression Model
ACCESTOMFI Coef. Std. Err. z P>z [95% Conf.Interval]
EDUCATION .2182335 .3702497 0.59 0.556 -.5074427, .9439096
GROUPLENDING 1.198419 .3285662 3.65 0.000*** .5544409, 1.842397
AGE 0.0639918 .0237026 2.70 0.007*** .0175356, .110448
MARITAL 0.7983846 .3614146 2.21 0.027** .0900251, 1.506744
RISKTAKING -1.138844 .3347788 -3.40 0.001*** -1.794998, -.4826895
LOANEXPRNCE 0.2070514 .0885081 2.34 0.019** .0335788, .3805241
MONTHLYINC~E 0.0001482 .0000898 1.65 0.099* -.0000278, .0003242
ADEQUACY 1.349518 .3505593 3.85 0.000*** .6624346, 2.036602
_cons -4.668179 .9883583 -4.72 0.000*** -6.605326, -2.731032
Number of obs 150
LR chi2(8) 125.29
Prob> chi2 0.0000
Log likelihood -41.328177
Pseudo R2 0.6025
Source: Source: own survey, 2017
***, ** and * means significant at the 1%, 5% and 10% probability levels, respectively.
Following that the second step in propensity score matching technique is to compute the
common support region. According to the procedure stated earlier, only observations in the
common support region are matched with out of the common support region and the other
groups should be out of further consideration. Once the common support region defined,
individuals that fall outside this region have to be disregarded and for these individuals the
treatment effect cannot be estimated. According to this study finding indicated in the figure
Page 64
63
below represents the distribution of the household with respect to estimated propensity scores. In
case of treatment households, most of them are found in the right direction of the distribution. On
the other hand, most of the control households are found in the left side of the distribution.
Source: own survey, 2017
Figure3: Kernel density of propensity scores
4.6.2. Matching program and non-program households
As stated before, three main tasks should be accomplished before one launches the matching task
itself. First, predicted values of program participation (propensity scores) should be estimated for
all households in the program and outside the program as done in the previous section in the
research method. Second, a common support condition should be imposed on the propensity
score distributions of household with and without the program. Third, discard observations
whose predicted propensity scores fall outside the range of the common support region.
0.1
.2.3
.4
kde
nsity lp
-4 -2 0 2 4x
Non Client Household Client Household
Total Household
Page 65
64
As shown in the table below, the estimated propensity scores vary between 0.0052 and 0.9995
(mean = 0.833924) for client or treatment households and between 0.0016 and 0.8712 (mean =
0.166074) for non program (control) households. The common support region would then lie
between 0.0016 and 0.9995. In other words, households whose estimated propensity scores are
less than 0.0016 and larger than 0.9995 are not considered for the matching exercise. As a result
of this restriction, there were no households discarded; because, there were no a score above and
below the extremes result. That is the study does not have to drop MFIs households from the
sample in computing the impact estimator.
Table12: Distribution of Sample Households by Estimated Propensity Scores
Group N Mean Std. Min Max
Treated Households 75 0.833924 0.2337 0.0052 0.9995
Control Households 75 0.166074 0.2374 0.0016 0.8712
Total Households 150 0.500000 0.4091 0.0016 0.9995
Source: own survey, 2017
The two serial graphs those are indicated in the below figures portrays the distribution of
estimated propensity scores, with and without the imposition of the common support condition,
for client and non-client households, respectively. Most of the program households have
propensity score greater than 0.8712; whereas, majority of the non-client households have
propensity score around 0.166.
Figure4: Kernel density of propensity score of program households Source: own survey, 2017
01
23
kden
sity _
psco
re
0 .2 .4 .6 .8 1propensity scores BEFORE matching
Treated households , before common support condition
Treated households, after common support condition
Page 66
65
Figure5: Kernel density of propensity scores of non participant households
4.6.3. Choice of matching algorithm
Different alternatives of matching estimators were conducted to match the treatment program
and control households fall in the common support region. The decision on the final choice of an
appropriate matching estimator was based on three different criteria as suggested by Dehejia and
Wahba (2002). First, equal means test (referred to as the balancing test) which suggests that a
matching estimator which balances all explanatory variables; i.e., results in insignificant mean
differences between the two groups, after matching is preferred. Second, looking into pseudo-R2
value, the smallest value is preferable. Third, a matching estimator that results in the largest
number of matched sample size is preferred. To sum up, a matching estimator that balances all
explanatory variables, with lowest pseudo-R2 value and produces a large matched sample size is
preferable.
The below table presents the estimated results of tests of matching quality based on Dehejia and
Wahba (2002) performance criteria. As the result indicated, the Kernel matching with a band
width of 0.25 is the best estimator for the data which have because the lesser pseudo-R square is
01
23
kde
nsi
ty _
psc
ore
0 .2 .4 .6 .8 1propensity scores AFTER matching
Untreated households, after common support condition
Untreated households, before common support condition
Page 67
66
registered in there. According to Caliendo and Kopeinig (2005), after matching systematic
differences in the distribution of covariates between controlled and treated groups are not
expected therefore, the pseudo- R2 should be fairly low. To select the right matching estimator
results in the largest number of matched sample size is preferred can be considered. Therefore,
under the above preferable band width the entire samples had been considered.
Taking the above consideration into account, the estimated result of tests of matching quality is
taking based on the above mentioned criteria. Looking into the result of the matching quality,
Kernel with band width 0.25 was found to be the best for the indicator of accessing
microfinance. Hence, the estimation results and discussion for this study are the direct outcomes
of the Kernel with band width 0.25.
Table13: Performance of different matching estimator
Performance criteria
Balancing test* pseoudo-R2 matched sample size
Nearest Neighbor (NN) Matching
First Neighbor (NN(1)) 6 0.069 146
Second Neighbor (NN(2)) 7 0.053 150
Third Neighbor (NN(3)) 8 0.042 148
Fourth Neighbor (NN(4)) 8 0.025 150
Radius caliper
0.01 8 0.033 148
0.25 7 0.025 150
0.5 7 0.095 150
Kernel
With no band width 8 0.129 150
Band width of 0.1 7 0.071 150
Band width of 0.25 7 0.041 150
Band width of 0.5 6 0.054 150
Source: own survey, 2017
*Number of explanatory variables with no statistically significant mean differences between the
matched groups of program and non-program households.
4.6.4. Testing the Balance of propensity score and covariates The below table illustrated the balancing test of covariates, before and after the matching.
According to the result most of the variables do not have any significant change before and after
the treatment applied except the age of the household head. Because, the variance ratio of this
Page 68
67
variable is out of the range of the expected threshold (0.63, 1.58), but the remaining variables
stay on the interval. This indicates age had a statistical significant different effect before and
after matching.
Table14: Results of the Balancing tests of Covariates Using Kernel band width 0.25 Estimator
Variable Mean t-test
Treated Control % bias t-value P>(t) V(T)/V(C)
EDUCATION 0.293 0.680 -87.8 -5.10 0.000 0.95 GROUPLENDING 0.667 0.720 -12.3 -0.70 0.482 1.10 AGE 42.17 32.35 130 8.08 0.000 1.65* MARITAL 0.427 0.733 -65.9 -3.98 0.000 1.25 RISKTAKING 0.173 0.707 -124.7 -7.75 0.000 0.69 LOANEXPRNCE 2.667 3.720 -59.0 -3.34 0.001 0.98 MONTHLYINC~E 5421 3653 98.7 6.13 0.000 1.57 ADEQUACY 0.853 0.907 -12.7 -1.00 0.318 1.48
Source: own survey, 2017* If variance ratio outside [0.63; 1.58]
4.6.5. Treatment effect on the treated
In this section, the study provides evidence as to whether or not the access of microfinance has
brought significant changes on household income status. The estimation result presented in the
table below provides a supportive evidence of statistically significant effect of accessing
microfinance institution on building the resilience of female headed households; because, the t-
statistical value lay out of the range -2<t<2. Therefore, after controlling for pre-intervention
differences in demographic, social, economic and risk characteristics of the client and non- client
households, it has been found that, on average, the program has increased the monthly income of
the participating households by 1,215 Birr. Stated in other words, the program has increased the
average monthly income of the participating households nearly 32.6%.Therefore, the Somali
Microfinance Institution plays a significant role for the resilience of female headed household.
Table15: Average treatment effect on the treated (ATT) for monthly income of the respondents
Variable Sample Treated Controls Difference. S.E t-stat
Monthly income ATT 4937.67 3722.66 1215.01 523.79 2.32
Source: own survey, 2017
Page 69
68
4.6.6. Factors influencing treatment effect on the treated
In this section, the microfinance impact on the outcome variable monthly income is evaluated for
his significant impact on participant households, after the pre-intervention differences were
controlled. As the above result elaborated there is no any unmatched data set in the considered
sample households. Therefore, all the treated participants were included in this portion. Factors
influencing the treatment effect on the treated were identified using multiple regression models.
The main objective here is to examine if the effect of the program varies among the households
in the treatment group.
Before estimating the model, the data were checked for occurrence of strong multicollinarity
problem using VIF. According to the result revealed in the below table, all VIF values of the
explanatory variables were very much less than 10 which suggesting that there is no strong
correlation among the independent variables.
The dependent variable of the model here is amount of monthly income of the households
earned, compared to their comparator households in the non-client group. In other words, 75
microfinance client households were used in the analysis here.
Table16: Variance inflation factor for all explanatory Variables
Variable VIF 1/VIF
AGE 1.18 0.846374
LOANEXPRNCE 1.16 0.864477
RISKTAKING 1.15 0.872025
MARITAL 1.08 0.929826
EDUCATION 1.07 0.930315
ADEQUACY 1.06 0.940014
GROUPLENDING 1.06 0.944041
Mean VIF 1.11
Source: own survey, 2017
Table17 illustrated the estimated multiple regression model of the factors influencing treatment
effect on the treated. The p-value of the F test result, 0.000, is lesser than the significant value
0.05. So, the model is statistically significant, i.e., the explanatory variables included in the
model jointly influenced the dependent variable. The estimated regression results suggest that the
Page 70
69
effect of the microfinance on the client households is not uniform. In other words, the effect of
the treatment appears to have a strong relationship with the age of the household head (AGE) and
educational states of the household (EDUCATION).
More specifically, controlling for all other factors, the program tends to provide more income for
older aged households than younger households. This impliesold age household-heads are more
responsible to increase access of microfinance to elongatetheir family states. In this regard,
increasing the participation of women in such a program would contribute to overcome
incomerelated problems in the study area.
On the other hand, more educated households tended to receive smaller gains of income from
microfinance institution. That is, education level of the household resulted in a negative relation
with monthly income in a significant way. A unit increase in the education level decreases the
income from microfinance institutionby 773 Birr per month. This implies that educated
households less likely to access microfinance to enhance the gain from additional income
sources rather they may get income from other sources.
Table17: Results of the Multiple Linear Regression Model for annual income Variables
Coef. Std. Err. T P>t [95% Conf. Interval]
EDUCATION -772.947 447.799 -1.73 0.089** -1666.758 120.864 GROUPLENDING -306.845 429.336 -0.71 0.477 -1163.803 550.112 AGE 121.044 25.909 4.67 0.000* 69.330 172.758 MARITAL -418.685 412.323 -1.02 0.314 -1241.685 404.315 RISKTAKING -428.427 556.308 -0.77 0.444 -1538.822 681.969 LOANEXPRNCE 58.618 110.924 0.53 0.599 -162.786 280.023 ADEQUACY 77.702 573.316 0.14 0.893 -1066.641 1222.05
_cons 777.956 1299.133 0.60 0.551 -1815.125 3371.04
Sample size 75
R2 0.312
Adj R-squared 0.240
F (7, 67) 4.34
Prob>F 0.000*
Source: own survey, 2017
* and ** means significant at 1% and 10% probability levels, respectively.
Page 71
70
4.5.7. The sensitivity of the evaluation results
In this section the issue of sensitive with respect to the choice of the balancing scores is
addressed. Matching estimators work under the assumption that a convincing source of
exogenous variation of treatment assignment does not exist. The sensitivity analysis is done on the
estimated average treatment effect using alternative matching estimators.
As the result indicated, except the nearest neighboring (NN) matching, all other matching
estimators resulted in statistically significant effects of the program on participating household
monthly income. So, the sensitivity analysis of the distribution evaluated based on the significant
criteria.
Therefore, with respect to the monthly income the average treatment effect on the treated was
found to be sensitive, i.e., changing the variables will change the income levels of the HH. In
other words, the average monthly income of the households on the microfinance client was
found to be sensitive or not robust to the dummy confounder, i.e., adding or removing the
variables will bring a change. Thus, the impact estimate of average treatment is not sensitive to
unobserved selection bias.
Table18: Sensitivity analysis result
Matching Method Treated Controls Difference S.E. T-value
NN 2179.8 1862.26667 317.533333 271.091527 1.17
Caliper 2390.1447 1709.25498 680.889756 194.105982 3.51*
Kernel 4409.22 3257.17 1152.05 281.72 4.09*
Source: own survey, 2017
*Significant with 0.05 level of significance
Page 72
71
CHAPTER FIVE
CONCLUSION AND RECOMMENDATION
5.1. Conclusion
In this study the impact of SMFI on building the resilience of female headed households has
been studied using cross sectional data from Jigjiga City, Fafan Zone of Ethiopian Somali
Region. To give the answer of the giver research questions the outcomes categorized into client
and non-clientof MFI. In here anyone should remained that it is impossible to observe the same
object in two states simultaneously. In other words, the fundamental problem in any project
evaluation is the missing data problem. While the program evaluator observes the factual for an
object, it is impossible to observe the counter-factual for the same object.
Therefore, the primary data for this study collected from 150 households from both client and
non-clientof SMFI in Jigjiga City, Fafan Zone of Somali Regional State using a structure
questionnaire. The study for a comparison purpose select samplerespondents from both loan and
non-loan receivers female households using multistage proportional sampling followed by
simple random sampling technique. Information about the loan receivers had been taken from
Somali microfinance institution. In fact, looking into changes on only client households by
simply asking about changes they observed in their live would be misleading, that is response
bias would have been expected; so,as such types of problem would create serious problems when
using these kinds of impact evaluation exercise. Hence, the study has applied a propensity score
matching technique, which is capable of extracting comparable pair of treatment-comparison
households.
As expected, participation in microfinance was determined by a combination of factors such as
preference for group lending, age of the household head, marital states of the household
experience of the household in loan use, average household income and adequacy of loan
repayment had positive significant effect on the access to microfinance. That is a unit increment
of these factors lead a loge odds increment of access to microfinance. Therefore, to improve the
microfinance accessibility the institution has to work in these areas.
Page 73
72
For the findings of a reliable estimate of the impact of microfinance institution on building the
resilience of female headed households, thus necessitates controlling for all such factors
adequately. In doing so, propensity score matching has resulted in 75 clients to be matched with
75 non-clients households. In other words, a matched comparison of access to microfinance was
performed on these households who took loan from SMFI. The resulting matches passed a
variety of matching quality tests and were fit for answering the study‟s main objectives.
Supporting the participants with MF programs is thought to enhance the resilience of female
headed households. From the finding, the MF clients fill the poverty gap and improve the life
style of the households at all. There is a good improvement recently in addressing the targeted
households with other none-client households. In fact the program has increased the average
monthly income of the participating households. Therefore, the Somali Microfinance Institution
plays a significant role for the resilience of female headed household.
After controlling for other characteristics, it has been found that the old age household-heads are
more responsible to increase access of microfinance to improve their family states in the study
area. Therefore, age is a very significant factor to the management of the household income and
improves the family life style. Contrarily, more educated households tended to receive smaller
gains of income from microfinance institution. That is educated persons are less likely to take a
risk than non educated households. Therefore, they are less beneficial to gain from MFIs.
Page 74
73
5.2. Recommendations
The current study was based on small sample size taken from Jigjiga City of Ethiopian
Somali region. Therefore, the results cannot be generalized to other districts of the region
especially in the analytical terms. Further research done on a bigger scale with large sample size
could shed light on how microfinance activities affect building the resilience of female headed
households of Somali region in particular and Ethiopia in general. Another area that has not been
investigated is the difficulties that the microfinance institutions face. These areas deserve to be
studied by future researchers in the field.
In fact to improve the society culture of accessing MFI, the institution should focus on the listed
factors.
All factors are important, but in particular the program should focuses on the loan repayment
period. It should be improved, because as the result elaborated one of the risk averters fear is
short duration of repayment.
The estimated results revealed that households who are old aged and less educated families were
more likely to gain more from the MF. So, to elongate the target and to include all the society
primarily awareness creation should be important.
Somali Microfinance Institution beneficiary female household headed participants build more
resilience than none beneficial. This indicated that the MFI play a great role for the building of
better life. Therefore, taking this consideration into account, the area should increase slimily
MFIs, to do so, they should appreciate investors, non government organization and governmental
bodies participate in loan provision.
Reducing the deposit money is also recommended. This is also one of the problems raised by the
respondents, and to solve this problem to stop or reduce the deposit amount percentage, and
taking only the collateral as asset from the borrowers. Moreover, this study recommends
microfinance non-users should be addressed and motivated to participate in microfinance loan to
increase their resilience by increasing their income.
Page 75
74
REFERENCES
AbdikadirHasan (2016).The impact of microfinance institution of the performance of micro and
small scale enterprises in Jigjiga city.Msc thesis, Ethiopian civil service university,
Addis ababa
AlemayehuYirsaw(2008). The Performance of Micro Finance Institutions in Ethiopia: A Case of
Six Microfinance Institutions. MSc. Thesis in Accounting and Finance, Addis Ababa
University. Addis Ababa.
Alexander (2011) the impact of microfinance on the livelihoods of women in rural communities:
a case study of Jaman south district, Ghana.A thesis submitted to the institute of
distance learning, Kwame nkrumah university of science and technology in partial
fulfillment of the requirements for the degree of Commonwealth executive masters in
business administration. Jaman, Ghana.
Anwar Ahmed Sadik, 2015. Impact Assessment of PSNP Program on Household Food Security: The
Case of FadisWoreda, East Hararghe Zone of Oromia Regional State, Ethiopia. A Thesis
Submitted to College of Agriculture, Department of Rural Development and Agricultural
Extension, School of Graduate Studies, Haramaya University, Haromaya, Ethiopia.
Arbon, 2013. Developing a Process and Tools for Building Resilience in Vulnerable Households,
Torrens Resilience Institute
AshwillMaximillian, et al 2011. Gender Dynamics and Climate Change in Rural
Bolivia.Washington, DC: The World Bank.
Beatriz Armendáriz and Jonathan Morduch. 2010. The Economics of Microfinance. Second
Edition,The MIT Press Cambridge, Massachusetts London, England. pp16.
BereketGebremedhin, 2008. The Impact of Microfinance Services on the Living Conditions of
Households with Low Income Earning: The Case of Addis Ababa.A Thesis Submitted to
the School of Graduate Studies of Addis Ababa University in Partial Fulfillment of the
requirements for the Degree of Master of Science in Economics, Addis Ababa, Ethiopia.
Bigsten,et al(2003).Credit constraints in manufacturing enterprises in Africa. Journal of African
Economics, 12(1): 104-125.
Page 76
75
Buvinic, M., & Gupta, G. R. (1997). Female-headed households and female-maintained families:
Are they worth targeting to reduce poverty in developing countries? Economic
Development and Cultural Change, 45(2), 259-280. doi:
http://www.jstor.org/stable/1154535?origin=JSTOR-pdf Retrieved on 4th March 2014.
Cagatay, N., (1998), “Gender and Poverty”, Social Development and Poverty Elimination
Division, Working Paper Series 5, UNDP.
Chant , Sylvia H. (2007). Children In female-headed households: Interrogating the
Concept of an „inter- generational transmission of disadvantage‟ with particular Reference to the
Gambia, Philippines and Costa Rica, London School of Economics, Gender Institute.
Chant, Sylvia H. (2003).Female household headship and the feminisation of poverty: Facts,
fictions and forward strategies from London School of
Economics.http://eprints.lse.ac.uk/574/1/femaleHouseholdHeadship.pdf Retrieved on 6th
March 2014.
Cheston, Suzy, & Kuhn, Lisa. (2002). Empowering Women ThroughMicrofinance.Published by
UNIFEM for Microcredit Summit 2002. Retrieved 28 February 2004
from….http://www.icrocreditsummit.org/papers/empowering_final.doc.
Coleman, B. (1999). The impact of group lending in northeast Thailand Journal of Development
Economics, 45, 105-141.
Cutter, S. L., Burton, C. &Emrich, C., 2010. Disaster Resilience Indicators for Benchmarking
Baseline Conditions.Journal of Homeland Security and Emergency Management.
Elsabeth L. and Little.F. 2003. Is microfinance an effective strategy to reach the Millennium
Development goals?
Ernest, A., 2008. From informal finance to formal finance in Sub-Saharan Africa: institute of
statistical, social and economic research university of Ghana paper presented at the
high-level seminar on African finance for the 21st century organized by the IMF institute
and the joint Africa institute Tunis. Tunisia.
Hossian.F, 2002. Small loans, Big claims. Jstore Foreign policy.132. Garnenege Endowment for
international peace.
Hoque, M. &Itohara ,Y., 2009. Participation and decision making role of rural women in
economic activities: A comparative study for members and non-members of the micro-
credit organizations in Bangladesh Journal of Social Sciences 4 (3): 229-236.
Page 77
76
Khan et al., 2007, Impact of Microfinance on Living Standards, Empowerment and Poverty
Alleviation of Poor People, M.A thesis, Umeå School of Business, Bangladesh.22 pp
Klein, r. j. t., Nicholls, R. J. &Thomalla, F. 2003.Resilience to natural hazards:How useful is this
concept? Global Environmental Change Part B: Environmental Hazards, 5, 35-45.
Kurmanalieva, E. Montgomery, H. and Weiss, J. (2003). “Microfinance and Poverty Reduction
in Asia: What is the Evidence?”. A paper presented for the 2003 Asian Development
Bank Institute Annual Conference on Microfinance and Poverty Reduction, Tokyo
December 5th 2003
Ledgerwood, J.(1999). “Microfinance Handbook: An Institutional and Financial Perspective,
Sustainable Banking with the Poor.”Washington, D. C., World Bank.
Lindvert, M.(2006). “Sustainable Development Work and Micro Finance: A Case Study of how
ECLOF Ghana is working towards Financial Sustainability.” Thesis submitted to the
Department of Social Sciences, Mid Sweden University.
Littlefield, E., Morduch, J., Hashemi, S., 2003. Is Microfinance an Effective Strategy to Reach
the Millennium Development Goals? Focus Note No. 24, CGAP, Washington, USA.
Lokshin, M., Harris, K. M., &Popkin, B. M. (2000).Single mothers in Russia: Household
strategies for coping with poverty. World Development, 28(12), 2183-2219.
Mayoux, Linda. (2002). Women‟s Empowerment Versus Sustainability?Towards a New
Paradigm in Microfinance Programs.In Lemire, Beverly, Pearson, Ruth, & Campbell,
Gail. (eds.). Women and Credit:Researching the Past, Refiguring the Future. New
York:Berg, 245-269.
Mayunga, J. S. 2009. Measuring the measure: A multi-dimensional scale model to measure
community disaster resilience in the U.S. Gulf Coast region.Ph.D. 3370760, Texas A&M
University.
Meehan F. (1999).“Micro-Credit: Sound Business or Development Instrument.”Voorburg:
Netherlands.
MeronAssefa (2003). Female headed households and poverty in urban Ethiopia, M.A thesis,
addisababa university, Ethiopia, 3 pp.
Mullings, L (1995). Households headed by women: The politics of race, class, and gender. In
Faye D Ginsburg &Rayna Rapp (Eds.), Conceiving The New World Order: The Global
Page 78
77
Politics Of Reproduction, (pp. 122-139). United States of America: University of
California Press.
Murray. U and Boros. R. (2002), “A Guide to Gender Sensitive Microfinance”, The Socio-
Economic and Gender Analysis (SEAGA) Program, FAO, pp. 10-11).Ostrom, E. (1998):
A behavioral approach to the rational choice theory of collective action. In: American
Political Science Review. vol. 92, pp. 1-22.
Quisumbing, A., Haddad, L., and Pena, C., (2001), “Are Women Over Represented among the
Poor? An Analysis of Poverty in Ten Developing Countries”, IFPRI: FCND Discussion
Paper No. 115.
Robinson, M. S. (1998). „Microfinance: the Paradigm Shift From credit Delivery to Sustainable
Financial Intermediation‟.12 pp
Rutherford, S(1999) The poor and their money: an essay about financial services for poor
people, Manchester, UK: Institute of Development Policy Management .Available at
[http://www.microfinancegateway.org/content/article/detail/1737]
Ruwanpura, K. (2003). The survival strategies of Sinhala female-heads in conflict affected
eastern Sri Lanka.(No. 360913).International Labor Organization.
http://www.ilo.org/public/libdoc/ilo/2003/103B09_91_engl.pdf Retrieved on 1st January 2015.
Sara adugna , 2014. Determinants of microfinance institutions loan portfolio quality: empirical
evidence from Ethiopia, M.A thesis, Addis Ababa University, Ethiopia
Schlaufer C.,2008. Accessing financial services in rural areas, Overview formal and informal
financial services challenges of rural finance.InfoResources Focus No 2/08. Bern
University of Applied Sciences. Overview of financial services for rural areas, 8(1): 2-
13.
Simanowit.A and Walter. A, 2000. Ensuring impact: Reaching the poorest while building
financially self-sufficient institutions and showing improvement in the lives of the
poorest women and their families. In (ed) Horris and Raley. 2000. Path ways out of
poverty: innovation in microfinance for poorest families. Kumawa press, inc.
Somali microfinance institution (2016) the Annual achievement report of the 2015/2016 plan of
Somali Microfinance institution.
Sumner, A.(2007). “Meaning versus measurement: why do ´economic ` indicators of poverty
still predominate?” Development in Practice, vol. 17, no. 1, pp. 4-13.
Page 79
78
Syeda, R. S, A. H. Amara, & T. B. Aqsa, 2008. Determinants of credit programme participation
and socioeconomic characteristics of beneficiaries: Evidence from Sargodha.
The ILO Thesaurus.(2005). Retrieved September 16, 2009, from International Labour
organization (ILO) Web Site: http://www.ilo.org/public/libdoc/ILO-
Thesaurus/english/tr3578.htmEconomic Development and Cultural Change, 45(2), 259-
280.doi: http://www.jstor.org/stable/1154535?origin=JSTOR-pdf Retrieved on 4th
March 2014.
Tsehay T. and Mengistu B.(2002). “The Impact of Micro finance Service among poor women in
Ethiopia.”Occasional paper No. 6, AEMFI, Addis Ababa.
UBOS. (2014). National Population and Census Housing 2014: Provisional Results. Uganda:
United Nations Population Fund.
United Nations. 1995. The Copenhagen Declaration and Programme of Action: World Summit
for Social Development. New York: United Nations.
United Nation Population Funds (UNFPA), (2002), “State of World Population 2002 People
Poverty and Possibilities: Making Development Work for the Poor”.
USAID, 2012.Building Resilience to Recurrent Crisis, Available at:
http://transition.usaid.gov/resilience/USAIDResiliencePolicyGuidanceDocument.pdf
UNISDR 2005.Hyogo Framework for Action 2005-2015: Building the Resilience of Nations and
Communities to Disasters. Available at:
http://www.unisdr.org/2005/wcdr/intergover/official-doc/L-docs/Hyogo-framework-for-
action-english.pdf
WoldayAmha and AntenehKifle (2013).AEMFI Performance Analysis Report Bulletin AEMFI.
Addis Ababa,
Wolday, 2000. “Networking Microfinance Activities in Ethiopia: Challenges and Prospects.”
AEMFI Proceeding of the Conference on Microfinance Development in Ethiopia:
Bahirdar.
Wolday, 2007. Empowering women thorough microfinance review of the experience of Ethiopia
deposit taking microfinance institutions. Vol.No.1. AEMFI, Addis Ababa.
World Bank 2011.Gender & Climate Change: Three Things You Should Know. Washington, DC
PolicyBrief:http://siteresources.worldbank.org/EXTSOCIALDEVELOPMENT/Resourc
Page 80
79
es/244362-1232059926563/5747581-1239131985528/5999762-1321989469080/Gender-
Climate-Change.pdf
World Bank, 2016. Available at http://www.worldbank.org/en/country/ethiopia/overview
(Accessed on 20/1/2017)
YilkalWassieAyen. 2016. Impact of micro-credit programs on female headed households in
jimma zone, Ethiopia,International Journal of Scientific and Research Publications,
Volume 6, Issue 1
YirgaWorkneh, 2012. Determinants of Access to Credit by Smallholder Farmers: The Case of
Gozamen District, East Gojjam Zone, Ethiopia. A Thesis Submitted to Collage of
Agriculture and Environmental ScienceDepartment of Rural Development and
Agricultural ExtensionSchool of Graduate StudiesHaramaya University.Haramaya,
Ethiopia.
Yunus, M. &Jolis, A. 2003, Banker to the poor: Micro-lending and the battle against world,
Public Affairs (Perseus), New York.
Yohannes.(2007). “Regulating Microfinance in Ethiopia: Making it more ive.”Essays on
Regulation and Supervision. Microfinance Regulation and Supervision Resource
Centre.UNDP (2005) The Microfinance Sector in Central Africa.
Zahrani (2011).Empowering of female headed households, case study “sedighin” charity
institution in iran
Wright, G. A. N., 2000. Microfinance Systems: Designing Quality Financial Services for the
Poor. Zed Books Ltd. London & New York, The University Press Limited, Dhaka.
Page 81
80
Appendix I
Questionnaire
Field Questionnaire for the interview.
QUESTIONNAIRE
My name is Abdurahman Kedir Ali, I am studying Masters Program at Institute of Disaster Risk
Management and Food Security Studies, Bahir dar University, Ethiopia. I have designed the
following questionnaire for the study of the impact of Somali microfinance institution on
building the resilience of female headed households in jigjiga city, fafan zone, Somali
regional state-Ethiopia, which required for my thesis work as an integral part of my study.
I would highly appreciate if you fill this questionnaire. It will take approximately 15-20 minutes.
I expect your kind cooperation in this respect. All information provided in this study will be
treated as confidential and your anonymity is assured.
Demographic, socio-economic, psychological and institutional factories of the respondents.
Demographic factors
1. Sex of the respondent
Male Female
2. Age of the respondent…………………
3. Religion
Christian Muslim Other, please specify …………………………..
4. Ethnic Group
Somali Amhara Oromo Other, please specify……………………
5. Marital status of HH head
Married Single Widowed Divorced
6. How many family members do you have? ……………………………
7. Do you have any educational achievements?
Not at all Primary Intermediate Secondary University
Socio- Economic Characteristics of Sample Household
Page 82
81
8. When did your business begin operations? Please specify………
9. How did you know about the Somali microfinance institution?
Friend Relative Media advertisement Proximity to home/ business
Other, please specify…………
10. What is the source of your initial capital?
Personal Savings Friends and relatives Loan from MFIs Others, please
specify………………………………………….
11. Experience of loan use from MFIs (Access to credit)
11.1. had you ever use credit for the last consecutive years: I had used but I left now I had
not used I have used still now.
11.2. Were you demanding for credit in the last 12 months? . Yes .No
12. What was the purpose of the loan?
To start business to expand existing business To pay family expenses Other, please
specify…………………………
13. What is your current business type?
Commerce agriculture manufacturing others, please specify……………
14. How much is your monthly income in average?.........................................
15. Were you consulted in making household decisions before joining the MFI?
Yes No
16. Are you consulted in making household decision after joining the MFI?
Yes No
17. How many times per day do you consume?
Once twice three times other, please specify.........................
Page 83
82
18. Would you tell me the monthly estimated expenditure of your household on each of the
following items?
Expenditure categories Estimated monthly expenditure( in birr)
Food
House rent
education
health
clothing
Entertainment
Communication
Others, please specify
total
Psychological factors
19. Risk taking ability of the households
19.1. In your view, is borrowing from financial sources risky? Yes No
19.2. Did you give-up to take loans from lending organization due to fear of risk?
Yes. No
Institutional factors
20. Adequacy of loan repayment period
20.1. Was the loan disbursement time by lending institutions appropriate to perform your
activity? Yes No
20.2. If no, indicate the appropriate duration? ____________________________
20.3. Is loan repayment period of lending institution adequate? Yes No
20.4. If you say no, how much month/year enough for loan repaying _______?
20.5. Did you re-pay your loan on time? Yes No
12.6. If no, what is/are the reason/s for not re-paying on time? _____________
Page 84
83
21. How long have you been a member of SMFI?
less than one year one year two years three years 4 years and above
22. How long did it take for you to receive your first loan from MFI?
1month 2months 3months 4 months other, please specify…………………
23. Which of the following basic requirements did you have to satisfy before the loan was
given to you?
Physical collateral group collateral savings guarantors other, please
specify………………………….
24. Preference for group lending
24.1. Is the group lending preferable to you? Yes No
24.2. If you say yes why? _________________________________________________
_____________________________________________________________________
24.3. If you say no why? __________________________________________________
____________________________________________________________________
24.4. How do you get loan? In group or in individual
24.5. If you get group, who form the group? _________________________
25. What are the major challenges you face in accessing loans from MFIs?
The collateral repayment time is short the amount is small interest/profit is high
other, please specify………………………………………………………….
Thank you very much for your cooperation!