FACTORS AFFECTING LOAN REPAYMENT PERFORMANCES:

A CASE STUDY IN DEVELOPMENT BANK OF ETHIOPIA,

JIMMA DISTRICT

A THESIS SUBMITTED TO THE SCHOOL OF GRADUATE STUDIES OF

JIMMA UNIVERSITY FOR A PARTIAL FULFILLMENT OF REQUIREMENTS
FOR AWARD OF MASTER DEGREE IN BUSINESS ADMINISTRATION.

BY:

NAOL SHIFERAW GERBA

JIMMA UNIVERSITY

COLLEGE OF BUSINESS & ECONOMICS

MBA PROGRAM
OCTOBER, 2017

JIMMA, ETHIOPIA

i
FACTORS AFFECTING LOAN REPAYMENT PERFORMANCES:

A CASE STUDY IN DEVELOPMENT BANK OF ETHIOPIA,

JIMMA DISTRICT

A THESIS SUBMITTED TO THE SCHOOL OF GRADUATE STUDIES OF

JIMMA UNIVERSITY FOR A PARTIAL FULFILLMENT OF REQUIREMENTS
FOR AWARD OF MASTER DEGREE IN BUSINESS ADMINISTRATION.

BY:

NAOL SHIFERAW GERBA

ADVISOR/UNDER THE GUIDENCE OF:

MR. TADELE TESFAYE (ASSISTANT PROFESSOR)

AND

CO-ADVISOR:- AMINA AHMED (M.SC.)

JIMMA UNIVERSITY
COLLEGE OF BUSINESS & ECONOMICS
MBA PROGRAM

OCTOBER 13, 2017

JIMMA, ETHIOPIA

ii
STATEMENT OF CERTIFICATE

i
ii
iii
Abstract
This study was conducted in Development Bank of Ethiopia Jimma District geographical area.
Development bank of Ethiopia is state owned and specialized financial institution with the mandate
of providing long, medium and short term loans to feasible and viable projects of commercial
agriculture, agro processing and manufacturing sectors following government priority area. This
study is conducted on the factors affecting loan repayment; a Case study of Development Bank of
Ethiopia, Jimma District. Accordingly, endeavors are made to contribute to the empirical gab
regarding factors affecting loan repayment performances specifically in Jimma District. The main
objective of this study was to identify and analyze the major factors of loan repayment performances
in DBE, Jimma District; more specifically from four different perspectives, borrowers related
factors, bank related factors, business/project related factors and factors related to external
environments. Both primary and secondary data were used in the study. The primary data was
collected from 150 selected borrowers through questionnaires and pre-tested structured interview
with staffs and bank managers. To define and select the population of the study, stratified random
sampling was used where borrowers were stratified based on their loan status. Both descriptive
statistics and econometric analyses particularly logistic regression (binary logit) were employed to
present the results and findings of the research. The study was basically conducted from four broad
perspectives; factors related to characteristics of borrowers, factors in the side of lender institution,
factors related to business/project and the other external factors were analyzed through descriptive
statistics such as frequencies, percentages, mean, and standard deviation. A total of twenty one
explanatory variables were included in the logistic regression and out of these eight were found to be
statistically significant to influence the dependent variables. The results of binary logistic regression
revealed that Educational qualification of borrowers, family size of borrowers, credit experience,
having other business, proper follow up, duration of service time/time horizon, loan size and loan
diversion were found significant and influenced loan repayment performances. Based on the
descriptive and econometric results/analysis, the researcher has recommended to the bank to
undertake proper screening, disbursing loan at the right time, conduct proper follow-up, provide
sufficient amount of loan as per the feasibility study of the project, solve other difficulties as
identified in this study and work on all other factors affecting loan repayment performances.

Key words: Bank, borrower, Loan Repayment.

iv
ACKNOWLEDGEMENTS

On top of all, I would like to thank the Almighty God, for the charity, kindness, help and forgiveness,

and caring; May your name be praised and blessed forever on me.

My sincere and deepest gratitude goes to my advisor Mr. Tadele Tesfay (Assistant professor) and co-

advisor Mrs. Amina Ahmed for friendly approach, unreserved assistance in giving me relevant

comments and guidance throughout the study.

I would like to offer my special thanks to my Father Shiferaw Gerba who always standby me and

supports who paid everything to give me everything. Thank you dad!!

I also want to extend my sincere gratitude for my fiancée and my brother Tolawak Shiferaw. You

guys were always there for me. You win my heart, thank you!

Family deserves everything. My beloved family, words cannot describe how wonderful I feel to have

you as a family. You have been doing all what you could do. I am always begging my God to pay

back all the favors you did for me.

Finally yet importantly, I would like to thank all friends, colleagues and all who helped me either

directly or indirectly. Thank you!

Naol Shiferaw
[email protected]

v
TABLE OF CONTENTS
Contents pages

ACKNOWLEDGEMENTS .................................................................................................................................. v
TABLE OF CONTENTS..................................................................................................................................... vi
LIST OF TABLES ............................................................................................................................................... ix
LIST OF FIGURES ............................................................................................................................................. ix
ACRONYMS/ABRIVATIONS: .......................................................................................................................... x
CHAPTER ONE ...................................................................................................................................................1
INTRODUCTION ................................................................................................................................................1
1.1. Background of the Study ..........................................................................................................................1
1.2 Statements of the Problems .........................................................................................................................3
1.3. Objectives of the Study ..............................................................................................................................6
1.3.1. General Objective ...............................................................................................................................6
1.3.1.1. Specific Objective ............................................................................................................................6
1.4 Research Hypothesis ...................................................................................................................................6
1.5 Significance of the Study ............................................................................................................................8
1.6 Scope of the Study ......................................................................................................................................8
1.7 Limitation of the Study ...............................................................................................................................9
1.8 Organization of the Study ...........................................................................................................................9
CHAPTER TWO: ...............................................................................................................................................10
REVIEW OF RELATED LITERATURE ..........................................................................................................10
2.1. Theoretical Review .............................................................................................................................10
2.1.1. Banks and Its Importance............................................................................................................10
2.1.2. Development Banks ....................................................................................................................11
2.1.3. The Difference between a Development Bank and Commercial Banks .....................................11
2.1.4. Basic Requirements to Access Credit .........................................................................................14
2.1.5. Performing Loans .......................................................................................................................15
2.1.6. Nonperforming Loans: Meaning and Nature ..............................................................................16
2.1.7. Loan classifications in Ethiopia ..................................................................................................19
2.2. Development Bank of Ethiopia ...........................................................................................................20
2.2.1. The Main Functional Areas of the Bank .....................................................................................21

vi
 2.3. Empirical Literature ............................................................................................................................25
2.4. Empirical Studies in Ethiopia .............................................................................................................30
CHAPTER THREE: ...........................................................................................................................................34
RESEARCH METHODOLOGIES ....................................................................................................................34
3.1. Research Design .................................................................................................................................34
3.2. Description of the Study Area.............................................................................................................34
3.3. Data Type and Source .........................................................................................................................35
3.4. Method of Data Collection..................................................................................................................36
3.5. Population and Sampling Method.......................................................................................................36
3.6. Method of Data Analysis ....................................................................................................................37
3.7. Model Specification ............................................................................................................................38
3.8. Variables of the Study.........................................................................................................................40
3.8.1. The Dependent Variable .............................................................................................................41
Dependent variable defined ........................................................................................................................41
3.8.2. Definition and Hypothesis on Independent Variables ................................................................42
Explanatory variables encoded ...................................................................................................................47
3.9. Conceptual Framework .......................................................................................................................50
CHAPTER FOUR...............................................................................................................................................52
RESULTS AND DISCUSSIONS .......................................................................................................................52
Introduction.....................................................................................................................................................52
4.1. Background Information of Respondents: ..........................................................................................52
4.2. Descriptive Analysis ...........................................................................................................................53
4.2.1. Borrowers Related Factors ..........................................................................................................54
4.2.2. Business Related Factors ............................................................................................................59
4.2.3. Institutional Related Factors .......................................................................................................63
4.2.4. External Related Factors .............................................................................................................69
4.2.5. Other Major Problems.................................................................................................................70
4.3. Econometric Analysis ..........................................................................................................................71
4.3.1. Model Tests.................................................................................................................................71
4.3.2. Results of Regression Analysis ...................................................................................................75
4.3.3. Discussions on Regression Results .............................................................................................78

vii
CHAPTER FIVE ................................................................................................................................................83
CONCLUSION AND RECOMMENDATION ..................................................................................................83
5.1. Conclusion ..........................................................................................................................................83
5.2. Recommendation ................................................................................................................................86
REFERENCE .....................................................................................................................................................88
APPENDIX .........................................................................................................................................................92

viii
LIST OF TABLES
Table 3. 1summary of Expected Sign (+/-) of Explanatory Variables in this Study ..................... 48

Table 4. 1 Questionnaire and interview success rate............................................................................ 53

Table 4. 2 Gender in repayment performances. ................................................................................... 54
Table 4. 3 Age and marital status of Borrowers and Loan Repayment ................................................ 56
Table 4. 4 Educational Qualification and family size of Borrowers and Loan Repayment ................. 58
Table 4. 5 Business sector and business form against repayment ........................................................ 60
Table 4. 6 other business and income vs loan repayment performances ............................................. 62
Table 4. 7 loan size and time horizon against repayment .................................................................... 64
Table 4. 8 collateral, equity and diversion against repayments. ........................................................... 66
Table 4. 9 interest rate, grace period, follow up and KYC with repayment ......................................... 67
Table 4. 10 whether condition and market condition against repayment. ........................................... 69
Table 4. 11 VIF of the Continuous Explanatory Variables used in the study ...................................... 72
Table 4. 12 Contingency Coefficients for Dummy Variables .............................................................. 73
Table 4. 13 Results of Binary Logistic regression, loan repayment performances. ............................. 76
Table 4. 14 Odds ratio of binary logistic regression, loan repayment performances. .......................... 77

LIST OF FIGURES
Figure 2. 1Loan process of DBE ......................................................................................................................23
Figure 2. 2 Conceptual framework ..................................................................................................................51

ix
ACRONYMS/ABRIVATIONS:
ADLI……………………………………………Agricultural Development Led to Industry
BSC……………………………………………..Business Scored Cared
CBE……………………………………………..Commercial Bank of Ethiopia
DBE……………………………………………..Development Bank of Ethiopia
ECG …………………………………………… Export Credit Guarantee
GDP …………………………………………… Growth Domestic Production
GTP …………………………………………….Growth and Transformation Plan
KYC……………………………………………..Know Your Customer
LAT……………………………………………...Loan Approval Team
MFI……………………………………………...Micro-finance Institution
MOFED…………………………………………Ministry of Finance and Economy Development
NBE……………………………………………..National Bank of Ethiopia
NPL……………………………………………...Non Performing Loan
PCFR…………………………………………….Project Completion Follow up Report
PIFR……………………………………………...Project Implementation Follow up Report
POFR……………………………………………..Project Operational Follow up Report
PRLR……………………………………………..Project Rehabilitation and Loan Recovery
RUFIP ………………………………………….. Rural Financial Intermediation Program
SME …………………………………………......Small and medium Enterprises
SWOT…………………………………………….Strength, Weakness, Opportunity and Threat
RRR…………………………………………….....Relative Risk Reference
VIF…………………………………………….…..Variance Inflation Factor
WTO………………………………………………......World Trade Organization

x
 CHAPTER ONE

INTRODUCTION

1.1. Background of the Study

Over the past two decades, the Ethiopian economy has gone through numerous changes; it substantially
outperformed the average of Sub-Saharan African countries. The Government of Ethiopia adopted market
oriented economic policy, made agriculture its primary priority in 1991, and implemented Agricultural
Development Led-Industrialization (ADLI) strategy. Following the change of the government by 1991 the
country introduced major economic reforms by accepting capitalist ideology contrary to the previous
communist set up in the economy of the country by 1992(MoFD, 2015)

Since then, the Ethiopian economy has gone through remarkable economic growth in all agriculture,
service and industrial sector according to the World Bank report of 2016(World Bank, 2016). Although
initially led by agriculture, the growth base is broadening, with increasing contributions to GDP from
services and industry year after years. In the same token the banking sector reveals dramatic progresses and
expansions in the past twenty years. Banks play a very important role in the economic development of
every nation. They have control over a large part of the supply of money circulation. Banks are the main
stimulus of the economic progress of a country. The financial sectors contribution to growth lies in the
central role it plays in mobilizing savings and allocating these resources efficiently to the most productive
uses and investments in the sector Tihitina, 2009)

The Ethiopian financial institutions have a long time history. The use of money and coins in Ethiopia has a
long history, and the introduction of modern banking is nearly a century old. The original bank of
Abyssinia started operation in February 1905 and its activities included keeping government accounts and
financing exports. Despite the long history, which precedes the advent of modern banking throughout most
of Africa, the Ethiopian financial sector has not progressed as it beginning. In the period of a shift from a
mixed to a state managed economy, the development of the financial sector was stunted. Although the
financial sector of Ethiopia has grown in the 1990’s, compared to its state during the preceding decades, it
is still in its infancy.

1
In Ethiopia, there are two government owned banks and sixteen private banks, seventeen insurance
companies (1 public and 16 private), thirty five (35) Micro finance institutions and five Capital Goods
Finance Companies in 2015/16. Commercial Bank of Ethiopia (CBE) is one of the dominating state-owned
banks whose assets represent about 70% of the sector and Development Bank of Ethiopia (DBE) is the
only development bank having a second place market position in the country (National bank report, 2016).

Development bank of Ethiopia (DBE) is the only bank of its kind in Ethiopia. It’s different from other
commercial banks in its nature and objectives endowed to it. Development Bank of Ethiopia is a
specialized financial institution established to finance and provide close technical support to viable projects
from the priority areas by mobilizing fund from domestic and foreign sources while ensuring its
sustainability. The Bank extends investment credit to creditworthy borrowers and projects that have
received a thorough appraisal and found to be financially and economical viable and socially desirable. In
addition to project financing and rendering technical support to the selected priority area sectors, DBE has
given great task in financing the Small and Medium Enterprises through Lease Financing program to
enable them to acquire capital goods and machineries (DBE annual report, 2015).

In 2015/16 fiscal year the Bank has set a target of approving, disbursing and collecting Birr 14.82 Billion,
Birr 13.54 Billion and Birr 5.78 Billion respectively. With regard to the actual performance of the year; the
approval, disbursement, and collection of Birr 11.8 Billion (80%), Birr 6.3 Billion (47%) and Birr 4.1
Billion (71%) were registered, respectively (DBE, annual report 2016)

The Non-performing loan size and ratio of the Bank in the year 2015/16 was Birr 5.6 Billion (17.71%)
which is 53% of the planned target of Birr 3.6 Billion (9.45%) at the corporate level. Compared to the
preceding year same period performance of 12.5%, it has increased by 44%.

The repayment performances of DBE Jimma district within the past three consecutive years showed that
the bank’s NPL is increasing and going against the plan to minimize the ratio into a single digit and
achieve a 100% performing loans by 2020. Even though, NPL ratio of the bank showed little improvement
from its historical performance at corporate level the NPL ratio of DBE in the year 2015/16 is 17.71% and
fresh entrants to NPLs has showed sharp increment during 2015/16 fiscal year. The 2013/14 annual report
of the bank indicates, DBE Jimma district NPL ratio is 14.35% while the corporate NPL ratio is 8.23%,
with the fresh entrant to NPL 2.58%. Similarly, the 2014/15 annual report of the bank revealed that the
NPL ratio of DBE is 12.54%, while NPL ratio in DBE Jimma district is 18.1%. The main reasons

2
contributing to such low performances, particularly in NPL and fresh entrants to NPL relates to different
factors. According to Nawai & Sharif (2013), Olomola (1998) and Micha'el (2006), Abraham (2002) and
kibrom (2004) and many other studies discussed in empirical studies in chapter two of this study, the main
causes of such low repayment and high NPL ratio performance would emanate from institutions related
factors, borrower related factors, business related factors and external factors. Hence, assessing and seeking
for solution of factors affecting loan repayment performances become imperative in providing credit
service for different governmental and non-governmental business.

1.2 Statements of the Problems

Development Bank of Ethiopia (DBE) is one of the major state owned institutions established to support
the economy development of the country through provision of project finance and technical support to
viable projects that are selected as priority areas by the government. As a policy Bank, it is entrusted to
serve as a tool for the country‘s development through availing medium and long term credit to agriculture,
industry, mining and energy and SMEs (DBE annual plan, 2016).

Development Bank of Ethiopia is well known and specialized in project financing. The Bank has been
offering medium and long term loans to different kinds of viable projects. Hence, it is known that the role
of Development Bank of Ethiopia is very important in the economy practically by financing government
development priority areas which are believed to be the engine of growth like Manufacturing Industry,
Agro-Processing, Commercial Agriculture, Mining & Energy and SMEs. In addition to project financing,
following the especial emphasis given by the government to Small and Medium Enterprises as they are
believed to be the foundation for the move to industrialization, recently DBE is entrusted to support SMEs
along with medium and large scale industrial projects.

Coming to the plan and performances of the bank, In 2015/16 fiscal year the Bank has set a target of
approving, disbursing and collecting Birr 14.82 Billion, Birr 13.54 Billion and Birr 5.78 Billion
respectively. Whereas the actual performance of the year; the approval, disbursement, and collection of
Birr 11.8 Billion (80%), Birr 6.3 Billion (47%) and Birr 4.1 Billion (71%) were registered, respectively. In
the geographical area where this research was conducted, Jimma District the case seems little different due
to the fact that Financing agricultural projects especially in Gambella District was suspended in the year
2015/16, following the reportedly land overlapping and other related problems in the area and the very

3
nature of projects in the District(which is primarily agriculture). Hence, the performance of the District was
highly affected not only in loan collection/repayment performances but in all aspects.

At corporate level the NPL ratio of the bank shows improvement from its historical performance which is
17.71% and fresh entrants to NPLs has showed increment during last 2015/16 fiscal year. At Jimma
District the NPL ratio increased from 12.5% in the year 2014/15 to 18% in 2015/16 and the new entrant
sharply increased from 1.65 to 2.5 in the same fiscal year.(Annual reports of DBE, 2014/15 and 2015/16
and national Bank of Ethiopia)

DBE has set a vision of having 100% successful projects by the end of the year 2020. The performance
reports of the bank, which includes the figures in the above paragraph from the year 2015/16 however
didn’t shows the same story, but even though the time table keeps running the NPL ratio was not reducing
from time to time as expected. Furthermore, the GTPI performance reports of the bank and the reports of
National Bank of Ethiopia and the supervising governmental agency, Public Financial Enterprises Agency
indicated that Development bank of Ethiopia was not achieving its targeted goal especially in approving,
disbursing and collecting loans as expected from the plan cascaded from GTPI. The performance in Jimma
District resembles the corporate performances of the bank and even worst in loan repayment performances.

In this study, focus is given to loan repayment performances which include both performances and
nonperformance of loans. The reasons and factors for performances of loans or increase in NPLs are related
to the cumulative effects of different factors. This is what necessitated and motivated the researcher to
focus in this area. The issue of Loan repayment performance and NPL has been a subject of major concern
for researchers for many years across the world and in recent years in Ethiopia.

In Ethiopia, there were researches conducted on the related topics by different researchers. For instance,
Wondimagn (2012) conducted a research titled ‘determinants of NPLs on commercial banks of Ethiopia’
and his study indicated that interest rate has no significant impact on the level of commercial banks loan
delinquencies in Ethiopia. On the other hand, Mitiku (2014) “Determinants of Commercial Banks
Lending” with the objective of assessing the relationship between commercial bank lending and its
determinants variables (bank size, credit risk, GDP, investment, deposit, interest rate, liquidity ratio and
cash required reserve) by taking financial statement of seven years and found that there was significant
relationship between loan size, credit risk, gross domestic product and liquidity ratio and commercial bank
lending. Kibrom T (2010) studied about determinants of successful loan repayment performances of private

4
borrowers in the North Region of Development Bank of Ethiopia by using binomial model and focused on
borrowers’ specific and business type specific in order to analyze successful loan repayments.

Firafis Haile(2015) conducted a study on related subject area under a title ‘determinants of loan repayment
performances; a case study of Harari Microfinance institutions’ mainly focusing on borrowers specific
factors using binary logit model. The result identified loan size, credit experience, training, business type
and family size were significantly affected the repayment performances. Similarly, Abraham G (2002)
conducted a study on DBE Batu branch on loan repayment and its determinants in small scale enterprises
financing in Ethiopia. The study mainly focuses on bank specific and borrowers’ specific factors.

Arega Seyoum et, al. (2016) studied about factors affecting non-performing loans in the DBE central
District by using descriptive statistics from bank specific factors and borrowers specific factors in order to
determine factors affecting non-performing loans in the DBE central District. The result of the study shows
that poor credit assessment and credit monitoring are the major causes for the occurrence of NPL in DBE.
Credit size (includes aggressive lending, compromised integrity in approval, rapid credit growth and bank’s
great risk appetite); high interest rate, poorly negotiated credit terms and lenient/lax credit terms, and
elongated process of loan approval were bank specific causes for the occurrence of nonperforming loans.

In all the above studies what affects loan repayment performances are evaluated from bank specific and
borrowers’ specific factors. In reality what affects loan repayment performances were not limited to bank
specific factors and borrowers specific factors but beyond these it includes bank specific, borrower
specific, business specific and other factors (macroeconomic factors). There were studies conducted on
Nonperforming loans in Development Bank of Ethiopia in other area of the bank like in North region or
Central region and Batu Branch but it’s not appropriate to generalize the findings of these studies
especially to Jimma District. Because, loans in the areas were mainly agricultural and the nature & types of
problems differs from the central and north Districts which are predominantly industry, service and agro
processing projects. In addition, there are internal and external changes since these researches were
conducted in the bank, including Changes of policies, organizational restructuring, change in interest rate
and the global climate changes are among the major occurrences.

Generally the researcher believes that the problems related to defining factors affecting loan repayment
performances were not properly addressed particularly in Development Bank of Ethiopia Jimma District
due to the fact that there were few empirical studies in this area and some previous studies conducted in

5
other areas of the bank were limited to bank specific and borrowers’ specific factors. Hence the researcher
is motivated to study and lay his own contribution on the factors affecting loan repayment performances in
a broader sense. Accordingly, endeavors are made to identify the major factors that contributed to loan
repayment performances specifically in DBE Jimma District from four broad perspectives, borrowers
related factors, bank specific factors, business characteristics and external factors.

1.3. Objectives of the Study

1.3.1. General Objective

The objective of this study is to identify Factors affecting loan repayment performance, identify the major
factors from four different perspectives particularly, from borrower side, from bank/lender side, from
business and from other external side of loans of Development Bank of Ethiopia Jimma District.

1.3.1.1. Specific Objective

To achieve the general objective, the following more specific objectives were identified under this study:

1. Identify the major borrowers specific factors (Education, Experience…..etc.) on loan repayment
performance of Development Bank of Ethiopia, Jimma District.

2. Identify the Bank specific factors (Loan size, Follow up, grace period, due diligence/KYC), Collateral
and equity) on loan repayment performance of Development Bank of Ethiopia, Jimma District.
3. Identify Business related factors (Business sector, business form…etc.) on loan repayment performance
of Development Bank of Ethiopia, Jimma District.
4. Identify other major factors (market and weather conditions) on loan repayment performance of
Development Bank of Ethiopia, Jimma District.

1.4 Research Hypothesis

To achieve the objective of this study the researcher would test the following hypotheses concerning the
factor affecting loan repayment performance of DBE, Jimma District. Empirical researches conducted in
the area found different results; for instance Kibirom (2010), in his study on determinants of successful
loan repayment Performance of private borrowers in Development bank of Ethiopia north region, identified
factors that determine loan repayment performance which includes; borrowers perceived need, that is
borrowers have to be given an opportunity to borrow for their perceived needs, competence, that is the

6
borrowers past personal and profit record, past prosperity etc. Based on this model, educational level of the
borrowers, repayment period, availability of other source of income, sector, purpose of the loan and type of
labor determine successful loan repayment performance of the borrowers positively and significantly.
Whereas, gender and family size have positive sign, but are not statistically significant. Moreover,
variables such as age, loan diversion, other source of credit show negative sign but not statistically
significant. The variable experience is statistically significant but show negative sign.

Awunyo-Vitor (2012) searched the determinants of loan repayment default among farmers in Brong Ahafo
District of Ghana. The study employed probit model to investigate factors that influence farmer’s loan
repayment default. Data used in this study was gathered through a survey of 374 farmers in five Districts
within Brong Ahafo District of Ghana. The results showed that farm size, and engagement in off farm
income generating activities reduces the likelihood of loan repayment default significantly. In addition,
larger loan size and longer repayment period as well as access to training are more likely to reduce loan
repayment default.

Abraham (2002) conducted a research with the aim of identifying the major factors behind the loan default
problem of small-scale enterprises with particular reference to Development Bank of Ethiopia (DBE), by
employing tobit model. Sample selection was based on stratified sampling and 102 borrowers were
selected. The result of econometric model revealed that having other source of income, education, work
experience in related economic activity before the loan and engaging on economic activities other than
agriculture are enhancing while loan diversion, being male borrower and giving extended loan repayment
period are undermining factors of the loan recovery performance of projects.

Firafis Haile(2015) conducted a study on related subject area under a title ‘determinants of loan repayment
performances; a case study of Harari Microfinance institutions’ mainly focusing on borrowers specific
factors using binary logit model. The result identified loan size, credit experience, training, business type
and family size were significantly affected the repayment performances.

Based on these and other empirical research findings the researcher wants to draw the following research
hypothesis;

H1: There is positive relationship between Borrowers’ specific factors (Education and Experience) and
loan repayment performance.

7
H2: There is positive relationship between Bank specific factors (Loan size, Follow up) and loan
repayment performance.

H3: There is negative relationship between Business specific/project related factors (business sector and
loan diversion) &loan repayment performance.

H4: There is positive relationship between macroeconomic factors) like market and weather conditions and
loan repayment performance.

1.5 Significance of the Study

This study and its finding is significant for many more reasons. The subject of the study remains problems
of every financial institution in our country this day. So, the findings of this research are expected to
contribute a lot for different stakeholders. The following are among the main significance of this study: it
benefits the researcher to obtain new knowledge about problems under the study and gives clear picture
about the issue of loan repayment performance, Present the current clear picture of NPLs in DBE Jimma
District and tries to show the significant factors (internal as well as external) that determine the repayment
performances, Use as starting point for other studies which may focus on similar topics and issues related
to factor affecting loan repayment performance in general and factors that influence the level of
nonperforming loan in baking industry in particular and also study will enable lenders of Development
bank of Ethiopia how to overcome potential factors that are highly affects the level of nonperforming loan
in the bank at general.

1.6 Scope of the Study

This study is conducted on Development bank of Ethiopia Jimma District under a title factors affecting
loan repayment performances. Hence the scope of the study is limited to the geographical limitation of
Jimma District which includes Jimma branch, Gambella branch, Mettu branch, Bedelle branch, Agaro
branch, Bonga branch, MizanTefari and Teppi branches. Jimma District is selected for geographical
proximity and accessibility for data collection. Among eight branches under Jimma District, only two of
them are graded as A branch (Jimma branch and Gambella branches) and its these two branches that are
empowered to handle active customers and provide loan for their customers for practical purposes. Hence,
the data used under this study is data from these two branches.

8
On the other hands, the subject matter of the study is limited to identifying major factors affecting loan
repayment performances in the Jimma District. This study mainly focuses on the issues that extracted in the
research objective and research hypothesis.

The other important issue is regarding kind and type of data used in the study. The study used both primary
data collected using questionnaire and interviews and secondary data from different source as defined in
methodology but such secondary data used in this study is limited to the past fiscal year, 2015/16, which is
one year only.

1.7 Limitation of the Study

In conducting this study, the researcher faced some challenges and shortages from methodological
limitations. The main problem is geographical limitation of the study in to Jimma district. The other
limitation related to the theoretical drawbacks emanated from the nature of the models used in this study
which is beyond the control of the researcher. The other limitation is regarding lists of independent
variables; the independent variables are not limited to those listed, discussed and presented in this work but
many more are not covered due to financial and time limitations.

1.8 Organization of the Study

This research report is organized in five chapters. Chapter one provides the general introduction about the
whole report. Chapter two presents the review of related literatures. Chapter three provide detail
description of the methodology employed by the researcher. Chapter four contains data analysis
presentation and interpretation. Finally, the last chapter concludes the total work of the research and gives
relevant recommendations based on the findings.

9
 CHAPTER TWO:

REVIEW OF RELATED LITERATURE

This chapter presents the theoretical foundation of the study critically with issues pertaining to factors
affecting the repayment performances of loans. A critical review of the existing literatures concerning the
repayment performances of borrowers and factors which influences the repayment capacity, and Various
literatures focusing both on non-performing loans and successfully performing loans are thoroughly
reviewed and presented in this part of the study. Accordingly, first there is the theoretical reviews on well
performing and non-performing loans which include nature and definition of NPLs, Classifications of
Loans and advances, theories on bank loan and cause for loan default and performing loans. The other
important part is reviewing different empirical results regarding the impact of various banks specific,
borrowers’ specific and macro level factors on the growth of nonperforming loan.

2.1. Theoretical Review

2.1.1. Banks and Its Importance

The term bank refers to an institution that deals with money and provides other financial services.
According to Heffernan (1996), banks are defined as intermediaries between depositors and borrowers in
an economy that are distinguished from other types of financial firms by deposit collection and offering
loan products. Banks role in the economy of any country is very significant. They play intermediation
function in that they collect money from those who have excess and lend it to others who need it for their
investment.

Banks mobilize deposits and allocate the mobilized money efficiently to the most productive uses of
investment in the real sector. Availing credit to borrowers is one means by which banks contribute to the
growth of economies. The banking sector makes a meaningful contribution to the economic growth of
every country. Banks contribution to the growth lies in the role they play in mobilizing deposits and
allocating the resources efficiently to the most productive uses investment in the real sector. So making
credit available to borrowers is one means by which banks contribute to the growth of economies. Banks
pool resources together for projects that are too large for individual shareholders to undertake (Bagehot,

10
1873). They are also considered the most important enabler of financial transactions in any country’s
economy and are the principal source of credit (Rose, 2002). Bank finance is the primary source of debt
funding. Commercial banks extend credit to different types of borrowers for many diverse purposes, either
for personal, business or corporate clients (Saunders & Cornett, 2003). Besides, banks are also the
custodians of nation’s money, which are accepted in the form of deposits and paid out on the client’s
instructions (Sinkey, 2002; Harris, 2003). Banks accept deposits, make loans, and derive a profit from the
difference in the interest rates paid and charged respectively. Some banks also have the power to create
money (Fasil and Merhatbeb, 2009).

Notwithstanding all other activities, banking industry considers lending as their most important function for
utilization of funds. Since the major portion of gross profit of the industry is earned from loans, the
administration of loan portfolios seriously affects the profitability of banks.

The most important financial institutions are commercial banks, mutual funds, security firms, insurance
companies, and pension funds.

2.1.2. Development Banks

A development bank is a ‘bank’ established for the purpose of ‘financing development’. A traditional
definition of a development bank is one which is a national or District financial institution designed to
provide medium-and long-term capital for productive investment, often accompanied by technical
assistance, in less-developed areas (Encyclopedia Britannica, 2003). Development Banks are financial
intermediation that provides financing to high priority investment projects in a developing economy. This
definition implies that the purpose of development banking is to bring the country to a higher level of
development. Development banks fill a gap left by undeveloped capital markets and the reluctance of
commercial banks to offer long-term financing.

2.1.3. The Difference between a Development Bank and Commercial Banks

There are several differentiating factors between a development bank and a commercial bank. Some
extreme observations below are made in order to emphasize “traditional” differences between the two in
order to emphasize the point. Actual practice, of course, differs from commercial bank to commercial bank
and from development bank to development bank. As the country’s capital markets develop, there shall be
less difference between these specialized institutions and the similarities shall become more apparent. With

11
this as a premise, the traditional differences between development and commercial banks are in the
following areas (compiled by Asian Development Bank, ADFIAP, 2007)

a. Impetus for the Creation of the Institution: A development bank is created as an instrument of
economic development while a commercial bank is created by business opportunities.
b. Posture Relative to Business Opportunities: A development bank is supposed to be pro-active as it
should take an active role to promote projects and to develop institutions (entrepreneurs). The projects
chosen are those that are consistent with the economic development priorities. A commercial bank is
known to be reactive to business opportunities. It requires bankability only after the entrepreneur’s
decision has been made; it waits for the idea to culminate into a funding requirement.
c. Types of Projects Supported: For a development bank, there is an explicit effort to support economic
development projects. The following desired ‘impact’ projects form the basis for scanning for
opportunities: import substitution (at competitive prices); exports; increased local demand; District
development (for example, tourism); and increased industrial efficiency through better technology. For
a commercial bank, the abovementioned goals are not the starting point for the identification of
projects. Rather, they would most likely be side-benefits. A commercial bank has little concern for
these objectives, except for the viability of the bank transaction. In short, a development bank’s
activities are project-based while that of the commercial bank are transaction-based.
d. Search Process for Projects Financed: A development bank goes into a planning cycle, identifying
which are the likely areas to go into. For example, if it determines that an export is an area to be
promoted, then it conducts a marketing study and seeks entrepreneurs to implement related projects.
For the commercial bank, the search process is different. It asks, “Are you an exporter?”, and then
looks at that entrepreneur’s cash balance to determine if there is a marketing opportunity for the
transaction.
e. Project Promotion Activities: A development bank offers counseling and advisory services for
enterprise development and promotion as part of its development lending process. A commercial bank
offers legal and business advice, appraisal services and credit investigation, usually for a fee. It
undertakes very little project promotion and institutional development. Its emphasis is on client
development and marketing.
f. Strategic Goals: A development bank has a more difficult strategic objective because it is involved with
the concerns of the country, specifically economic development. Aside from this, after providing
financing, it is also concerned with developing the enterprise. Developing them explicitly would mean

12
 additional costs to the bank. Enterprise development dramatically limits the number of accounts that a
development can handle because this is time-consuming. A commercial bank’s main concern is to
generate profits. Other benefits are only incidental. With a commercial bank’s cost-consciousness,
economic development would be its last priority.
g. Criteria for Financing: A development bank assumes project risks and does not insist on too much
collateral. It will provide financing as long as the other criteria are met. A commercial bank pays less
attention to the project in relation to the collateral requirements. However, the more progressive banks
are lending against project cash flow and without collateral.
h. Assessment of the Loan Proposal: A development bank employs project appraisal as a means to
determine the viability of the project submitted for financing.
Project appraisal looks at the technical, financial, marketing, management, environmental and
economic aspects of the project. Loan repayment is based on the cash flow to be generated by the
project. A commercial bank uses risk asset management as tool to assess the borrower. It looks at the
so-called 5 C’s of credit, i.e., character, capacity, capital, collateral and condition. It bases loan
repayment on the capacity of the borrower to pay (even from other sources) than from the ‘project’
itself. Thus, it can be said that development bank financing is project-focused while that of a
commercial bank is borrower-oriented.
i. Term of Loans Extended: A development bank provides mainly term loans (maturity of more than one
year). On the other hand, a commercial bank provides mainly short-term loans (less than one year
maturity).
j. Sources of Loan Funds: A development bank is dependent on concessionary, long-term funds, e.g.
pension funds, funds from multilateral financial institutions like the World Bank, Asian Development
Bank, etc. It has traditionally limited access to domestic or commercial funds. A commercial bank has a
strong deposit base and its corporate borrowers are also depositors. They can match its commercial
borrowing against its own short-term loans.
k. Lending Policies for Cyclical Industries: A development bank supports its clients in spite of short-term
cycles while a commercial bank does not like cyclical industries.
l. Resource Mobilization: A development bank undertakes project promotion work to match
concessionary long-term financing while a commercial bank mobilizes deposit funds from small
depositors which are lent out to large companies.

13
m. Client Relationship: A development bank relates more to clients as borrowers. There is less day-to-day
business relationship. Trade transactions of a commercial bank allow for frequent monitoring and close
client relationship.
n. Scope of Institutional Mandate: A development bank is essentially a specialized institution. It has
limited branching and range of products. The commercial bank has a generalized charter. It can offer a
wide range of products (especially in the case of universal banks) and can open more branches.

2.1.4. Basic Requirements to Access Credit

In order to at least minimize the inevitable credit risks, according to (Ghatak and Guinnane, 1999) a
thorough credit assessment should be conducted by the lenders especially concerning the borrowers`
character, collateral, capacity, capital and condition (what is normally referred to in the banking circles as
the 5C`s) should be conducted if they are to minimize credit risk. Such gathering of information is possible
primarily from your credit application and a credit bureau report, to determine whether borrowers are able
and willing to repay the debt. In the final analysis, every credit grantor attempts to answer the question:
how risky is it to lend or extend credit to this applicant? This decision is relatively easy for most because
the applicants will fall at one end of the continuum or the other of the six “C‟ s of credit.

Capacity: - is a factor in determining creditworthiness. It is assessed by weighing a borrower is earning

ability and the likelihood of continuing income against the amount of debt the borrower carries at the time
the application for credit is made.

Capital:-Factor in determining creditworthiness consisting of a borrower’s tangible assets and resources.

The presence of sufficient capital in a borrowers profile is an assurance that a debt could be paid from the
borrowers assets if the need arose.

Character: - Character is determined by analyzing how a borrower has handled past obligations.

Collateral:-is a real or personal property that a borrower pledges for the term of loan. When the borrower
fails to repay, the creditor may take ownership of the property by following legally mandated procedures.

Conditions:-A factor often considered with the factors of capacity, capital, and character when creditors are
analyzing an applicant’s creditworthiness. This factor consists of economic conditions that could affect a
borrower’s ability to repay, such as unemployment, seasonal work.

14
2.1.5. Performing Loans

The principal profit- making activities of banks are loans. In allocating funds, the primary objective of bank
management is to earn income while serving the credit needs of its community. Therefore, Lending
represents the heart of the industry. Loans are the dominant asset and represent 50-75 percent to total
amount of assets at most banks, that generate the largest share of operating income and represent the banks
greater risk exposure (Mac Donald and Koch, 2006).

Loans and advances are defined in the respective laws of different countries. In Ethiopia, under Article 13
(FDRE 592/2008) and (NBE/43/2008) Article (4.6) loans and advances are defined as:

“… Any financial assets of a bank arising from a direct or indirect advance (i.e. unplanned overdrafts,
participation in a loan syndication, the purchase of loan from another lender etc.), or commitment to
advance funds by a bank to a person that are conditioned on the obligation of the person to repay the
funds, either on a specified date or on demand, usually with interest. The term includes a contractual
obligation of a bank to advance by the bank on behalf of a person. The term does not include accrued but
uncollected interest or discounted interest.”

Loans and advances are the most profitable of all the assets of a bank. These assets constitute the primary
source of income by banks. As a business institution, a bank aims at making a huge profit. Since loans and
advances are more profitable than any other assets, it is willing to lend as much of its funds as possible.
However, banks have to be careful about the safety of such advances.

Because of controllable and uncontrollable factors, it is unlikely to have 100% of collection of loan.
Controllable factors are bank specific factors that are controlled by firm level and reflect overall bank
credit policy as well as inadequate credit analysis, loan structuring, and loan documentation, etc.
Uncontrollable factors are external factors or macro-economic factors that are not controlled by firm level.
It reflect adverse economic conditions, adverse change in regulation, environmental change surrounding
the borrower’s operation, and catastrophic events. So, in reality some of the loan will be nonperforming
(Daniel T, 2010).

So, loan may be considered as performing if payments of both principal and interest charges are up to date
as agreed between the creditor and debtor. Therefore, managing loan in a proper way not only has positive
effect on the banks performance but also on the borrower firms and a country as a whole. Failure to

15
manage loans, which make up the largest share of banks assets, would likely lead to the episode of high
level of non -performing loans.

2.1.6. Nonperforming Loans: Meaning and Nature

Providing loans to their customers are the principal functions of banks. In allocating funds, the primary
objective of banks was to earn income while serving the credit needs of its community. Lending represents
the heart in banking industry. According to (Mac Donald and Koch, 2006) Loans are the dominant asset
and represent fifty percent to seventy five percent to the total amount of banks assets. In most banks loans
generate the largest share of operating income and represent banks greater risk exposure.

The lending function is considered by the banking industry as one of the most important function for the
utilization of funds. Loans and advances are the most profitable of all assets of banks and constitute the
primary source of income by banks. Banks are business institution; hence, want to make as much profit as
possible through extending loans and advances. But due concern has to be given and banks have to be
careful about the safety of such advances, according to (M. Radha, and SV. Vasudevan. 1980).

Banks provide loans and advances in the existences of asymmetric information, certain level of risks are
inevitable. Accordingly, due to controllable and uncontrollable factors, it is unlikely to have 100% of
collection of loan and advances in reality. Loan defaults are inevitable given the uncertainty of the future
economic conditions and the existences of other controllable and uncontrollable factors. The main issue is
how to minimize the rate of this risk? How to increase asset quality of financial institutions, or minimize
the rate of non-performing loans by identifying factors that causes it?

Non-performing loans are closely associated with banking crises. Many authors argue that the magnitude
of non-performing loans is a key element in the initiation and progression of financial and banking crises.
Unless properly managed and kept at reasonable standard non-performing loans (NPLs) often associated
with bank failures and financial crises in both developing and developed countries (Gebru Meshesha,
2015). The issue of non-performing loans (NPLs) has gained increasing attentions in the last few decades
across the globe because the immediate consequence of large amount of NPLs in the banking system is
bank failure. The issue of loan default is related with none recovery/repayment of loans. When a borrower
cannot repay interest and/or installment of the loan after it has become due, then it is qualified as default

16
loan or non-performing loan. It is known as non-performing, because the loan ceases to “perform” or
generate income for the bank.

Our world has experienced banking crises in different times. Banking crises in turn cause very bad
economic conditions. Historically, the occurrence of banking crises has often been associated with a
massive accumulation of non-performing loans which can account for a sizable share of total assets of
insolvent banks and financial institutions, especially during a period of systemic crises. Nonperforming
loans generally refers to loans, which for a relatively long period of time do not generate income; that is the
principal and/or interest on these loans has been left unpaid for at least 90 days. The economic and
financial costs of bad loan are significant. Potentially, these loans may negatively affect the level of private
investment, increase deposit liabilities and constrain the scope of bank credit to the private sector through a
reduction of banks’ capital, following falling saving rates as a result of runs on banks, accumulation of
losses and correlative increased provisions to compensate for these losses. Impaired loans also have
potential for reducing private consumption, and in the absence of deposit guarantee mechanisms to protect
small depositors can be a source of economic contraction, especially when coupled with declining gross
capital formation in the context of a credit crunch caused by erosion of banks’ equity and asset (Fofack,
2005).

The definition of NLP varies across countries; there is no global standard to define nonperforming loans at
practical level. The concept has been defined in different literatures and by different scholars using
different parameters. Criterion for identifying non-performing loans varies throughout the world even
between countries. Some countries use quantitative criteria to distinguish between “good” and “bad” loans
like the number of days overdue, schedule payments while others rely on qualitative standards like the
availability of information about the client’s financial status, and management judgment about future
payments as used by (Teshome, 2010).

According to the International Monetary Fund, a non- performing loan (NPL) is any loan in which interest
and principal payments are overdue for 90 days or more. A number of other literatures have also tried to
define NPLs in their own ways. Even though, attempts are made to define NPL by different institutions and
scholars in different ways, still all of them indicate NPLs are Loans that are outstanding in both principal
and interest for a long period of time contrary to the terms and conditions contained in the loan contract.

17
Different endeavors are also made by a number of writers and authors to define what is meant by bad or
Nonperforming loans as per their understanding of the subject matter. Machiraju (2001) for instance,
expresses NPLs as a leading indicator of credit quality. NPLs or bad loans arise in respect of the loans and
advances which are given by banks to the whole range of different projects including but not exclusively
retail or wholesale, personal or corporate or short, medium or long term projects. NPLs are very sensitive
elements of a bank’s operations.

Another writer that attempted to define nonperforming loans are Caprio and Klingebiel (1996), cited in
Fofack (2005). They defined non-performing loans as those loans which for a relatively long period of time
do not generate income that is, the principal and or interest on these loans have been left unpaid for at least
ninety days. The authors further supported that non-performing loans are the loans which are not
generating income. According to (Guy, 2011), Nonperforming loans are also commonly described as loans
in arrears for at least ninety days and nonperforming loans have been widely used as a measure of asset
quality among lending institutions and often associated with failures and financial crises in both developed
and developing world.

Non -performing loans can also be defined as defaulted loans, which banks are unable to profit from it
(Tihitina, 2009). Usually loans fall due if no interest has been paid in 90 days, but this may vary between
different countries and actors. Defaulted loans force banks to take certain measures in order to recover and
securitize them in the best way.

From all these definition, it’s very clear that nonperforming loans occurs when a debtor has not met his or
her legal obligations according to the debt contract like where debtor has not made a scheduled payment, or
has violated a loan covenant of the debt contract.

Likewise, Ethiopia has also defined what is meant by nonperforming loans under National Bank of
Ethiopia’s (NBE’s) Directive no, SSB/43/2008.

It defines nonperforming loans as; “loans or advances whose credit quality has deteriorated such that full
collection of principal and/or interest in accordance with the contractual repayment terms of the loan or
advances in question”

It further provides that:

18
…, loans or advances with pre-established repayment programs are nonperforming when principal and/ or
interest is due and uncollected for 90 (ninety) consecutive days or more beyond the scheduled payment
date or maturity”.

In our country, therefore, loans become nonperforming when it cannot be recovered within certain
stipulated period of time that is governed by some respective laws. Accordingly, the following conditions
must be met to categorize some loans under nonperforming one:

a. A loan that is not earning income;

b. Full payment of principal and interest is no longer anticipated;
c. Principal or interest is 90 days or more delinquent or;
d. The maturity date has passed and payment in full has not been made.
Hence, in Ethiopia, if a loan is past due 90 consecutive days, it will be regarded as non- performing.
The criteria used in Ethiopian banking business to identify non-performing loan is a quantitative criteria
based on the number of days passed from loan being due.

2.1.7. Loan classifications in Ethiopia

The classification of loans into performing and nonperforming loan is not appropriate in reality. Loans may
take different other status than these two extreme classifications. As per directive number SBB/43/2007
loans are classified into five classes.

1. Pass loans: these are the loans that have not become any problem, present no special risk than the
normal risk inherent to any loan. Short term loans past due for less than 30 (thirty) days and medium
and long-term loans past due for less than 90 (ninety) days.
2. Special mention loans: these are the loans that have shown some early signs of trouble, such as missing
one payment, missing a few financial statements, deterioration of the collateral, etc. Some other events
not under the borrowers control may also trigger some alarm, such as deterioration of the labor or
political or security situation in the area where the business is located.

 Short term loans past due for 30 (thirty) days or more, but less than 90 (ninety) days and medium
and long-term loans past due 90 days or more, but less than 180 days.

19
3. Substandard loans: - these are the loans that have become real problems, missing payments for two
consecutive payments. They also present real weaknesses that jeopardize the orderly liquidation of the
loan. The following non-performing loans at a minimum shall be classified substandard:

 Short term loans past due 90 days or more, but less than 180 (one-hundred-eighty) days;
 Medium and long term loans past due 180 days or more, but less than 360 days
4. Doubtful loans: There are very serious questions about the borrowers capacity to repay, leaving the
bank with a strong possibility of loss, at least partial loss. The following non-performing loans at a
minimum shall be classified doubtful:
 Short term loans past due 180 (one-hundred-eighty) days or more, but less than 360 days;
 Medium and long term loans past due 360 (three-hundred-sixty) days, but less than 3 years.
5. Loss Loans: these are loans that are beyond hope after all means of recovery have been exhausted, or
loans that have not been performing for over 1 year. The only course of possible action is to take legal
actions to foreclose and write the loans off the book as a loss.
 Short term loans past due 360 (three-hundred-sixty) days or more;
 Medium and long term loans past due 3 (three) years or more;
Based on the above classification the loan of the banks considered as performing and nonperforming. If the
loan fall under pass and special mention category they are classified as performing loan otherwise it is
considered as non-performing loan (DBE, 2014).

2.2. Development Bank of Ethiopia

Among the formal source of credit in Ethiopia, Development Bank of Ethiopia is providing loan and
technical support for viable projects on the bases of individual credit. In line with this, Development Bank
of Ethiopia, Jimma District has scored the following performance during the fiscal year that ended June 30;
2015. The District has approved 2.123 billion birr and has achieved 386 percent of its plan. During the
same budget year, a total of 1.351 billion birr has disbursed to different sectors of the economy especially
agricultural projects. This revealed 223% of its plan has achieved. Regarding to loan collection, a total of
106.87 million birr was collected with registering 60% achievement (DBE annual report, 2015).

Based on DBE annual reports of 2011/12, 2012/13, 2013/14 and 2014/15, Jimma District has scored some
how good trend of non-performing loan ratio where 47% in 2011/12 fiscal year towards 9.58% in 2014/15,
2015/16 which is in line with the vision of the bank to be achieved by 2020. However, it needs also some

20
reduction in the coming years by using different rehabilitation mechanism. On the other hand, loan
repayment performance from whole projects including both healthy and unhealthy projects were face
difficulty as we seen relative to its own plan as well as number of projects that actually approved, where
scored performance not less than 100% of its plan within in those periods.

2.2.1. The Main Functional Areas of the Bank

The Bank is extending investment credits to creditworthy borrowers and projects that have received a
thorough appraisal and found to be financially and economically viable and socially desirable. Based on the
nature of the projects, DBE is providing long and medium term loans as well as short-term working capital
as a package. The term of loan is, however, to be determined based on the specific needs-and-requirements
of the projects. According to revised credit policy on 2016, the bank is providing:

New Loans: As per the current working credit policy and procedure of the bank all borrowers who wish to
obtain financing for new priority area projects are required to provide the minimum equity contribution of
25% of the total project cost in cash. The cash contribution placed upfront or gradually over a period not to
exceed 6 months from the loan contract signing date. The Bank will finance the remaining balance up to a
maximum of 75% of the total project cost after utilization of the 25% equity contribution by the borrower.

Expansion Loan: as per current working credit policy of the bank, all borrowers who wish to obtain
financing for the expansion of an existing priority area project and whose assets of the existing project are
not collateralized can access 100% financing of the expansion cost provided that the value of the existing
asset covers 40% of the total project cost. This means that the debt to equity ratio stands at 60:40. For any
cash contribution made by the promoter to cover the shortfall, the promoter can access additional loan from
the bank according to the debt to equity ratio of 60:40.

Working Capital Loans: in addition to the permanent working capital that is part of project cost, working
capital loan serves as bridge finance and is availed based on the cash flow of the project itself. The purpose
of working capital finance is for extension of inventory cycle, increase capacity utilization and cover short-
term cash flow problem of existing customers.

Co-financing (Syndicate Financing): in order to maintain the exposure limit, minimize risks and to
overcome occasional liquidity problems, the Bank may finance projects involving very large amount of

21
investment capital under co-financing arrangements with other national or international financial
institutions.

Guarantee Services: the Bank is providing financial guarantee services to its reliable clients. Export credit
guarantee service, on the other hand, is provided to well performing clients of other banks/financial
institutions with reliable and or good record of accomplishment.

Lending Managed Funds: the bank may undertake lending operations for supporting development
projects from managed fund at the request of governmental or non-governmental agencies.

Loan Transfer: healthy loans (loans performing as per the contract entered between the borrower and the
Bank) can be transferred to new clients upon the written request of both the original and the new clients.
However, the new clients’ credit worth capability to run the project should be confirmed by conducting the
required due diligence or KYC assessment and the request should be approved by the loan approval team.

Loan Buy-out: the Bank may buy-out loans extended by other local banks and local microfinance
institutions. However, the loans to be purchased under the buyout loan facility should be: Viable/ongoing
concern (operational) and Priority sector project loans.

Lease Financing: lease financing is a service in which the Bank provides financial service for purchases of
machineries, equipment and accessories for priority area projects. Whereby the lessee pays 10% of the
purchasing price of these assets to the Bank in advance and the lessee either returns the assets to the Bank
or purchases them at agreed price at the end of the lease period (revised credit policy, 2016).

Loan Processes
According to DBE loan procedure and manual (2016), the Loan Process of the Bank is designed to serve
the customer with a shortest possible time, minimum cost and high quality. This process starts its function
by attracting and persuading customers to apply for investment loans and ends at loan collection. This loan
process encompasses the following four independent loan-processing teams at corporate and District level
to handle loan-processing activities at various stages and responsibility levels:

22
Figure 2. 1Loan process of DBE

CREDIT TEAM APPRAISAL

• Due deligence TEAM
• screening • Feasibility study
• Propose amount

REHABILITATIO APPROVAL
N TEAM TEAM
• Sick projects • Over all policy decission
• approve/reject

Credit Process;
Project Appraisal teams at District;
PRLR team at Districts; and
Loan Approval team

Credit process
The Bank accepts applications from both recruited and walk-in customers if they fulfill the Bank’s loan
requirements branches (credit process) undertake. The recruiting customers by attracting and persuading
potential applicants using appropriate means of communication. On the other hand, due diligence or KYC
assessment undertake by the Bank to identify the integrity of the borrower. This is done to protect the Bank
from entering into relationships with inappropriate borrowers and to check the borrower’s creditworthiness.

23
Project Appraisal The project appraisal document focus on assessments and evaluations of the technical,
market, financial, and managerial viabilities as well as socio-economic benefits of projects. Within their
study report, market aspect of the project such as demand and supply analysis, market prospects and major
marketing areas of the proposed product or service, marketing strategy and arrangements for the products
or Services, price analysis for the products/services of the project and strengths, weaknesses, opportunities
and threats (SWOT Analysis) of the project.

Technical assessment includes location and site, project engineering, availability of utilities, availability of
inputs, production process and system of production, project implementation plan and environmental
impact assessment of the project. Project management aspect includes organization, management and labor
issues. Financial analysis (investment cost, working capital requirements, pre-operating costs, financing
scheme, equity requirement of the customer, revenue estimates and operating costs, projected financial
statements, viability and measures of project worthiness, working capital determination ). Economic and
social benefits and costs, conclusion and recommendation, loan repayment schedule, etc. are also included
as a part of appraisal document.

Loan Approval: Once loan applications for financing of development projects are received and screened
for appraisal by the Credit Process/branches, the Projects Appraisal District appraisal teams appraises the
project, it needs to be decided by the District Loan approval team. The LAT is to make decisions on the
approval or rejection of the loan. In this process, the Loan Approval Team deliberates and decides on the
loan approval document to accept or reject the loan proposal. Once the loaning decision is made in the
Loan approval Process/team, the case goes back to the Credit Process/branch for subsequent actions. Once
the loan is approved through the LAT, it comes back to the branch for facilitation and follow-up of loan
contract signing between the Bank and the borrower. After signing and registering of the loan contract is
compliance check for disbursement/equity utilization release to ensure that the borrower has fulfilled all the
requirements as per the agreement stipulated in the loan contract. In checking loan disbursement/equity
utilization, the credit process or branches use the Equity Release and Loan Disbursement Request Approval
Formats of the Bank.

Project Rehabilitation and Loan Recovery (PRLR)

Once a project is identified as sick in accordance with the above definitions and could not be resolved by
the branch, it will formally transferred to the District PRLR team, where a thorough analysis will be
conducted in order to decide whether the project can be rehabilitated. Such projects can be given relief and

24
concessions, as and when necessary, depending on their specific nature. The following are some of the
relief and concessions that may be extended by the Bank to sick projects for rehabilitation.

Management Intervention: the Bank may intervene into the management role of a given project by
overtaking its day-to-day management tasks fully or partially.
Provision of Additional Working Capital Loans: if provision of additional investment and/or working
loans is deemed necessary for rehabilitation of projects, it should be entertained accordingly.
Loan Repayment Rescheduling: whenever loan repayment rescheduling is found to be the best project
rehabilitation mechanism, it should be implemented accordingly.

Loan Transfer to Third Party: as an option for project rehabilitation task, upon agreement among the
Bank, the customer and a person willing to take over the project, projects (loans) can be transferred to
third party as a rehabilitation mechanism.
Different debt-equity ratio: In some cases, the PRLR team may propose higher debt-equity ratio. The
newly higher debt-equity ratio to be proposed based on the existing realty of sick projects and it must
justify that the sick projects will be rehabilitated. The PRLR team can also propose a different mode of
loan repayments so that the sick projects may not face cash constraints for the rehabilitation tasks.

2.3. Empirical Literature

In this part of the proposal different related literatures and studies will be critically analyzed and presented.
Accordingly, the first section emphasizes on any literatures and studies on factors affecting loan repayment
performances anywhere in the globe followed by related literature reviews conducted in Ethiopian context
and finally attempts will be made to reveal the reason why this study is found essential.

Different studies have been conducted regarding the determinant of loan repayment delinquency and
default especially on commercial banks, microfinance and agricultural credit borrowers by using different
technique of analysis. Among others binomial logit or probit regression model, multinomial logit or probit
and tobit are widely use and some of them are presented show the effect of those factors on loan repayment
delinquency and default.

To begin with, Munene, et al.(2013), in his study of Factors Influencing Loan Repayment Default in
Micro-Finance Institutions: The Experience of Imenti North District, revealed that there was significant
relationship between the type of business, age of the business, number of employees, business profits and

25
loan repayment default. There is strong link between technical training for loan beneficiaries and the
performance of entrepreneurial businesses among the remote communities. The study was conducted on
Microfinance institutions in Kenya to establish the causes of repayment defaults in Imenti North District,
Kenya using a descriptive survey design by incorporating 400 respondents of individual microfinance loan
beneficiaries and microfinance institution officials using census and cluster sampling procedures for micro
finance institutions officers and loan beneficiaries respectively. The data collected using both structured
and unstructured questionnaires and analyzed using descriptive and inferential statistics.

Another related study by Vigano, Lawra (1993), under a title “A Credit Scoring Model for Development
Banks: An African Case Study” has identified some important factors using a credit-scoring model. Taking
the case of Development Bank of Burkina Faso, Vigano found out that being women, married, aged,
proximity to the bank, use of better technology and being flexible to adjust to market changes, proper use
of the loan, project diversification, frequency of loan maturity, collateral, personal guarantee and being a
pre-existing depositor are negatively related to loan default risk. Loans in kind, long weighing period from
application to disbursement and being younger firm, past default, existence of other loan are those
positively related to loan default rate.

A study made on loan repayment determinants under the Social Emergency Loan Scheme (SEALS) in
Nigeria by Njoku and Odii (1991) employing multiple regression model based on

300 sample beneficiaries (9.3% of the total population) indicated that poor loan repayment performance
was due to late release of loan funds, cumbersome loan application and disbursement procedures and
emphasis on political considerations in loan approvals. In addition, loan diversion to non-agricultural
enterprises as well as low enterprise returns resulting from low adoption rate of improved agricultural
technologies contributed to poor loan repayment performance of small holders. Loan volume, years of
farming experience, farming as major occupation, years of formal education, family size, loan period, farm
size, farm output, value of assets and interest paid on loan were all highly significant determinants of loan
default. The coefficients of loan volume, years of formal education, family size and interest paid on loan
are positive while the coefficients for years of farming experience, loan period, farm size, farming as major
occupation, farm output and value of assets are negative.

Olomola (1998) was also utilized multinomial logit regression model for the analysis of loan repayment
determinant by grouping the dependent variable as well as explanatory variables. i.e., the dependent

26
variable is loan repayment performance (good credit risk paid or due < 30day, delinquent those who were
paid within 90 day from the due date and defaulter who were no fully paid after 3 month from the due date.
Whereas the independent variables are factor of loan repayment were classified as borrower related, loan
related, lender related and extraneous factor.

There was a research conducted by Chirwa (1997) to assess the determinants of the probability of credit
repayment among smallholders in Malawi using a model of probit. This model allows for analysis of
borrowers as being defaulters or non-defaulters. The result indicted that crop sales, income transfers,
degree of diversification and quality of information are positively related while size of club is negatively
related to the probability of repayment. Other factors like amount of loan, gender, family size and club
experience were found to be insignificant.

Using probit model of data analysis, Yacob (2014) analyzed the socio-economic factors that affect the
institutions loan repayment performance Eritrean Saving and Micro Credit Program of Dekemhare Sub-
Zone using the stratified sampling technique. The data collected from a sample of 134 respondents, which
were 67 defaulters and 67 non-defaulters. A structured questionnaire was used to collect the primary data
and descriptive statistics and the probit model were employed to analyze the data. The socio-economic
characteristics of the respondents were described using averages, percentages while the factors influencing
loan repayment performance of the saving, and Micro Credit Program loans were analyzed using the binary
probit regression model. Results of the regression analysis revealed that the level of education, loan size
and loan category have insignificant effect on the probability of the loan repayment. On the other hand,
age, gender, type of business and credit experience are significant determinants where age and type of
business have negative relationship and gender and credit experience have positive relationship with the
loan repayment probability.

Another research conducted in Ghana using probit model to identify factors affecting loan repayments by
the farmers of Brong Ahafo District of Ghana. Awunyo-Vitor (2012) searched the determinants of loan
repayment default among farmers in Brong Ahafo District of Ghana. The study employed Probit model to
investigate factors that influence farmers loan repayment default. Data used in this study was gathered
through a survey of 374 farmers in five Districts within Brong Ahafo District of Ghana. The results showed
that farm size, and engagement in off farm income generating activities reduces the likelihood of loan

27
repayment default significantly. In addition, larger loan size and longer repayment period as well as access
to training are more likely to reduce loan repayment default.

Theresa, et al. (2014) examined the determinants of loan repayment among cooperative farmers in Awka
North L.G.A of Anambra state, Nigeria. This study examined the determinants of loan repayment using
SPSS version 17. The study provides empirical evidence on the farmers “socio-economic characteristics as
well as determine which of the characteristics that influence loan repayment, the range of amount of loan
applied for, amount received and amount repaid by the cooperative farmers and organizational factors
affecting the farmers” credit repayment ability. Two coefficients (educational qualification and farm size)
are significant at 5%; and (loan application cost and collateral value) are significant at 1% respectively.
Age, membership duration, and income of the farmers were not significant but it shows a positive
relationship with loan repayment. There was a significant difference between the amount of loan received
and amount repaid by the cooperative farmers. All the organizational factors affecting the farmers‟ credit
repayment ability were significant at 0.000 significant levels.

The study by Oladeebo and Oladeebo (2008) confirmed that income, gender, farm size, age of farmers,
years of farming experience with credit, size of loan, family size, timeliness of loan disbursement, level of
education of farmers, sales of crops, degree of diversification, income transfer and the quality of
information were positive and significant determinants of agricultural credit repayment.

The other important study was done by Arene(1992). He evaluated the credit delivery system of Supervised
Agricultural Credit Schemes among smallholder maize farmers in Nigeria employing multiple regression
analysis. The result based on 95 sample maize farmers showed that high repayment farmers had larger loan
size, larger farm size, higher income, higher age, higher number of years of farming experience, shorter
distance between home and source of loan, higher level of formal education, larger family size, higher level
of adoption of innovations, and lower credit needs than low repayment farmers. The regression analysis
showed that size of loan, farm size, income, age, number of years of farming experience, level of formal
education and adoption of innovations are significantly and positively related to repayment rate, Distance
between home and source of loan, family size and credit needs were found to be negatively related to
repayment rate.

Causes and treatment of NPLs were studied in detail by Bloem and Gorter (2001). They agreed that “bad
loans” may considerably rise due to abrupt changes in interest rates. They discussed various international

28
standards and practices on recognizing, valuing and subsequent treatment of NPLs to address the issue
from view point of controlling, management and reduction measures.

A study conducted by Espinoza and Prasad (2010) focused on macroeconomic and bank specific factors
influencing NPLs and their effects in the Banking System. After a comprehensive analysis, they found that
higher interest rates increase NPLs but the relationship was not statistically significant.

The other study evaluated the factors influencing on repayment performance of farmers in Khorasan
Razavi province of Iran. The logit model seeks to explain the probability of loan on time repayment
because of any of the identified independent variables. The signs of the coefficient of independent variables
and significance of the variables were used determining largely the impact of each variable on probability
of dependent variable. Results showed that farmer‟s experience, income, received loan size and collateral
value have positive effect while loan s loan interest rate, total application costs and number of installment
implies a negative effect on repayment performance of recipients (Kahansal & Mansoori, 2009)

The authors like Sinkey and Greenwalt (1991) by employing a simple log-linear regression model and data
of large commercial banks in the United States from 1984 to 1987 investigate the loan loss-experience of
large commercial banks in the US; they argue that both bank specific and macro-economic factors explain
the loan-loss rate (defined as net loan charge offs, charge off rate which is also known as NPL rate) plus
NPLs divided by total loans plus net charge-offs of these banks. The authors find a significant positive
relationship between the loan-loss rate and internal factors such as high interest rates, excessive lending,
and volatile funds. Similar to other study, the authors further report that depressed District economic
conditions also explain the loss-rate of the commercial banks.

The findings above revealed that the probability of loan repayment depends on the borrowers’ specific
characteristics (i.e. age, education, experience, gender, family size, loan utilization, e.t.c.), institutions
specific (i.e. repayment installment, collateral, frequency of maturity, grace period, loan volume, interest
rate, number of disbursement, e.t.c.) and other factors such as business type, political influence, technical
advice, level of social cohesion (for micro enterprises), e.t.c. The strong side of the empirical studies
reviewed above is that they assessed al1 sources of loan default, which is the borrowers` willingness and
ability of repayment, the lenders` loan administration capacity, and other external economic factors.

From all studies discussed above, one can observe different financial institutions in Nigeria, Spain, China,
Kenya, Iran, Tanzania and others are assessed in relation with different factors that causes nonperforming

29
loans and forwarded their finding and recommendations for these countries. For example, In Kenya banks
shift away from concentration on security based lending and put more emphases on the customer ability to
meet the loan repayment and in China NPLs is transferred to Asset management (Waweru and Kalani,
2009).

So, it is appropriate to study what exactly affects loan repayment performances in Ethiopia.

2.4. Empirical Studies in Ethiopia

In this part, the Ethiopian literatures will be reviewed. Actually, there is no sufficient literature (published)
in our country, but some of the relevant studies are reviewed in the next section of this paper.

Michael (2006) has analyzed the impact of factors on loan repayment performance in informal sector of
financial institutions in Addis Ababa by grouping the independent variable (i) Borrower related causes; (ii)
Causes related to business operation; (iii) Lender related causes and (iv) Extraneous causes, A positive
coefficient shows that the variable is associated with a higher probability of being in the delinquent
category than that of being in the good credit risk category. On the other hand, a negative coefficient
indicates that the variable is associated with a lower probability of being in the delinquent category than
that of the good credit risk category.

In another relevant study by Abreham (2002) an investigation of determinants of repayment status of

borrowers and criteria of credit rationing were conducted with reference to private borrowers around
Zeway area who are financed by the DBE. The estimation result employing tobit model revealed that
having other source of income education, work experience in related economic activity before the loan and
engaging on economic activities other than agriculture are enhancing while loan diversion, being male
borrower and giving extended loan repayment period are undermining factors of loan recovery
performance.

Jemal, (2003) make a research on Microfinance and loan repayment performance, which was a case study
of the Oromia Credit and Savings Share Company (OCSSCO) in kuyu, the study area, Kuyu is found in
Oromia National Districtal State (ONRS). In his research methodology, he employed a logit model to find
the factors influencing on loan repayment performance in the micro finance institution. The sample size is
203, which is 9.3 percent of the total beneficiaries of the micro finance institution.

30
The independent variables used on the research includes, age of borrower, gender of borrower, educational
level of borrower, loan size in Birr, timeliness of loan release, loan diversion rate (ratio of loan diverted to
total loan receive, income from activities financed by loan (annual), annual income from other activities
(not financed by the loan), value of livestock in Birr, suitability of repayment period, use of financial
records, adequacy of supervision visits made to a borrower, location of residence of borrower, number of
dependents number of times borrowed.

The estimation results of the descriptive statistics and the tobit model show that education, income, loan
supervision, suitability of repayment period, availability of other credit sources and livestock are important
and significant factors that enhance the loan repayment performance, while loan diversion and loan size are
found to significantly increase loan default. In addition, female borrowers were found better in terms of
loan repayment.

Wondimagegnehu Negera (2012) in his study “determinants of NPLs on commercial banks of Ethiopia”
revealed that underdeveloped credit culture, poor credit assessment, aggressive lending, botched loan
monitoring, lenient credit terms and conditions, compromised integrity, weak institutional capacity, unfair
competition among banks, willful defaults by borrowers and their knowledge limitation, fund diversion for
unexpected purposes and overdue financing has significant effect on NPLs. Conversely, the study indicated
that interest rate has no significant impact on the level of commercial banks loan delinquencies in Ethiopia.

In order to analyze such determinant factors for successful loan repayment performance at bank, researches
has done at north District of DBE. According to the study of Kibrom (2010), factors that determine loan
repayment performance include; borrowers perceived need that is borrowers have to be given an
opportunity to borrow for their perceived needs, competence that is the borrowers past personal and profit
record, past prosperity etc. Borrowers personal character which were related with personal qualities of the
borrower including age, gender, educational level, house hold size, management capacity, loan utilization,
availability of other sources of income, bank experience etc. Factors which are related with the bank such
as structure of the bank, change in the lending policy, way of appraising the project, responsibility and
accountability of the staff members and external factors related with the macroeconomic condition of the
country, government policy and natural factors had analyzed.

Million, et al. (2012) examined the determinants of loan repayment performance among smallholder
farmers in East Hararghe Zone, Ethiopia specifically Kombolcha and Babile Districts. Structured

31
questionnaire was used to gather information from 140 smallholder farmers. Quantitative data was
analyzed using descriptive statistics such as mean, standard deviation, and percentage used. Moreover, a
two-limit tobit model was used to select variables which most significantly distinguish between non-
defaulters and defaulters of agricultural loan, from a set of personal and socio-economic variables
hypothesized to influence repayment behavior. The Two limit tobit regression model results indicate that
agro ecological zone, off-farm activity and technical assistance from extension agents positively influenced
the loan repayment performance of smallholder farmers, while production loss, informal credit, social
festival and loan-to-income ratio negatively influenced the loan repayment of smallholder farmers.

The study by Arega seyoum(Dr) and Tadele Tesfaye(2016) on factors affecting nonperforming loan in
central District of Development Bank of Ethiopia, using descriptive statistics including mean, frequency
and percentages and processed through computer loaded SPSS software and by collecting both primary and
secondary data. The study mainly focuses on bank specific and borrowers’ specific factors to establish the
relation between different variables and the nonperforming loans. This study was conducted in central
District of the bank, where the money borrowed from the bank was invested mainly on industry and service
sector of the economy. Hence, it is not appropriate to deduct the finding to the cases of Jimma District.

Generally, the Empirical researches on loan repayment performance in Ethiopia are not comprehensive
enough to cover all scenarios. For instance, the research by Abraham 2003 was conducted in Development
Bank of Ethiopia; however it is limited to loan repayments and its determinants in small scale enterprises.
The study by kibrom Tadesse (2010) conducted on successful loan repayments was on Development Bank
of Ethiopia in North District. It mainly focuses on determinants of successful loan repayment performances
in Northern part of the country where the loans are mainly allocated for industrial investments, and did not
cover the other side of repayment performances (which is NPLS). Another study conducted on factors
affecting nonperforming loans by Arega Seyoum(Dr) and Tadelle Tesfaye(2016) (both are Jimma
University lecturers) was conducted on Development Bank of Ethiopia Central District where the
economic sector the loan was utilized was mainly industry and commercial horticulture. The study focuses
basically on non-performing loans and limited to bank specific and borrowers’ specific factors to establish
relationship between variables and non-performing loans.

32
All of the above empirical studies fail to address the fact that factors other than those bank specific and
borrowers specific would also affects repayments of loans. In addition all of the studies were conducted in
a geographically industrialized area of the country and the unique characteristics of agricultural
investments which are prevailing in Jimma District is not covered by the studies. Furthermore, now a day
there are many changes regarding the bank policy, interest rate, introduction of new products,
organizational structure as well as the general macroeconomic condition of the country and the world.

Therefore, this research will contribute towards filling the gap by identifying and analyzing the factors
affecting loan repayment performances in Jimma District, in which no research is conducted and most of
the borrowers are engaged in the agricultural and service giving sectors.

33
 CHAPTER THREE:

RESEARCH METHODOLOGIES
In the preceding chapter the review of related literature on factors affecting loan repayment performances,
the empirical studies and their respective findings are presented. As loan repayment performances basically
comprises of performing and non-performing loans, different studies related to both scenarios were
thoroughly reviewed under the preceding chapter. This Chapter presents the methodology that provides a
detailed direction about the methods that the researcher used in conducting the research. Hence, the
research design, description of the study area, data type and source, methods of data collection, sampling
techniques, methods of data analysis and definition variable, measurement and description of variables are
discussed.

3.1. Research Design

Research design is a comprehensive plan. It is a blueprint for empirical research aimed at answering
specific research questions or testing specific hypotheses (Anol Bhattacherjee, 2012). Research design is
the program that guides the researchers in the process of collecting, analyzing and interpreting the data.

Therefore, the nature of problem and objective of any study usually determine the type of research design
adopted by researcher. A choice of research design reflects the priority of a researcher about the
dimensions of the research process and methods. The objective of this research is to identify the factors
affecting loan repayment performances in Development Bank of Ethiopia, Jimma District. The collected
data mainly focused on description of borrower’s characteristics, lending institution/bank related factors,
business/project related factors and external factor that affects loan repayments and their relationship
among the dependent and explanatory variables. Therefore, both qualitative and quantitative research
method were used in the study.

3.2. Description of the Study Area

The geographical study area of this research is located in the south western part of the Ethiopia and
comprises some parts of three National regional states. Accordingly the geographical coverage of the area
includes Jimma zone, Buno Bedelle Zone and Illu Abbabor Zone from Oromia National regional state,
Kafa zone, Benchi Madji Zone, Shaka and Dawro zone from South Nations and nationalities regional state

34
and the whole region of Gambella Peoples regional state. This area is categorized as ever green and
suitable area for agricultural and commercial farming investments according to DBE loan manual.
Currently Development Bank of Ethiopia has about 12 Districts all over the country. Jimma District is one
of such Districts covering the above listed administrative zones from three regional states. Development
Bank of Ethiopia Jimma District has two grade “A” branches (Jimma branch and Gambella branch) and six
grade “C” branches (Matu, Bedelle, Agaro, Bonga, Mizan and Teppi branches).

Pursuant to the currently existing organizational structure of the bank grade A branches can provide credit
to their borrowers and administer their active loans, while the role of grade C branches were limited to
collecting repayments from old loans and handling inactive loans. There is also a team called Project
Rehabilitation and Loan Recovery Team (PRLR) responsible to handle sick/nonperforming loans under the
District. In this study, grade C loans are excluded from the population of the research for they are totally
dead and write-off and hence, the population of the study is limited to loans under the District office (grade
A branch) borrowers and sick loans handled by PRLR team under the District.

3.3. Data Type and Source

The data employed in this study is both primary and secondary data. Accordingly, the primary source of
data was collected through questionnaire and interviews from the sample population. A structured and semi
structured questionnaire with open ended and closed ended type was distributed and collected from 150
borrowers identified using stratified sampling from the loan position of the District as of June 30, 2016. It
excludes borrowers whose repayment installment has not yet matured because it would be premature to
assess the real performance of the projects as well as credit worthiness of the borrowers unless they are
practically tested by their repayment record. The staff borrowers were also excluded from the population of
the sample because, such loan is either personal loan or housing loan which are basically the privilege to
the staff members of the bank and the repayment performance for such credits are very much secured as
long as the employee remains the worker of the bank. Further, such loans didn’t qualify the requirements of
financing for projects/working capital loans, hence excluded. The financial position of the borrowers at the
end of fiscal year i.e. June 30, 2016(the bank’s financial statement closing date) is considered was used as
another of source data.

35
Primary data: The primary data was collected from original source (borrowers) through questionnaire. The
primary data collected through semi-structured questionnaire distributed to the borrowers; and interviews
conducted to the bank officials and staffs.

The questionnaire included both close and open-ended questions. The close-ended questions covered the
personal information, institutional, external factors, loan and repayment related questions. The open-ended
questions dealt with the perception of clients towards the bank and their feelings. All questionnaires
translated into Amharic. The questionnaire was pre-tested by three borrowers before conducted for the
whole sample. Besides, interviews were made with selected loan officers and managers, and relevant
documents were reviewed.

Secondary data: secondary data were used as a source of data in this work to determine the repayment
performances of the bank in the previous consecutive years and to determine the sample size population of
the study. Especially, annual reports, loan portfolio of the banks and others publications of the bank were
used as a secondary data.

3.4. Method of Data Collection

In order to achieve the objectives stated in the preceding section and considering the nature of the problem
and the research perspective, the researcher used both quantitative and qualitative data. The Primary data
were collected through primary data collection techniques mainly using structured and semi structured
questionnaire and interviews with the officials and senior officers of the bank. Secondary Data were
directly gathered from records and published documents of the bank. The data collected include aggregate
loans outstanding balances, NPLs as at the annual closing date, June 30 2016 and others as found
appropriate. For the purpose of comparison the surveyed banks data for the years preceding and following
years performances were also considered.

3.5. Population and Sampling Method

Determining type and method of sampling mainly depends on the types of population that the study covers.
According to (Kothari, 2004), if the population from which a sample is to be drawn does not constitute a
homogeneous group, then stratified sampling technique is applied to obtain a representative sample. The
usual method, for selection of items for the sample from each stratum, resorted to is that of simple random
sampling.

36
Hence, Sample selection was based on stratified sampling where borrowers were selected based on the
diversified investment activities they are carrying on and in proportion to the population classification in
terms of their loan status. The loan position of the District shows that as of June 30, 2016, the total number
of borrowers including the staff loans in Jimma District (Jimma Branch, Gambella branch, PRLR team of
the District and the remaining six branches) was three hundred seventy five (375). From the total loans 84
of them were staffs loans (short and long staff loan), and 18 were not matured yet, the remaining 273 loans
were listed in the loan position of the District having different loan status and in all cases their first loan
repayments is mature. Thus, the total population of the study is 273. Out of these, 217 loans were
performing loans/non-defaulters and the remaining 56 were defaulters (loan status including substandard,
doubtful and loss). According to their loan status, 79.49% of the total populations were performing loans
while the rest 20.51% were nonperforming. Borrower whose maturity date is not yet due and Staff
borrowers are excluded from sampling of population because loans for employees is mere privilege and
didn’t qualify the requirements of provision loans for other businesses and its repayment is almost granted
as long as the employee remains in the bank. On the other hand, in case of loans which are not matured yet,
it was excluded from sampling because it is difficult to study about the factors affecting loan repayment
performances while the loan is not yet matured, it will be prematurity. Therefore, based on proportional
stratified random sampling the samples of 150 borrowers were selected, out of which 110 were performing
loans and 40 were defaulters. In addition, two principal officers from each team (credit team, appraisal
team and approval team), two senior officers from different teams, Jimma and Gambella branch managers
and Jimma District manager was interviewed, their comments and ideas are used in interpretation and
recommendation of the study.

3.6.Method of Data Analysis

The data collected through the above stated techniques were thoroughly coded and checked for consistency
and analyzed and interpreted using both descriptive statistics and econometric analysis. Accordingly, the
researcher analyzed the data using descriptive statistics (frequencies, percentages, mean, and standard
deviation) to obtain information on the factors affecting loan repayment performances and binary logistic
econometric model (logit) was used to identify the factors of loan repayment ability in Development Bank
of Ethiopia Jimma District. Descriptive statistics was employed to analyze the data and the results were
tested with non-parametric tests of significance, whereas econometric analysis, specifically binary logistic
regression was used to identify statistically significant variables in relation to the dependent variable. A

37
loan repayment performance refers to the ability/capability of borrowers to duly repay loans or fail to repay
their loans. Hence, the dependent variable is dummy variable. If Borrowers experienced well repayment
performances the dependent variable takes a value of 1, and if the borrowers fail to repay their loans as per
the terms of agreements/contracts the dependent variable take the value of 0. So, the level of significance
and influence of each independent variables where defined and identified using both descriptive and
econometric analysis against the dependent variable.

Finally the analyzed data was presented in the form of table and percentage in order to make the data
understandable and attractive detailed statement would support these tools.

3.7. Model Specification

Data collected through the above stated methods were analyzed using different techniques. According to
(Kothari, 2004) data analysis requires a number of closely interrelated operations such as establishment of
categories, the application of these categories to raw data through coding, tabulation, and then drawing
statistical inferences.

Hence, the researcher analyzed the collected data using descriptive statistics (frequencies, percentages,
mean, and standard deviation) to obtain information on the factors of loan repayment performances
especially to summarize and conclude the implications of qualitative data and binary logistic econometric
model (logit) was used to analyze the determinants of loan repayment ability of the District borrowers.

According to Vasisht (n.d), logit analysis produces statically sound results, which can be easily interpreted,
and the method is simple to analyses. Assume the following basic model, it can be express the probability
that y = 1 as a cumulative logistic distribution function.

The cumulative Logistic distributive function can then be written as:

38
Pi= prob(Y = 1| X) is the response probability. The non-response probability (1- Pi) is also evaluated as:

Note that the response and non- response probabilities both lie in the interval [0, 1]; Zi ranges from - ∞
to + ∞ , and hence, are interpretable. There is a problem with non-linearity in the previous expression, but

this can be solved by creating the odds ratio and its log-transformation.

(Gujarati, 2004)

Li is called the logit, thus, the log-odds is a linear function of the explanatory variables. The above
transformation has certainly helped the popularity of the logit model. Note that for the linear probability
model it is Pi that is assumed to be a linear function of the explanatory variables. The odds ratio can be
interpreted as the probability of something happening to the Probability it will not happen. Accordingly,
the estimated models used in this study presented as follow.

LRP =β1 + β2(Gedr) + β3(Ag) + β4(Mar) + β5(Educ) + β6(Hhs) + β7(Exp) + β8(Othbus) + β9(Busfrm) +
β10(Bussctr) + β11( Income) + β12(Lnamt) + β13(Div) + β14 (Eq) + β15 (Grprd) + β16 (Folup) + β17
(Coll) + β18 (Int) + β19 (KYC) + β20 (Timhzn) + β21 (Mrkt) + β22 (Wthr)

Where LRP, Gedr, Ag, Mar, Educ, Hhs, Exp, Othbus, Busfrm, Bussctr. Incm, Lnamt, Div, Eq, Grprd,
Folup, Coll, Int, KYC, Timhzn, Mrkt, Wthr denotes Loan Repayment performance, Gender, Age,
Marriage, Education, House hold size, Experience, Other business, Business form, Business sector,
Income, Loan size, Diversion, Equity, Grace period, Follow up, Collateral, Interest, Kyc, Time horizon,
Market and Whether respectively.

39
β1 = an intercept, Where β2, β3, β4, β5, β6, β7, β8, β9, β10, β11, β12, β13, β14, β15, β16 β17 represent
estimated coefficient

On the other hand, binomial logit regression model of regression was used for econometrical or statistical
analysis of the study. The statistical analysis of model with qualitative dependent variables can be viewed
as the problem of predicting probabilities for the various possible values (responses) of the dependent
variable. Probit and Logit are well-known techniques for the case when there are only two responses,
typically the occurrence or non-occurrence of some event. Both have essentially the same interpretation -
the probit is based off an assumption of normal errors and the logit off of extreme value type errors. The
logit has slightly fatter tails than the probit possibly making it slightly more robust. In binary studies probit
and logit are largely undifferentiated.

In this research, the researcher selected logit model because it is slightly easier to introduce random
parameters to and estimate as a simulated maximum likelihood regression. Even though, both have
simple and fairly elegant representations in the binary case on paper, in cases of choice with more than
two alternatives the logit quickly becomes the preferred choice as the probit model is difficult to estimate
when alternatives are above 3. So, logit was used to present the econometric results and analysis of the
research.

Odd ratios were used to explain the degree of influence of variable. Odd ratio/ logistic regression
coefficients provide information on the probability of being on time payer and default as we change the
independent variable by on unit reference to on time payer category

Furthermore, Likelihood Ratio (LR) Chi-Square test used to show that at least one of the predictors'
regression coefficients was not equal to zero in the model than the observed statistic under the null
hypothesis; the null hypothesis was that all of the regression coefficients in the model are equal to zero.
To see the significant of each explanatory variable t- test was applied. Whereas to detect the presence of
violation on basic classical linear assumptions of the model, different techniques were applied regards
autocorrelation, multicolinearity, hetroschedasticit and normality test.

3.8. Variables of the Study

In order to achieve the extracted objectives of this research, the researcher selected different variables
based on literatures that could affect the dependent variables either positively or negatively. Hence, based

40
on availability of data the variables selected in this research are to signify the loan repayment performance
and the variables which are attributable and likely to influence the dependent variable were listed down
with their respective expected sign.

Selection of variables was based on empirical literature review as presented under the preceding section to
establish the factors affecting loan repayment performances. While guided by the literature review, the
researcher also considered other factors likely to influence loan repayment. To establish the factors
affecting loan repayment, the researcher summarizes these variables under four broad categories, (i)
variables that are related to the characteristics of the borrowers, (ii) factors related to the lending institution
(the bank), (iii), factors related to the business/project itself and (iv), factors emanated from the external
environments.

3.8.1. The Dependent Variable

The dependent variable of the study is loan repayment performances. Loan repayment performance (LRP)
is the ability to repay the loan as per the loan agreement and/or inability to repay the loan by either failing
to complete the loan as per the loan agreement or neglect to service the loan. Several African studies on
loan repayment performance have estimated the factors of loan repayment performance with a binary loan
outcome – defining borrowers as either current on their loan repayments or in default. There are a number
of different factors that would affect this dependent variable either positively or negatively.

Dependent variable defined

Taking in to consideration the loan status classification by national bank of Ethiopia, according to the loan
manual of Development bank the dependent variable of this study loan repayment performances is dummy
variable and all other independent variable are encoded as dummy as well as categorical explanatory
variables, which is appropriate to use STATA software in the following form. Hence, the study has encoded
both dependent and independent variables in the following way by taking in to consideration of being dummy and
categorical variable.

Loan repayment performances – encoded as dummy as follows;

1. Pass: due date less 90 (ninety) days

2. Special Mention: due date 90 days‟ ≤Y<180days Performing loan ……..…. 1

3. Substandard: due date 180 days‟ ≤Y<360 days

41
4. Doubtful: due date 360 days‟ ≤ Y<3 year Non performing loan …….. 0

5. Loss: due date 3 years ≤ Y

3.8.2. Definition and Hypothesis on Independent Variables

Selection of variables was based on empirical literature on the factors affecting loan repayment
performances. While guided by the literature review, the researcher also considered other factors likely to
influence loan repayment. To establish the factors affecting loan repayment, the researcher summarized
variables in to four categories, factors related to characteristics of the borrowers, factors related to lending
institution, factors related to the nature of the business/project and other external factors. So, dependent
variable (loan repayment performance) is expected to be explained by the following independent variables:

3.8.2.1.Factors Related To Characteristics of the Borrowers

These factors are very personal and attached to the behavioral attributes or personal integrity of the
borrower. Such type factors are many in numbers; the followings are some to be evaluated under this
research;

Gender: determines whether male or female borrowers perform better than the other. It is a dummy
variable taking, 0 for female and 1 for male. The female borrowers have a tendency for better loan
repayment. This means that lending to women can lead to their economic empowerment and inculcate them
a culture of hard work and financial discipline, which can lead to high loan repayment rates, thus women
borrowers may have high loan repayment performance. Thus being women expected to have a positive sign
on loan repayment.
Age: age of borrower in years. It is a continuous variable but rearranged as 1) young age (15-30) 2)
mature age (31-50) and 3 old age (above 51). It is argued that older borrowers are wiser and more
responsible than younger borrowers. On the other hand younger borrowers are argued to be more
knowledgeable and more independent. That means on the other way round, the older person may have a lot
of experience on business, which may lead to loan repayment, and the younger one may have limited
experience attributed to his age and this may lead to loan repayment. Hence, age contribution to loan
repayment performance cannot be predetermined.
Marital status: this variable evaluates whether single, married or divorced borrowers showed any
difference in repayment performances. It’s generally believed that marriage brings stability to once life and

42
equips how to act towards something responsibly. It is a continuous variable but rearranged as a dummy
variable; taking 1 if the borrowers are single, 2 if married and 3 if the borrowers are divorced/widowed.
The borrowers who engaged in marriage can have financial management experience in their home. Thus,
having such managing experience can be reflected in their loan utilization. The expected sign is negative to
being default loan.
Education: generally education is among the primary tool that has transformed the world order as it stands
today. Education improves once performance quality. Higher educational levels enable borrowers to
comprehend more complex information, keep business records, conduct basic cash flow analysis and
generally speaking, make the right business decisions. So, it is important to test whether education level
difference between and among borrowers have brought any change in their loan repayment performances.
This is a continuous variable but arranged as categorical variable, taking 1 if the borrowers have no formal
education, 2 where the borrowers attended primary educational, 3 if borrowers attended secondary
educational and 4 if the borrowers attended college/university education. This factor is expected to have a
positive impact in loan repayment performance, because higher educational levels enable borrowers to
comprehend more complex information, keep business records, conduct basic cash flow analysis, and make
the right business decision. Hence borrowers with higher levels of education may have higher repayment
performances.

Family size: this variable is all about the number of dependents on the borrower. Hence, using this variable
comparison is made between borrowers having small family size with those having medium or large family
numbers against their loan repayment performances. It is a continuous variable(measured in number of
members of farm family but arranged as 1 for small family size(1-3) 2 for medium family size(4-5) and 3
for large family size(above 6); it is assumed that the larger the family size the negative influence on loan
repayment performance which is attributed to higher house hold expenses. There is a possibility of loans
diverted to unintended purposes because of many responsibilities resulting from meeting the needs of many
members of the family. Hence borrowers with large family sizes may have lower repayment performances.

Credit Experience: it is a continuous variable but rearranged as dummy taking 0) where borrower have no
any credit experience and 1) if borrower have credit experience. Borrowers who have been in business
longer are expected to be more successful with their enterprise. They have more stable sales and cash flows
than those who have just started. Thus, those who are more experienced will have high repayment rates.
Hence, it is expected that experience will positively affect loan repayment performance of borrowers.

43
3.8.2.2.Factors Related To Lending Institution

These factors are mainly related to technical capacity and strength and/or weakness of lending institution in
providing credit services to its borrowers particularly in screening, appraising, approving, supervising and
collecting loans from their borrowers.

Loan size: It is the amount of money permitted for the borrowers. In case, the amount of money
permitted/lent to borrowers have any influence or not was evaluated using this variable. In order to operate
the investment with its full capacity and cover all necessary costs, sufficient amount of money is required.
Von Pischke (1991) noted that efficient loan sizes fit borrowers’ repayment capacity and stimulate
enterprise. If amount of loan released is enough for the purposes intended, it will have a positive impact on
the borrower’s capacity to repay. If on the other hand the amount of loan exceeds what the borrower needs
and can handle, it will be more of a burden than help, thereby undermining repayment performance. Also
positive or negative sign may be expected if the loan is too small. If the loan is too small it may be easy to
repay such loans thus enhancing performance (i.e. positive sign). However, too small loan may not bring
commitment on borrowers to use the loan productively (Von Pischke, 1991). It may also encourage
borrowers to divert the loan to other purposes, increasing credit risk and undermining performance, in
which case a negative sign for the variable is expected (Vigano, 1993). It is continuous variable but re
arranged as 1 if the amount is from 1-10 million, 2 if the amount is from 11- 20 million and 3 if the amount
is above 21 million birr.

Follow up: It is a dummy variable that proper follow up taking as 1 and otherwise 0. It is done at different
stage of the project. Project follow up can be done at the stage of project under implementation, during
implementation and commencing to commission. Undertaking of fledged follow up as per the schedule
boost the projects /customers to accomplish their task duly and the project can generate revenue. The
chance of being a default loan is low if proper follow up has done.

Grace period: As of Abreham (2002) if large grace period is given, the project will have sufficient time for
implementation so that borrowers could properly utilize the loan for the intended purpose and to generate
adequate income after it starts operation. Therefore, it will not face repayment problem when the loan due
later. Grace period is a dummy variable, 0) if sufficient grace period is not given and 1) if sufficient grace
period is given to the project. The expected sign of grace period to nonperformance/defaulting is negative.
Know Your Customer/Due diligence (KYC): It is a screening stage evaluation of the borrower and the
business whether it is creditworthy or not. It is an entry point assessment. Conducting proper due diligence

44
is require in every applications to access credit from Development bank of Ethiopia. In this stage the
borrowers all round aspects are assessed in relation to its personal characteristics from past to present,
fulfillment legal documents to be a creditor, project management , capital adequacy , credit relation and
experience , availability of inputs and identification of risk. Know Your Customer/Due diligence (KYC) is
a dummy variable that well done due diligence taking as 1 and otherwise 0. Therefore, adequate due
diligence, the expected sign for being default is negative.
Lending Interest rate: It is a dummy variable taking 1 if increase in interest rate negatively affected
repayment performance and 0 if otherwise. Increase in lending interest rate, increases the amount of loan to
be repaid per installments. Hence, it is expected to have positive relation with the default.
Timeliness/time horizon: Timeliness of release of loan (measured as a dummy, 1 for loans released at the
right time and 0 for loans not released at the right time). Investments in DBE Jimma District area were
predominantly, farming an activity which is mostly seasonal and rain fed, hence if the loan is not released
at the right time yield will be affected and repayment performances may below. Johnson and Rogaly
(1997) noted that timeliness of loan disbursement is important when loans are used for seasonal activities
such as agriculture. They argued that complicated appraisal and approval procedures, which might delay
disbursement, influence a program of seasonal loans for farmers who use to buy inputs. Further they noted
that this could in turn worsen the prospects of repayment by diverting loan to non-intended purpose. In
such cases a positive sign is expected.

Collateral: Collateral is the guarantee for repayment performances. If the borrower secures high valued
collateral relative to the loan size, the lender may feel that it will not be a loser in case the borrower
defaults. Borrowers exert their maximum effort to repay the loan if the collateral towards the loan size is
high and vise versa. Collateral is dummy variable taking 0) if sufficient collateral is not attached for the
loan and 1 if the collateral is sufficient to restore the lent money in case of default. So, collateral will have
positive sign towards repayments of money.

Equity: For this research purpose equity is the share of borrowers in the total investment capital of a
project. The amount of equity of owners in the investment determines the sense of belongingness and the
extent of responsibility. If the amount of equity is higher, the borrower feels a sense of ownership and will
strive to recover the loan and make the whole asset his sole property. Equity is a dummy variable taking 0
if equity less or equal to 30 percent of total investment and 1 if the equity is greater than 30 percent of total
investment. Thus, a positive sign is expected.

45
Loan diversion: Diverted loans miss the targets of the investments. The projects will not generate
necessary earning and benefits to repay loans if loan diverted. According to Abraham (2002), loan
diversion is problematic only if the business which received the diverted money fail to pay back. But the
loan manual of DBE prohibits any kinds of loan diversion. Loan diversion is a continuous variable but
arranged as dummy variable taking 0) if not diverted and 1) if diverted. So, loan diversion contributes
positively to default.

3.8.2.3.Business Specific/Project Related Factors

Business Sector; It is clear that different types of projects have different level of risks. Thus, borrowers
with different types of projects may have different repayment rates. However, it is clear that borrowers who
engage in agriculture and agricultural related product sectors are expected to have default loan, this is
because agriculture and agricultural related projects are seasonal and more exposed to different risks than
service sectors. Business Sector is continuous variable but arrange as 1 for agriculture, 2 for service and 3
for industry. The expected sign for agriculture is positive for default.

Other business:-It is a dummy variable taking 0 if didn’t have another business and 1 if having another
business. It is all about experiencing some other business in addition to the current project that the
borrower involved. If the borrower has other source of income, he may not spend the income that is
generated from the current business other than loan repayment, or vice versa. It was expected positive or
negative sign because if the promoter has additional business other than the project, he or she will divert
loan and expend more time on other business. On the other hand, borrower who has other business might
use it as the source of short fall of capital or loan repayment.

Business form: The ownership and the level of responsibility of the business matters in operating any
business. This is to identify whether the ownership is sole proprietorship, PLC or other business form.
Because some form of business leave individual responsibility and accountability may cause for business
failure and hence low repayment performance. It is continuous variable but rearranged as 1 if sole owner 2
if PLC and 3 if SHC and others.

Income/profit: it is dummy variable taking 0 if sufficient income is not generated and 1 if sufficient
income compared to feasibility study is gained from the business. Income/profit is amount of money
generated from the business itself with a given fiscal year. Hence the sign it takes may not be single,
because if sufficient income is generated the variable shows positive sign to repayment and otherwise.

46
3.8.2.4.Other Factors (Macroeconomic Factors)

Market conditions (inflations): this is dummy variable taking 0 if market condition didn’t affected
repayment and 1 if market condition affected repayment of the borrower. In contemporary world the
market condition is unpredictable. There fluctuation from time to time, hence, shows negative sign to
repayments of loan.
Whether conditions: this is dummy variable taking 0 if whether condition didn’t affected repayment and 1
if whether condition affected repayment of the borrower. Now days the global whether condition is
threatening the life of human beings. El Niño for instance, caused some unpredictable and unbelievable
disasters in the last year. So, shows negative sign towards repayments of loan.

Explanatory variables encoded

Explanatory variables are encoded as dummy and categorical variables as follows;

1. Gender ………………………….....male= 1 and otherwise = 0
2. Age ………………. 1= Young age (15-30) 2 = mature age (31-50) 3= old age (51 above)
3. Marital status ……………. Single = 1 married = 2 divorced/widowed = 3
4. Family size --------------------- small = 1 medium = 2 large = 3
5. Education level …….. 1= no formal educ. 2= primary educ. 3= tertiary educ. 4 = coll/above
6. Other business ………………....Have other business = 1 and not have other business = 0
7. Experience ------------------------- if yes = 1 and if not = 0
8. Loan size ……………… small (1-10m) = 1 medium (11-20m) = 2 large (21 above) = 3
9. Kyc/due diligence …………….. If properly undertaken =1 if not = 0
10. Business sector ………………...Agriculture = 1 and service = 2 Industry =3
11. Business form ………………. Sole owner = 1 Plc. = 2 and SC and others = 3
12. Equity …………….. Equity greater than 30 percent = 1 and less/equal to 30 percent = 0
13. Time horizon……………………….timely = 1 delayed = 2 and too late = 3
14. Loan diversion ------------------- if diverted = 1 and if not = 0
15. Collateral …………………....sufficient collateral = 1 and not = 0
16. Grace period ……………….. Sufficient time given = 1 and if not = 0
17. Follow up …………………... proper follow up = 1 and if not = 0
18. Lending interest rate ……….....change in interest rate affected repayment = 1 and if not = 0

47
19. Income/profit ---------------------- if sufficient income generated = 1 and if not = 0
20. Market condition ……………..Poor market condition = 1 and were not poor = 0
21. Weather condition …………....Bad weather condition = 1 and were not bad = 0

To sum up, discrete dependent variable (loan repayment performance) was expected to be explained by listed
discrete and categorical independent variables with their sign as shown in the table 3.1.

Table 3. 1summary of Expected Sign (+/-) of Explanatory Variables in this Study

No Explanatory Measurement Definition Expected

Variables Sign

1 Age Categorical(young, The older the age having high experience +

mature, old age) contributes a lot for loan repayment
2 Gender Dummy, male & female lending to women, lead to high loan +/-
repayment rates
3 Marital status Categorical(single, Married borrowers can take great care +/-
married and than non-married for default
divorced/widowed)
4 Education Categorical (no educ, Being literate borrowers well informed +
primary, high school, and contributes for default negatively
coll.)
5 Credit Dummy,(yes or no) borrowers who have no or less +
experience, will contribute for default
Experience
6 Household Categorical(small, The smaller family size less probability -
medium and large family) being default
Size
7 KYC Dummy(properly made performing due diligence thoroughly less +
and otherwise) probability being default
8 Follow up Dummy(yes or no) Performing fledged follow up as per the +
schedule the probability of defaulting is
less
9 Business Categorical (agriculture, agricultural projects are seasonal, the rate -
Sector service and industry) for default so high
10 Equity Dummy (less than thirty The larger the equity of owners, the less +/-
percent or above) the probability of being default.
11 Time horizon Categorical(timely, disburse the loan timely, less probability +/-

48
 delayed and too late) being default
12 Grace Period Dummy(sufficient time large grace period is given for projects, +
given or otherwise) less probability being default
13 Collateral Dummy (sufficient When bank loan provided with sufficient +/-
grantee provided or not) collateral, the probability of being default
decrease.
14 Income/ Dummy (sufficient The more the profitability of projects, the +
income or otherwise) less the probability of being default.
Profit
15 Loan size Categorical(small, Increasing loan size ,increasing capital , +/-
medium and large) generates revenue, less probability being
default
16 Loan Dummy (loan diverted or Diverted loan miss the target, hence -
diversion not) probability of being default is high
17 Other Dummy (whether the Other business may help to repay or be a -/+
business borrower have another source of diverting loan; so the
business or not) probability may depend on scenarios.
18 Business Categorical (sole owner, Some business form limit the liability of -
form plc and Shc) owners, probability of default is high in
such cases.
19 Interest Dummy (increase of Increase in lending interest rate increases -
lending interest affect or the probability of being default
no)
20 Whether Dummy (product affected If whether condition was not normal in -
by whether or not) the probability of being default is high.
21 Market Dummy(the price of +
products affected by
market or not)

Source: compiled from the definition and literatures, 2017

49
3.9. Conceptual Framework

According to Shields (2013), a conceptual framework was expressed as the way ideas are organized to
achieve a research project’s purpose. It is connected to the research purpose.

A conceptual framework is a basic structure that consists of certain abstract blocks which represent the
observational, the empirical and the analytical/ synthetically aspects of a process or system being
conceived. The interconnection of these blocks completes the framework for certain expected outcomes.

Hence, the researcher developed the following conceptual framework to easily assess the relationships
between those factors and loan repayment performances. In addition to methods of data presentation and
analysis, the research frame work has described in the following diagram considering both explained and
explanatory variables. Thus, the simple diagram shows the effect of explanatory variables (borrower side,
lender side, business side and external) of loan repayment performance of borrowers. This conceptual
frame work was developed based on literatures reviewed under previous chapter and the variables selected
under this chapter of this study.

50
Figure 2. 2 Conceptual framework

Borrowers specific
factors
Age,
Gender,
marriage,
Family size,
Experience,
Education

Bank specific
factors
*Follow Up,
*Grace Period,
Business specific *Loan Size,
factors
Loan *Diversion,
*Income,
Repayment
*other business, Performances *Equity,
*business form,
*business sector *Collateral,
*Time Horizon,
*Kyc And
*Interest.

Other external
factors
*Whether condition
*Market condition

Source: Prepared by the researcher.

51
 CHAPTER FOUR

RESULTS AND DISCUSSIONS

Introduction

This chapter reports the results of the study conducted to identify the factors affecting loan repayment
performances. The data collected from survey questionnaire were carefully coded and checked for
consistency and prepared for analysis and interpretations. The analysis was performed using descriptive
statistics and with the help of stata. Therefore, this chapter presents analysis of the result and discussion to
achieve research objectives based on data obtained from the questionnaire respondents and interview made
with senior staffs and managers.

The first section of this chapter discusses the back ground of respondents followed by the result of
descriptive statistics of explanatory variables. In this part of analysis, factors of loan repayment
performances included under four categories (borrower related, business related, lender related and external
related) and other challenges of borrowers which affects repayment performance were analyzed by using
descriptive statics like percentiles, means, standard deviation and frequency. Besides, the second section
discusses the econometrics result of binary logistic & the analysis of significant variables.

4.1. Background Information of Respondents:

Questionnaire response rate and interview success rate: The questionnaire was distributed to a population
selected using stratified random sampling. Accordingly, there are two groups of population, performing
loans (borrowers) and nonperforming loans (borrowers), 110 and 40 respectively. Out of the one hundred
fifty questionnaires physically distributed to the target population, one hundred fifty usable responses were
collected. This represented a response rate of 100 percent and implies there is no unreturned questionnaire.

Out of the ten projected interviews, nine of them were successfully conducted, giving a success rate of 90
percent. The left interviews was unsuccessful due to the targeted interviewees were time constrained.
Despite this, the target population was fairly represented considering that managers who are relevant to the
study were interviewed. The results are shown in table 4.1 below.

52
Table 4. 1 Questionnaire and interview success rate

Target respondents Actual respondents Success rate

Questionnaire 150 150 100%
Interview 10 9 90%

Source: Own computation from primary data, 2017

4.2. Descriptive Analysis

The descriptive statistics for dependent and independent variables are presented below. The dependent
variable of the study is loan repayment performances and measured by performances of loans and/or
nonperformance of loans/impaired loans. Scholarly literatures presented under chapter two of this work,
classified the factors affecting loan repayment performances in to four broad categories.
Customer/borrower related factors (include age, gender, experience, family size and education level of the
borrower), lender institution related factors (includes loan size, time horizon, Collateral, equity, follow up
and grace period), business related factors (such as having other business, business form, business sector,
business income,) and finally eternal factors (like market problem, weather condition and others). The
detail descriptive and discussions were presented for every individual factors under all these groups.
As discussed under sampling techniques in previous chapter, the total population of the study was 273, out
of which 217 were performing loans and the remaining 56 were non-performing. Using stratified random
sampling techniques 150 population were selected for this study purpose and 110(73.3%) were performing
loans (according to loan status classification it includes pass, special mention and substandard) and the
remaining 40(26.7%) were nonperforming loans (doubtful and loss status). From this we have loan
repayment category of performing and Nonperforming/defaulters. In presenting the descriptive statics
analysis of the variable, in addition to percentage and frequencies, chi square test of independence allows
the researcher to determine whether variables are independent of each other or whether there is a pattern of
dependence between them. If there is dependence, the researcher can claim that the two variables have a
statistical relationship with each other. So, Pearson Chi-Square used in this study to indicate the level of
association of the independent variables with loan repayment.

53
4.2.1. Borrowers Related Factors

To begin with, borrower’s specific factors are the first most important factor related with personal
characteristics of the borrower and it’s important in determining performing and nonperforming loans
based on the personal behavior of the borrower. Under this research, gender, age, marital status, education
level, credit experience and family size were identified to evaluate their contribution in loan repayment
performances of the borrower. From among these variables, gender and credit experience were encoded as
dummy explanatory variables whereas age, marital status, education level and family size were encoded
and treated as categorical explanatory variables. So, now let us see all discrete and categorical variables
from loan repayment performances perspectives.

Gender of Borrower: There is a belief among many credit analysis/specialists that female are better payers
than male borrowers, taking into consideration of their being more entrepreneurial that results from
assuming more responsibilities in the internal affairs of a household. (Vigano, 1993) Also Khanker et al.
(1995) explains that loan recovery rates have been higher for women than for men.

Table 4.2 below shows the relationship between genders of borrowers with their repayment performances.
In terms of gender composition, from the total 150 survey population of the study the super majority of
them 137(91.3%) were male borrowers. The detail information is presented in the table below.

Table 4. 2 Gender in repayment performances.

Repayment performances
Explanatory Variable Performing Defaults Total X2 value
Male 99 38 137
72.3 27.7 100%
Female 11 2 13 X2 = 3.2603

84.6 15.4 100% P = 0.071

Total 110 40 1500
73.3 26.7 100%

Source; computed based on own survey, 2017

The table revealed that the number of female borrowers is much lesser than male. Accordingly, Out of the
total respondents, 137(91.3%) were male and 13(8.67%) were female. More specifically Out of the total
male sample respondents, 72.3%, and 27.7% of male respondents were non-defaulted and defaulters,

54
respectively. Whereas 84.6% and 15.4% of female respondents were non-defaulters and defaulters,
respectively.

This reveals that from their respective sex composition, females’ respondents were found having more
repayment performance than male respondents. Accordingly, Female borrowers generally delight in the
hard work ethics and the culture of financial discipline in repaying their loan and being committed to the
contractual agreement. The chi-square result also shows that the association between sex and loan
repayment is significant (X2=3.2603, at P=0.071) table 4.2. This indicates that female borrower had strong
positive relation to loan repayment performances.

Age: is one of the independent variables related with borrowers’ characteristics and determined loan
repayment performance of the borrowers. The survey results revealed that from total respondents 54(36%)
were at their young age/less than the age of 30 years, 62(41.33%) respondents were at their maturity
age/ranging 31-50 years and the reaming 34(22.67%)respondents were old/above 60 years old. The age
distribution of borrowers shows no significant difference as the number of one group is not that much
greater than the remaining.

Performance wise, from total young age borrowers 70.4% were the non-defaulters and 29.6 % were
defaulters. From mature age borrowers 67.7% respondents were the non-defaulters and 32.3% were
defaulters. From old age group of population 88.2% respondents were the non-defaulters and 11.8% were
defaulters.

55
Table 4. 3 Age and marital status of Borrowers and Loan Repayment

Variables Category Frequency Non- Defaulter Defaulter Total

N % N % N %
Age of Less than 30 years 54 38 70.4% 16 29.6% 54 36%
respondent 31-51 years old 62 42 67.7% 20 32.3% 62 41.3%
Above 60 years old 34 30 88.2% 4 11.8% 34 22.6%
Total 150 110 40 150 100%
Marital Single 61 43 70.5% 18 29.5 61 40.7
status
Married 81 61 75.3% 20 24.7 81 54
Divorced/widowed 8 6 75% 2 25 8 5.3%
Total 150 110 40 150 100
Have 85 68 80% 17 20% 85 100
Experience Credit experience(1) (56.7%)

No credit 65 42 64.6% 23 35.4% 65 100

experience(0) (43.3%)
Total 150 110 73.3% 40 26.7% 150 100

Source; computed based on own survey, 2017. N=number of respondents

The survey revealed that as the age of borrowers increases the probability of defaulting decreases and the
repayment performances increases. This emanates from the logic that as age increases the knowledge,
experience and skills of handling and administering businesses and thereby credit management enhances,
hence, the probability of defaulting decreases.

Marital status of borrowers: Regarding the marital status of the borrower’s, out of the 150 sample
borrowers, as depicted on table 4.2, 40.67%, 54%, and 5.3% respondents were single, married, and
divorced respectively. The single respondents were accounts for non-default and default 70.5% and 29.5%.
Married respondents were 75.3% and 24.7% non-defaulter and defaulter respectively. Among of Divorced
respondents, 75% non-defaulters and 25 defaulters. This indicated that compared to single borrowers
married and or divorced borrowers were better in paying their credit. The reason may related to the social
responsibility level of married and divorced borrowers. This result is same with the result of Josephat, et al.
(2013), Wongnaa1, et al, (2013)

56
Credit Experience: Another borrower related factor is credit experiences of respondents. The credit
experience of respondents shall be expressed in terms of years or months, hence it is a continuous variable,
but rearranged and encoded as dummy variable taking 1 if credit experience exists and o otherwise. Credit
experience helps borrowers in utilizing the loan for intended purpose and on how to prepare payments as
per the schedules.

Coming to survey results, From table 4.3 out of the total respondents majority of them 85 respondents or
56.7% of total borrowers have credit experience and 68(80%) of these experienced borrowers were paying
their loan as per the schedule of the contract, while the remaining 17 respondents or 20% were defaulted.
On the other hand, 65 respondents or 43.3% were completely new for inexperienced and did not have any
experience. Out of such borrowers 42 respondents, i.e. 64.6% were performing loans and 23 respondents
i.e. 35.4% were nonperforming loans. So, experienced borrowers are better in repaying their loan than
those who didn’t have any experience.

However, the chi-square result shows that the association between marital status and experience and the
dependent variable loan repayment is insignificant. This indicates that being either in any marital status
doesn’t statistically determine loan repayment.

Education: The survey on the educational characteristics of the borrowers shows that 22 (14.67% )of the
borrowers didn’t attended any formal education, some 53/35.33% borrower attended lower level/primary
education, the rest 48(32%) and 27(18%) of the borrowers attended secondary school or tertiary level and
joined college or university respectively as shown in table 4.4.

The loan repayment performance of the borrowers relative to their educational level as shown on table 4.4
showed that among 22 borrowers who do not have formal education 40.9% of them were repaying their
loan successfully whereas the remaining 59.1% were defaulters. From borrowers whose educational level is
at primary level, majority of them 71.7% repaid their loan duly and 28.3% of them defaulted. From
48(32%) borrowers who attended tertiary education, 81.2% of them were non-defaulters while 18.8% were
defaulters. Finally, from 27(18%) borrowers who attended college education and above, 88.9% were non-
defaulters and the remaining 11.1% were defaulters. This result indicates as education level increases, the
probability of defaulting decreases and vice versa. This result contradicts the result of Yacob (2014) that
the clients with lower education have fewer financial options and thus they would improve on their loan
repayment performance in order not to lose their only formal source of credit.

57
The chi-square result also shows the presence of strong and significant association between educational
level and dependent variable at 1% significance level (X2=66.5646, at P=000). This results of chi square
revealed that level of education has strong and significant relationship with the dependent variable.

Table 4. 4 Educational Qualification and family size of Borrowers and Loan Repayment

Variables Categories Non-Default Default Total Chi square

N % N % N % X2=66.5646
No formal education 9 40.9% 13 59.1% 22 14.67% P=0.000
Educational Primary education(1-8) 38 71.7% 15 28.3% 53 35.3%
Qualification secondary education(9-12) 39 81.2% 9 518.8% 48 32%
College and above 24 88.9% 3 11.1% 27 18%
Total 110 73.3% 40 26.7% 150 100.0%
Small 46 76.7% 14 26.3% 60 40% X2=6.8306
P=0.033
Family size Medium 43 81.1% 10 18.9% 53 35.3%

Large 21 56.8% 16 43.2% 37 24.7%

Total 110 73.3% 40 26.7% 150 100%

Source; computed based on own survey, 2017. N= number of respondents

Family size: In this study family size is used to express the number of dependents on the borrower.
Accordingly, the influence of family size on repayment performances of borrowers is assessed as follows;
out of the total sample borrowers, 60 of them have small family size and the repayment performances of
small size household is 26.3% defaulted and 76.7% non-defaulted. The other 53 borrowers were having
medium size family, out of which 18.9% of them were defaulted and the remaining 81.1% non-defaulted.
Lastly, 37 borrowers were responded having large family from which 43.2% defaulted and 56.8% non-
defaulted. The statistical survey from the above table showed as family size increases the likelihood of
being default increase and vice versa.
The regression result of chi-square shows presence of strong and significant relationship between family
size level and dependent variable at 5% significance level (X2=66.5646, at P=000). This results of chi
square revealed that increase or decrease in family size has strong and significant relationship with the
dependent variable.

58
4.2.2. Business Related Factors

Respondents were found to engage in various business sectors. For this study purpose the most important
businesses are categorized in to three sectors namely agricultural type businesses, service type business
sector and industry type business. The other source of factors that affects loan repayment performances
emanates from the business itself. That is, among the group of factors affecting loan repayment
performances; business related factors were other important factors in determining what factors were
affecting performing and non-performing loans. Thus, in this research business form, business sector, other
business and business income are selected as business related factors. Out of which business form and
business sector were encoded as categorical explanatory variables whereas other business and business
income were encoded and treated as dummy explanatory variables of loan repayment performance.

Business sector: this variable evaluates which economic sector from agricultural, service and industrial
sector of the economy affects loan repayment performances of the borrowers. As discussed under chapter
two of this study, literatures show that agricultural investments are vulnerable to different natural and
manmade problems than other projects.

With respect to the business sector on which loans were invested out of the total 150 sample populations
majority of them were agricultural projects. Located in the south western direction of the country
Development Bank of Ethiopia Jimma district and surrounding areas were conducive for agricultural
investments. As such, out of 150 respondents participated in the questionnaire of this research, 108 or 72%
of them were invested on agricultural projects. Compared to the remaining two economic sectors,
agricultural investments are dominated the loans of the district and the role of the sector determines the
performances of the district as a whole.

The repayment performances of agricultural loans as shown in table 4.5, below 72 borrowers or 66.7% of
sample respondents took the loan to engage in agricultural type business, whereas, 32 or 21.3% invested on
service investments and 10 borrowers or 6.7% took the loan to invest on industrial sector of the economy.
Now, out agricultural loans 66.7% were performing and of them or 33.3% of them were nonperforming.
On the other hand, from among loans invested on service sector of the economy, 90.6% loans were
performing and 9.4% were nonperforming loans. The loans invested on industry sector indicate that 90% or
9 performing and the remaining 10% or 1 loan was non-performing loans. As the table depicted, the
agriculture sector showed that the percentage of default higher than service and industry sector. In same

59
line research, Besley and Coate (1995) reveal that agricultural loans were risky and the probability of high
loan default.

Table 4. 5 Business sector and business form against repayment

Explanatory Loan repayment performances Chi square

Variable Category Performing Non-performing Frequency
Agriculture 72 36 108
X2 = 10.0736
66.7% 33.3% 100%
Business sector Service 29 3 32 P = 0.006*
90.6% 9.4% 100%
Industry 9 1 10
90% 10% 100%
Total 110 40 150
73.3% 26.7% 100%
Sole owner 62 7 69 X2 = 25.2614
89.9% 10.1% 100% P = 0.000*
Business
PLC 33 30 63
Form
52.4% 47.6% 100%
SC 15 3 18
83.3% 16.7% 100%
Total 110 40 150
73.3% 26.7% 100%

Source: own survey of data, 2017. * = Significant at 1%

The results of the survey as depicted in the above table showed that agricultural sector as compared to
others sectors has contributed 80% of nonperforming loans of the bank. The main reasons of lion share
contributions of agricultural investments in the nonperforming loans emanated from the nature of rain fed
agriculture. Rain fed agricultures is basically dependent on the natural factors like rain, drought and many
others. Statistically, chi-square also confirms the presence of strong and a significant association between
business sector and dependent variable at 1% level of significance (X2 = 10.0736, at P=0.006). This shows
that business sectors, especially agricultural loans have strong negative relationship with that of dependent
variable.

Business form: Ethiopian commercial law recognizes different types of business form having their unique
features. Share companies, Private limited companies, cooperatives, and different kinds of partnerships are

60
widely used forms of business in our country. Among many different features of business forms, one
distinguishing feature is the level of responsibilities the owners share in cases of debt recovery.

Borrowers of DBE, were categorized under three business forms; sole proprietorships, private limited
companies and share companies. Based on the results of the survey indicated in the above table, 69
borrowers or 46 percent of the sample population were sole owners/ private borrowers of their business,
while 63(42) borrowers were Private Limited Company(PLC) and the remaining 18 or 12% of borrowers
were Share Companies.

In terms of their Loan repayment performances, from total private borrowers/sole owners as shown in the
table, 89.9% of them were performing loans and 10.1% were nonperforming. In cases of Private Limited
company borrowers, 52.4% were performing loans while the remaining 47.6% were nonperforming loans.
Lastly 83.3 % and 16.75 were performing and non-performing loans respectively in case of Share
Company.

Maximum numbers of performing loans were found in private borrowers/ sole owners’ business form and
service and industry business sector. Based on row relative frequency of business forms and business
sectors, maximum numbers of defaulters were found in private limited company business form and
agriculture business sector, respectively. Maximum numbers of defaulters were from private limited
company business form and service business sector. The researcher believes that the reason why greater
numbers of defaulters are from private limited company is related to the level of responsibilities the
shareholders bear in cases of failure to repay the debt of the company. This result is the same with Arega
Seyoum(2016). The regression result of chi square also confirms the presence of strong and a significant
association between business form and dependent variable at 1% level of significance (X2= 25.2614 at P =
0.000). This shows that business form very specifically Private limited Company’s loans have strong
negative relationship with that of dependent variable.

Having another business: this variable is designed to evaluate the exposure, awareness and familiarity of
borrowers to operate the current business and whether such exposure and experience helps them in
repayment performances of their loan.

Accordingly, the survey result of the study as depicted on the table 4.6, out of total sample borrowers only
66 borrowers or 44% have other business and 88 borrowers or 56% of them didn’t have any other business
prior to the current project/business established with help of Development Bank of Ethiopia. The

61
repayment performance statistics as tabulated in the table below shows that out of those borrowers having
other business 50 borrower’s or 75.8% have performing loans and the left 16 borrowers or 24.2% were
defaulted. Likewise out of those borrowers having no other business 60 borrowers or 71.4% were paying
their loan as per the schedule of their contract with the lending bank, while the remaining 24 borrowers or
28.6% were defaulted. Obviously, this result shows that having other business contributes to repay the loan
as per the schedule of the contract whereas having no other business contributes to fail to repay the loan.
Similarly, the chi-square result reveals the strong and significant association between having other business
and loan repayment at significant level 5% (X2=25.8248, at P = 000). This implies that having business
experiences enhances the probability to repay loan more than no business experience.

Table 4. 6 other business and income vs. loan repayment performances

Explanatory Frequency Loan repayment performances

variable
Performing Nonperforming total
Loan Loan
Have Other 66 50 16 66 X2= 25.8248
business(1) 44% 75.8% 24.2% 100
Have no other 84 60 24 84 P = 0.000*
business(0) 56% 71.4 28.6 100
Total 150 110 40 150
73.3 26.7 100
Income generated 61 43 18 61 X2 = 0.4219
Sufficient(1) 40.7% 70.5% 29.5 100
P = 0.516
Not sufficient(0) 89 67 22 89
59.3% 75.3 24.7 100
Total 110 40 150
73.3 26.7 100%
Source: own computation of data, 2017. * Significant at 1%

Income/ profit generated from the project is a continuous variable, in this study it was rearranged as
dummy variable. It is generally believed that if a business generated sufficient income/profit, the
probability of repayment performance is high and vice versa. So, based on the results of the survey from
the above table 4.6, 67 borrowers or 44.7% answered their business generated sufficient profit. The left 83
borrowers or 55.3% answered their business did not rewarded them with sufficient income.

62
Accordingly, among the respondents that generated sufficient income 67 borrowers, 70.5 %(43) were
performing while 29.5% (18) were defaulters. From the projects that fail to generate sufficient income 89
borrowers, 75.3% (67) were performing/ non-defaulted and the remaining 24.7% (22) were defaulters. The
table showed that projects that have not generated sufficient income/profit expected to repay the loan from
the other source of finance otherwise the probability of loan to default is high. Despite most respondents
who answered this questionnaire, there were some respondents who left it blank space but they responded.
The respondents, who got profit from their loan, were high loan repayments rates. The result is the same as
Stephen (2012); and Wongnaa and Awunyo (2013). However, the chi-square result shows that the
association between income and loan repayment is insignificant (X2= .4219, at P=.516) table 4.6. This
indicates that being generating income by itself doesn’t determine loan repayment performances.

4.2.3. Institutional Related Factors

Development Bank of Ethiopia is a public financial institution that has been financing for viable projects in
line with country’s developmental policy and programs. The bank has its own policy and procedural
manual for the flow of work that can be used during providing credit services on those feasible projects.
There are different factors affecting loan repayment performances while using Policy and procedures of the
bank. Among institutional related factors, loan size, loan diversion, sufficient equity, grace period, follow
up, collateral, interest rate, KYC and time horizon problems were taken as factors of loan repayment
performance.

From these variables, loan size and time horizons were encoded as categorical variables and the remaining
were encoded and treated as dummy variables having their own features. Now, let’s discuss the results and
present the survey of the data.

Loan size: this is the amount of money the bank permitted for the project and whether the size of such loan
determines the repayment performances of loan is assessed as follows. Loan size is a continuous variable to
be expressed in terms of currency or figure but for the purpose of this study it is categorized and coded as
(1= 1million-10 million, 2=11 million to 20 million, 3= 21million and above).

From the total respondents 40 borrowers or 26.67% borrowed from 1-10 million birr, 58 borrowers or
38.67% were borrowed from 11-20 million birr and the remaining 52 borrowers or 34.67% were borrowed
above 21 million birr. When it comes to repayment performances, out of those who borrowed 1-10 million
birr, 27 borrowers, i.e. 67.5% were paying their loan as per the terms of their contract with the bank, and

63
the remaining 13 borrowers, i.e. 32.5% were defaulted. The next respondents were those who borrowed
from 11-20 million birr, out of 58 borrowers, 42 of them i.e. 72.4% were performing and the remaining 16
borrowers, i.e. 27.6% were defaulted. Lastly, from 52 borrowers who borrowed 21 million and/or above,
41(78.8%) were performing loans and the left 11 borrowers or 21.2% were nonperforming loans.

Generally, the results of this statistical analysis show that when loan size increases, the probability of
default decreased. It can be the fact that an increase in the loan size, borrowers can do their project in a
wide range with the inclusion of quality and quantity of products. Therefore, their project can generate
huge revenue and can repay the due amount of loan on time. The chi-square result reveals the strong and
significant association between having loan size and loan repayment at significant level 5% (X2=9.1793, at
P = 0.010). This implies that getting sufficient loan amount contributes to repayment performances. This is
the same as Ali AL-Sharafat, et al. (2013) that the volume of loans borrowed the most important factor and
had a positive effect on the repayment performance of the investigated agency. This is also the same as
(Ifeanyi and Blessing, 2012).

Table 4. 7 loan size and time horizon against repayment

Explanatory Category Loan repayment performances

variables
Frequency Performing Nonperforming Mean
1-10million 40 27 13
26.67% 67.5% 32.5% X2 = 9.1793
Loan size 11- 58 42 16 P = 0.010*
20million 38.67% 72.4% 27.6% 2.08
≥ 21 52 41 11
million 34.67% 78.8 21.2%
Total 150 110 40
100 73.3 26.7
Time Timely 39 37 2
horizon 26% 94.9% 5.1%
Delayed 79 68 11 1.953333 X2 =70.3838
52.7% 86.1% 13.9% Pr = 0.000**
Elongated 32 5 27
21.3% 15.6% 84.4%

Source: Output from Survey Data, 2017. ** Significant at 1% and significant at %5

Time horizon: this variable assess whether the lending institution deliver its services (includes services
like credit team processes, appraisal and approval) within a shortest possible time or otherwise. Time

64
horizon is a continuous variable, but rearranged as categorical variable and encoded as 1 for timely
response 2 for delayed services and 3 for elongated services. According to Johnson and Rogaly (1997),
timeliness of loan disbursement is important when loans are used for seasonal activities such as agriculture.
It’s argued that complicated appraisal and approval procedures, which might delay disbursement, influence
a program of seasonal loans for farmers who use to buy inputs. Further they noted that this could in turn
worsen the prospects of repayment by diverting loan to non-intended purpose.

The survey of data as stipulated in the table above shows that 26% of total population was served within a
reasonable time expected from customers, while 52.7% get delayed service and the remaining 21.3 got
service after such length dalliance. Out of timely served borrowers, 94.9% were performing their loan,
while the 5.1% failed. On the other hand out of borrowers who got service (especially disbursement and
approval) after some dalliance, 86.1% were repaying their loan as per the requirements of the bank and
13.9% were defaulted. Finally among those borrowers who got service after lengthy dalliance, 15.6% were
performing and 84.4% were nonperforming. The result from the survey is also backed by statistical chi
square. Hence, chi-square result reveals the strong and significant association between time horizon and
loan repayment at significant level 1% (X2=70.3838, at P = 0.000).

The statistics results in this survey indicate the fact that getting service within the shortest possible time
contributes to well performance and vice versa.

Collateral: this is a continuous variable, but arranged as dummy variable taking 0 if the collateral is
sufficient in cases of failure and 1 if the collateral is not sufficient. Lending institutions needs grantee for
the money they provide for their customers. The value of such collateral is believed to be more or equal to
the amount of money permitted for the borrower. The case is little different in development bank of
Ethiopia, because the loan policy of the bank do not require for collateral that is more or equal to the
amount of money lent to customers. The only collateral of the bank is the business/project itself, which in
some cases particularly in agricultural projects were open fields or empty store.

The table below revealed that, 54.7% respondents believes that the bank has no sufficient mortgage in case
the loan defaulted, if the banks go for recovery and the remaining 45.3% believes the bank has sufficient
collateral. In terms of performances, out of those who believe the bank has no sufficient collateral, 69.5%
of them were performing their duty of repaying their loan and 30.5% were non-performing. From those
who believe the bank has sufficient collateral, 77.9% were no defaulted while 22.1 were defaulted.

65
Equity: is a continuous variable, but encoded as dummy taking 1 where the equity is greater than 30%
percent of total investment of the project and 0 when the equity amount is less or equal to 30% percent of
total investment cost. Accordingly, from the table below, 76% of equity contribution is less or equal to
thirty percent of the total investment cost and 24% were more than thirty percent of the investment.

Further the table revealed that, out of 114 borrowers or 76 %whose equity contribution is less or equal to
thirty percent, 66.7% were performing and 33.3 were nonperforming loan. And from those 36 borrowers or
24% whose equity contribution is greater than thirty percent, 94.4% were performing loans and the left 5.6
were non-performing loans. This indicates that the probability of defaulting decreases as equity
contribution of borrower increases.

Table 4. 8 collateral, equity and diversion against repayments.

Explanatory Loan repayment performances

variables
Frequency Performing Nonperforming Total
Sufficient 68 53 15 68 X2= 1.3636
Collateral(1) 45.3% 77.9% 22.1% 100
Insufficient 82 57 25 82 P = 0.243
Collateral(0) 54.7% 69.5% 30.5% 100
Total 150 110 40 150
100 73.3% 26.7% 100
Equity>30 36 34 2 36 X2 = 0.8485
percent(1) 24% 94.4% 5.6% 100 P = 0.357
Equity≤30 114 76 38 114
percent(0) 76% 66.7% 33.3% 100
Total 150 110 40 150
100 73.3% 26.7% 100
Diverted 8 3 5 8 X2= 4.7944
loan(1) 5.3% 37.5% 62.5% 100 P = 0.029*
Not diverted 142 107 35 142
(0) 94.7% 75.4% 24.6% 100
Total 150 110 40 150
100 73.3% 26.7% 100
Source: Output from Survey Data, 2017. * Significant at 5%

Regarding loan diversion, out of the total sample respondents, 94.7% borrowers were not diverted their
loan to other business, but 5.3% respondents diverted their loan from intended business to some other
purposes. Out of those who diverted respondents 62.5% was defaulted and 37.5% was performing. And out
of those who did not diverted 75.4% were performing and the remaining 24.6% was defaulted. The

66
statistical result of this survey shows that, loan diversion contributes to the probability of defaulting and
vice versa. Statistically, chi-square result also confirms that there is strong and significance association
between loan diversion and loan repayment at 5% (X2= 4.7944, at P =0.029) could find from Table 4.8.

Table 4. 9 interest rate, grace period, follow up and KYC with repayment

Explanatory Loan repayment performances

variables
Frequency Performing Nonperforming Total
Given grace period(1) 59 44 15 59 X2 = 0.0771
39.3% 74.6% 25.4% 100% Pr = 0.781
Not given grace period 91 66 25 91
(0) 60.7% 72.5% 27.5% 100%
Total 1500 110 40 150
100 73.3 26.7 100%
Interest rate affected (1) 86 61 25 86 X2 = 0.6001
57.3% 70.9% 29.1% 100% Pr = 0.439
Interest rate not 64 49 15 64
affected(0) 42.7% 76.6% 23.4% 100%
Total 150 110 40 150
100 73.3 26.7 100%
Proper follow up (1) 41 32 9 41
27.3% 78% 22% 100%
No follow up(0) 109 8 31 109 X2 = 9.4665
72.7% 71.6% 28.4% 100% Pr = 0.002**
Total 150 110 40 150
100 73.3 26.7 100%
Proper KYC (1) 70 57 13 70 X2= 4.4836
46.7% 81.4% 18.6% 100% Pr = 0.034*
No proper KYC (0) 80 53 27 80
53.3% 66.2% 33.8% 100%
Total 150 110 40 150
100 73.3% 26.7% 100%
Source: Output from Survey Data, 2017. *Significant at 5% ** significant at 1%

As read from the table, 60.7% respondents said they were not given sufficient grace period so as to begin
operation with full capacity and establish itself strongly. Only 39.3% respondents answered their business
is given grace period. The business/projects that had no grace period contribute for default and non-default
loan was 27.5% and 72.5% respectively. In the other hand, projects that was given grace period contributes
for default and non-default loan was 25.4% and 74.6% respectively. The table below demonstrates that
when the projects have no grace period, projects faced repayment problem when the loan due later due to

67
insufficient time for implementation. But the chi square didn’t show the existence of strong and significant
relationship between grace period and the dependent variable.

In the same token, increase in lending interest rate also has affected repayment performances. Accordingly,
57.3% respondents answered their repayment capability was affected by increase in the lending interest rate
and 42.8% respondents said such increase in interest rate did not affected their repayment capacity. Out of
those affected by increase in lending interest rate change, 70.9% were performing their repayment duty as
per the schedule arranged in their loan contract, while 29.1% failed to carry on their duty of repayment as
per the schedule. From those who did not affected by change in lending interest rate, 76.6% were
performing their duty while 23.4% were nonperforming. The statistical analysis of chi square didn’t show
the presence of any strong and significant association between increase in interest rate and dependent
variable.

With regard to follow up, 72.7% of total sample population responded that there is no proper follow-up to
their business and some 27.3% responded the bank was conducted proper follow up with their
business/project. As shown in the above table, out of 72.7% respondents whose projects did not have
appropriate follow up, 71.6% were performing loans and 28.4% were defaulted. Similarly, out of those
projects that appropriate follow up was conducted, 78% were performing and the left 22% were defaulters.
Therefore there is a significant statistical difference between defaulters and non-defaulters in these
averages, at 1% significance level (Table 4.9). This indicates that the continuous follow up of borrowers
reminds them to pay attention toward their business and enables to increase their perception of
responsibility toward loan repayment

Know your customer (KYC) also known as due diligence is a screening stage sorting of credit worthy
customers. Reading of the table above, 53.3% respondents said such screening is not properly conducted
while the left 46.7% answered their business/project have appropriate KYC study. Out of businesses that
KYC was not conducted properly, 66.2% were performing loans while the remaining 33.8% were
defaulters. Likewise, out of those KYC properly conducted, 81.4% were performing and 18.6% were
defaulters. Statistically, chi-square result also confirms that there is strong and significance association
between Kyc and loan repayment at 5% (X2 = 4.4836 P = 0.034) could find from Table 4.9. This reveals
that the due diligence/KYC plays very important role in repayment of loan.

68
4.2.4. External Related Factors

External factors are factors of repayment performance which are beyond borrower, lending institution and
the business itself. Most of the time these factors are unpredictable and uncontrollable by these bodies, thus
it makes difficult in case of decision making in different institutions. In this study weather condition and
market condition were selected as external factors of loan repayment performance on development Bank of
Ethiopia, Jimma District customers. Since the majority of projects financed in this District were
agricultural, the possibility of being affected by such external factors is natural.

Therefore, the table below has provided number of respondents who were challenged or were not
challenged due to bad weather condition during running their business. Accordingly, 88 respondents or
58.7% were affected by bad weather condition and 41.3% were not affected by such conditions.

Regarding the repayment performances, out of 88(58.7%) respondents that were affected by bad weather
conditions, 57(64.8%) were performing loans and 31(35.2) were defaulters. Out of those 62(41.3%)
respondents who were not affected by bad weather conditions, 53(85.5%) were performing loans and the
remaining 9(14.5%) were defaulters. The results of chi square didn’t show the presence of significant
relationship between bad weather condition and dependent variable.

Table 4. 10 whether condition and market condition against repayment.

Explanatory Frequency Loan repayment performances

variable
Performing Loan Nonperforming Loan Total
Bad 88 57 31 88 X2 = 0.1555
weather(1) 58.7% 64.8% 35.2% 100%
No whether 62 53 9 62 P = 0.693
problem(0) 41.3% 85.5% 14.5% 100%
Total 150 110 40 150
100 73.3% 26.7% 100%
Market 75 52 23 75 X2 = 1.2308
problem(1) 50% 69.3% 30.7% 100%
P = 0.267
No market 75 58 17 75
problem (0) 50% 77.3% 22.7% 100%
Total 150 110 40 150
100 73.3% 26.7% 100%
Source: Output from Survey Data, 2017

69
As regards to market conditions, 75 (50%) of sample respondents answered that there were a problem of
poor market condition especially selling at lower price than expected one and the main reason of
unfavorable market condition were due to international price fluctuation and less demander for the product
and high amount of supplies and the remaining 75(50%) were not affected by market fluctuation.
Therefore, out of 75 respondents affected by market conditions, 52(69.3%) were performing loans and
23(30.7%) were nonperforming. And out of 75 respondents that were not affected by market situations,
58(77.3%) were performing loans and the remaining 17(22.7%) were defaulted. The chi square also didn’t
recognize the existence of significant relationship between market conditions and the dependent variable.

4.2.5. Other Major Problems

In addition to the lists of variables discussed above, there are different challenges that hinders the
repayment performances of loans by borrowers. So, respondents have stipulated different reasons, factors
and challenges they faced and what they think was affecting their repayment performances. Here are some
of them; majority of the responders sited that unavailability of skilled and unskilled labor, price fluctuation
of the product, poor quality of the product and land overlap were other challenges of sampled customers.

Availability of skilled and unskilled labor: the establishment of projects especially agricultural farming
projects needs the availability of skilled man power both for direct agricultural activities and administrative
works. However, even if casual workers particularly during the time of harvesting and sowing are very
important in farming, majority of the farm found in desert area were challenged by this factor.

Pricing and quality of product: the price of cotton, sesame and coffee product is largely degree depended
on demanded quality, number of suppliers, production, market condition etc. (the price maker is the market
itself). Thus, most of agricultural producers; especially cotton, coffee and sesame producers were
challenged by price fluctuation of the market. Due to the presence of new entrants, unexpected weather
condition and uncontrollable diseases and pest which is common for all agricultural products, had affected
quality of their produce and hence production.

Land overlap: especially, projects found in Gambella region were been challenging by this problem
because of using unreliable way of land providing to those potential investors from concerned institutions.
Thus, based on citation from majority of borrowers even if it was not significant factor in econometrical
analysis, land overlap problem was one of the causes for lag of loan process hence weak project
performance leads to delinquency.

70
4.3. Econometric Analysis
In contrast to descriptive analysis, an econometric analysis or statistical analysis is the method of data
analysis where mainly focus on coefficients, R-square, chi-square, standard error, tests, log likelihood ratio
etc., which can be done using different software’s such as STATA, SPSS and others. In this study STATA
version 12 was adopted for the analysis of binary logistic regression coefficients and different tests. So,
before running the binary logistic regression, the explanatory variables were checked using the following
tests.

4.3.1. Model Tests

According to (Gujarati, 1995), for the econometric estimation to bring about best, unbiased/reliable and
consistent result, it has to fulfill the basic linear classical assumptions. The basic assumptions include:
linearity in parameters of the regression model, for given explanatory variables the mean value and the
variance of the disturbance term (Ui) is zero and constant (homoscedastic) respectively, there is no
correlation in the disturbances, no correlation between the regressors and the disturbance term, no exact
linear relationship (multicollinearity) in the regressors and the stochastic (disturbance) term Ui is normally
distributed. Naturally, therefore, if these assumptions do not hold well on what so ever account, the
estimators derived may not be efficient. Based on the type of data (cross sectional type) used in this study,
the most important tests such as heteroscedasticity, multicollinearity are conducted and the appropriate
remedies were taken.

4.3.1.1. Test for Multicollinearity Assumption

If an independent variable has exact linear combination with the other independent variables, then we say
the model suffers from perfect collinearity. This assumption is concerned with the relationships which exist
between explanatory variables. In the construction of an econometric model, it may happen that two or
more variables giving rise to the same piece of information are included, that is, we may have redundant
information or unnecessarily included related variables and such situation is called a multicollinearity
(MC) problem.

One of the assumptions of the CLRM is that there is no exact linear relationship exists between any of the
explanatory variables. When this assumption is violated, we speak of perfect MC. If all explanatory
variables are uncorrelated with each other, we speak of absence of MC. Multicollinearity usually exists in

71
most applications. Therefore, the question is not whether it is present or not; it is a question of degree! In
addition, MC is not a statistical problem; it is a data (sample) problem. Therefore, we do not test for MC;
but measure its degree in any particular sample (using some rules of thumb). There is no consistent
argument on the level of correlation that causes multicollinearity.

There are two measures that are often suggested to test the presence of multicolinearity. These are:
Variance Inflation Factor (VIF) for association among the continuous explanatory variables and
contingency coefficients for dummy variables. The technique of variance inflation factor (VIF) was
employed to detect the problem of multicolinearity among the continuous variables. According to Gujarati
(2003), VIF can be defined as:

Where, VIF is variance inflation factor, Ri2 is multiple correlation coefficient and is explanatory variables
Xi is explanatory variables. The result of VIF test is annex as Annex 3 at the end of this paper.

Table 4. 11 VIF of the Explanatory Variables used in the study

Variable | VIF 1/VIF
-------------+----------------------
Education | 1.71 0.583593
Kyc | 1.44 0.693435
Loansize | 1.44 0.695333
Timehorizon | 1.41 0.711450
Graceperiod | 1.34 0.747682
Otherbusin~s | 1.29 0.772964
Businessse~r | 1.26 0.791117
Income | 1.26 0.791368
Marriage | 1.24 0.804159
Collateral | 1.23 0.813179
Gender | 1.22 0.817818
Interest | 1.22 0.818259
Businessform | 1.22 0.822914
Age | 1.20 0.830964
Households~e | 1.19 0.839626
Whether | 1.16 0.859426
Followup | 1.13 0.883061
Equity | 1.13 0.884816
Market | 1.13 0.887345
Experience | 1.12 0.891935
Diversion | 1.09 0.913764
-------------+----------------------
Mean VIF | 1.26

72
Larger value of VIF shows co-linearity across variables, thus if VIF exceeds 10 indicates that there was
multicolinearity within continuous variables. The result shows that no categorical and dummy explanatory
variables which have variance inflation factor near to 10 i.e. the maximum value among those dummy
explanatory variables was 1.50 in case of loan size while 1.26 was an average VIF of all variables.

In addition to VIF, contingency coefficients were computed to check the existence of multicolinearity
problem among the discrete explanatory variables. A contingency coefficient is a measure of the degree of
relationship, association of dependence among variables included in the study. The contingency coefficient
is calculated as follows (Garson, 2008 cited in Fikirte, 2011):

Where: C = contingency coefficients, X2 = the value of Chi-square, N = total sample size.

The decision rule for contingency coefficient is the larger the value of this coefficient, the greater the
degree of association. The maximum value of the coefficient is never greater than 1. The results of
contingency coefficients reveal that there was no serious problem of association among the discrete
variables.

Table 4. 12 Correlation matrix of coefficients of regress model

e(V) | Gender Age Marriage Educat~n Househ~e Experi~e Otherb~s Busine~m Busine~r Income Loanam~t
-------------+-------------------------------------------------------------------------------------------------------------
-
Gender | 1.0000
Age | 0.0123 1.0000
Marriage | 0.0119 0.1132 1.0000
Education | -0.1016 0.0805 0.1678 1.0000
Households~e | -0.1766 -0.2098 -0.1919 -0.0241 1.0000
Experience | 0.0499 0.0796 -0.0379 0.0028 0.0430 1.0000
Otherbusin~s | -0.0107 -0.1843 -0.0094 -0.1512 0.0063 0.0688 1.0000
Businessform | -0.1868 0.0715 -0.1180 0.0908 0.0261 0.1947 0.0404 1.0000
Businessse~r | 0.0417 -0.0780 -0.0571 -0.2016 0.0607 -0.0116 -0.0415 -0.0690 1.0000
Income | 0.0888 -0.1700 -0.0591 -0.2683 0.0464 0.0007 0.0406 -0.1400 0.1151 1.0000
Loanamount | -0.0001 0.0459 -0.2583 -0.0890 -0.0358 -0.0142 -0.0166 0.1724 -0.3131 -0.1162 1.0000
Diversion | -0.0680 -0.1369 -0.0075 0.1029 -0.0013 -0.0530 -0.0022 0.0074 0.0265 -0.0277 0.0207
Equity | 0.0886 -0.0742 -0.0693 -0.1436 0.1023 -0.0128 -0.0264 -0.1039 0.0921 0.1862 -0.0282
Graceperiod | 0.0291 0.1231 -0.1207 -0.0873 0.0859 0.1042 -0.2372 0.0207 0.0983 -0.0490 -0.1448
Followup | -0.0379 0.0030 0.0315 -0.0572 -0.0324 -0.0139 0.0240 0.1050 -0.0874 -0.1106 -0.0017
Collateral | -0.0424 -0.0762 -0.0517 -0.0296 0.0273 -0.0512 -0.0909 0.0981 0.0643 0.0451 0.0303
Interest | -0.0943 -0.0327 0.1511 -0.1199 0.0053 -0.0717 0.0976 -0.0510 0.0308 -0.0127 -0.2071
Kyc | 0.0648 -0.0523 -0.0289 -0.2337 -0.0726 -0.1079 -0.1155 0.0374 0.0033 0.0883 0.0891
Timehorizon | -0.1318 -0.0355 0.0771 0.3210 -0.0571 0.0919 0.0993 -0.0321 -0.0122 0.0936 -0.1239
Market | 0.1758 -0.0265 -0.1309 -0.0591 0.0704 0.0601 0.0460 -0.0052 0.0902 0.0317 0.0671
Whether | 0.1610 0.0824 0.0861 0.0529 -0.1941 0.0710 0.0632 -0.0234 -0.0334 0.0452 0.0125
_cons | -0.2784 -0.2898 -0.2912 -0.4279 -0.0671 -0.2707 -0.0251 -0.3133 -0.0877 0.0123 -0.1038

e(V) | Divers~n Equity Gracep~d Followup Collat~l Interest Kyc Timeho~n Market Whether _cons
-------------+-------------------------------------------------------------------------------------------------------------
-
Diversion | 1.0000

73
 Equity | -0.0204 1.0000
Graceperiod | 0.0586 0.0188 1.0000
Followup | 0.1182 -0.1482 -0.0211 1.0000
Collateral | -0.0273 -0.0439 0.0848 -0.0563 1.0000
Interest | -0.0292 0.0409 0.0072 -0.0076 -0.0034 1.0000
Kyc | -0.0235 0.0079 -0.1584 -0.0304 0.3433 0.1955 1.0000
Timehorizon | -0.0044 -0.1111 0.0417 0.0936 0.0722 -0.0777 -0.1131 1.0000
Market | -0.0649 0.0547 -0.0336 -0.0709 -0.0230 0.0606 0.0034 -0.1829 1.0000
Whether | -0.0097 0.1174 -0.0986 0.0323 -0.0531 0.0927 0.0748 0.1305 -0.0049 1.0000
_cons | 0.0029 0.0038 -0.0223 -0.0419 -0.1741 -0.0382 -0.0546 -0.4358 -0.1285 -0.3015 1.0000

4.3.1.2. Measures of Goodness of Fit

The conventional measure of goodness of fit, R2, is not particularly meaningful in binary regress and
models. A measure similar to R2, called pseudo R2, is available, and also ranges between 0 and 1(Gujarati,
2004).

According to Kibrom (2010), the use of conventional R2 for goodness of fit when the dependent variable
takes either 1 or 0 is not appropriate. “A summary measure used similar to the conventional R2 that have
been suggested for models with qualitative dependent variable is pseudo R2. It should be noted, however,
that in binary regress and models, goodness of fit is of secondary importance. What matters are the
expected signs of the regression coefficients and their statistical and/or practical significance? As noted
previously more meaningful interpretation is in terms of odds, which are obtained by taking the antilog of
the various slope coefficients” (Gujarati, 2004, p .605-606). Thus for this study, the model pseudo R2 is
81% or 0.81 (as it is depicted in the logistic regression). This result indicates that, the logit model explained
about 81% of the variation and it lies in the [0, 1] interval.

4.3.1.3. Test for Normality assumption

Normality assumption (ut ~ N (0, σ2) state that a normal distribution is not skewed and is defined to have a
coefficient of kurtosis 3. Bera-Jarque formalizes this by testing the residuals for normality and testing
whether the coefficient of skewedness and kurtosis are zero and three respectively. Skewness measures the
extent to which a distribution is not symmetric about its mean value and kurtosis measures how fat the tails
of the distribution are. To make sure that this assumption is valid or not, the residuals generated out of the
regression model is plotted against the fitted values of the dependent variables. If this curve is like bell
shaped distribution it can be concluded that the disturbance term is normally distributed with mean zero
and constant variance one (i.e. N~ (0, 1)).

To get the residuals normally distributed first we have to make sure that each variables employed are found
to be normally distributed. In this case, most of the variables are found to be normally distributed, the

74
variables that are not normally distributed were transformed to logarithmic form, and the disturbance term
becomes normally distributed. Therefore, normality test was checked out by using Kernel density estimate
test. According to Kernel, using command “kdensity r, normal” whiles after the command of ‘predict r’ the
graph shows normal distribution on estimated residual as compared with normal distribution reference line.

4.3.1.4. Heteroschedasticity
In general, heteroschedasticity is one of the problems of cross sectional data where it has assumed that
homoscedasticity or constant variance in basic classical linear regression assumptions. Due to the
indication for presence of such defects in the data were collected according to White's test, Breusch-Pagan
test and residual plot test, the study was applied robust technique of estimation in the STATA set up which
can easily detect the problem. The result is annexed at the annexation part of this paper.

4.3.2. Results of Regression Analysis

As shown in chapter three, the model used to find out and explain the association between the dependent
variable and the independent variables is:

LRP =β1 + β2(Gdr) + β3(Ag) + β4(Mar) + β5(Educ) + β6(Hhs) + β7(Exp) + β8(Othbus) + β9(Busfrm) +
β10(Bussctr) + β11( Income) + β12(Lnamt) + β13(Div) + β14 (Eq) + β15 (Grprd) + β16 (Folup) + β17
(Coll) + β18 (Int) + β19 (KYC) + β20 (Timhzn) + β21 (Mrkt) + β22 (Wthr)

Where LRP, Gedr, Ag, Mar, Educ, Hhs, Exp, Othbus, Busfrm, Bussctr. Incm, Lnamt, Div, Eq, Grprd,
Folup, Coll, Int, KYC, Timhzn, Mrkt, Wthr denotes Loan Repayment performance, Gender, Age,
Marriage, Education, House hold size, Experience, Other business, Business form, Business sector,
Income, Loan size, Diversion, Equity, Grace period, Follow up, Collateral, Interest, KYC, Time horizon,
Market and Whether respectively.

Under the following regression outputs the beta coefficient may be negative or positive; beta indicates that
each variable’s level of influence on the dependent variable. P-value indicates at what percentage or
precession level of each variable is significant. R2 values indicate the explanatory power of the model.

75
Table 4. 13 Results of Binary Logistic regression, loan repayment performances.
--------------------------------------------------------------------------------
Repayment | Coef. Std. Err. Z P>|z| [95% Conf. Interval]
---------------+----------------------------------------------------------------
Gender | -2.234012 64.67539 -0.03 0.972 -128.9954 124.5274
Age | 1.498252 .938601 1.60 0.110 -.3413725 3.337876
Marriage | .1965257 1.236763 0.16 0.874 -2.227485 2.620537
Education | 4.994262 2.29616 2.18 0.030* .4938714 9.494653
Familysize | -2.783878 1.394197 -2.00 0.046* -5.516454 -.0513033
Experience | 4.997728 2.54503 1.96 0.050* .0095619 9.985895
Otherbusiness | 5.08024 2.396175 2.12 0.034* .3838236 9.776657
Businessform | -.664479 .9807302 -0.68 0.498 -2.586675 1.257717
Businesssector | -1.532271 1.395945 -1.10 0.272 -4.268272 1.203731
Income | 1.793644 1.533704 1.17 0.242 -1.21236 4.799648
Loansize | 2.806922 1.503764 1.87 0.062* -.1404005 5.754245
Diversion | -8.941219 4.74869 -1.88 0.060* -18.24848 .3660423
Equity | 1.589673 1.459438 1.09 0.276 -1.270772 4.450119
Graceperiod | 4.900312 31.71605 0.15 0.877 -57.262 67.06263
Followup | 3.399658 1.992316 1.71 0.088* -.5052096 7.304525
Collateral | 2.710324 1.934826 1.40 0.161 -1.081866 6.502513
Interest | -2.105947 1.666111 -1.26 0.206 -5.371464 1.15957
KYC | .8590335 1.453748 0.59 0.555 -1.990261 3.708328
Timehorizon | -6.440273 2.561279 -2.51 0.012* -11.46029 -1.420258
Market | 2.231671 1.952738 1.14 0.253 -1.595625 6.058967
Whether | -.0583181 1.189738 -0.05 0.961 -2.390161 2.273525
_cons | -1.443807 64.80903 -0.02 0.982 -128.4672 125.5796
--------------------------------------------------------------------------------

Source: STATA version 12 Logistic regression result 2017,

Note: Coef. = coefficient, Std. Err = standard error, Pseudo R2 = 81%, Log likelihood = -18.25, Logistic
Regression Chi-square =141.26

* Significance at 5% ** significant at 10%

76
Table 4. 14 Odds ratio of binary logistic regression, loan repayment performances.
---------------------------------------------------------------------------
Repayment | Odds Ratio Std. Err. Z P>|z| [95% Conf. Interval]
---------------+-----------------------------------------------------------
Gender | .1070979 6.926597 -0.03 0.972 9.51e-57 1.21e+54
Age | 4.47386 4.199169 1.60 0.110 .7107941 28.15924
Marriage | 1.217167 1.505347 0.16 0.874 .1077992 13.7431
Education | 147.564 338.8306 2.18 0.030 1.638648 13288.48
Familysize | .0617984 .0861591 -2.00 0.046 .0040201 .9499905
Experience | 148.0764 376.8588 1.96 0.050 1.009608 21717.96
Otherbusiness | 160.8127 385.3354 2.12 0.034 1.467887 17617.66
Businessform | .5145415 .5046264 -0.68 0.498 .0752699 3.517381
Businesssector | .2160446 .3015863 -1.10 0.272 .014006 3.332528
Income | 6.011317 9.219578 1.17 0.242 .2974944 121.4676
Loansize | 16.55887 24.90063 1.87 0.062 .8690101 315.5271
Diversion | .0001309 .0006215 -1.88 0.060 1.19e-08 1.442016
Equity | 4.902148 7.154379 1.09 0.276 .2806149 85.63712
Graceperiod | 134.3317 4260.473 0.15 0.877 1.35e-25 1.33e+29
Followup | 29.95385 59.67753 1.71 0.088 .6033791 1487.014
Collateral | 15.03414 29.08845 1.40 0.161 .3389625 666.8155
Interest | .1217303 .2028162 -1.26 0.206 .0046473 3.188561
Kyc | 2.360878 3.432122 0.59 0.555 .1366598 40.78555
Timehorizon | .001596 .0040877 -2.51 0.012 .0000105 .2416515
Market | 9.315421 18.19058 1.14 0.253 .2027818 427.9332
Whether | .9433498 1.122339 -0.05 0.961 .0916149 9.713582
_cons | .2360275 15.29671 -0.02 0.982 1.61e-56 3.46e+54

Table.4.13 presented the regression result of Loan repayment Performances (LRP) as dependent variable
and six borrowers specific, nine bank specifics, four business specific and two external factors as
independent variables for the sample of 150 borrowers in Development Bank of Ethiopia Jimma District.
The adjusted R-squared value for the model is around 81%, suggesting that almost 81% variance in
repayment performances were explained by all mentioned explanatory variables. And also adjusted R2
value show that the overall goodness of the model. Accordingly, the value of R2 showing that model
used in this study has good statistical health. F-statistics of the model has a p-value of 0, suggesting that
all explanatory variables jointly can influence the repayment performances.

As shown in the above regression table the output of variables like Education, family size, Credit
experience, Other business and time horizon were statistically significant factors affecting the repayment
performances in Development Bank of Ethiopia at 5% significant level while loan size and loan diversion
are statistically significant factors affecting the repayment performances in Development Bank of Ethiopia
at 10% significant level. The coefficients of three significant variables, time horizon, family size and loan
diversion were negative and the left five, education, experience, other business, loan size and follow up
were positive. The negative coefficient indicates that the dependent variable was associated with the

77
independent variables negatively and the positive coefficient shows the positive influence of the variable
on the dependent variable.

On the other hand, thirteen (13) variables were found insignificant on dependent variable namely gender,
age, marital status, business form, business sector, income/profit, equity, grace period, collateral, interest
rate, KYC/ due diligence, market condition and weather conditions were statistically insignificant influence
on loan repayment performances. From these insignificant variables gender, family size, business form,
business sector, interest rate and weather conditions are having a negative sign and the remaining
insignificant variables bear positive sign. Overall, the binary logistic model predicted factors contributing
to 81% of Development Bank of Ethiopia Jimma District loan repayment performances as revealed in the
above table.

4.3.3. Discussions on Regression Results

The preceding sections present the overall results of the study. Thus, this section presented detail analyses
of the results for each explanatory variables and their importance in loan repayment performances in
accordance with the above regression result. In addition, the discussions analyses the statistical findings of
the study in relation to the previous empirical evidences.

According to the binary logistic result, the significant variables were significant at different level of
significance and discussed as below.

Time horizon: The result of logit model in table 4.13 indicated that time horizon have a negative influence
on loan repayment performances and statistically became significant predictor of borrowers’ loan
repayment performance at 5% significance level. As indicated table 4.13 and 4.14, timely disbursement of
loan increases the borrowers’ loan repayment probability by 5.6%. Thus, the result is in accordance with
the research hypothesis (time horizon has negatively influence repayment performances). This implies that
getting service timely or after long time waiting, keeping the other thing constant has a resultant change of
5.6% increase or decrease the repayment in the opposite direction. I.e. The odd ratio of the econometric
result indicates that disbursing the loan timely can reduce the probability of being default by 65% other
things remain constant (table 4.13). There are a number of studies found negative relationships between
time of disbursement and loan repayment performances. The result was consistent with the descriptive
analysis result in preceding section of this same study and consistent with the study and findings of other

78
research like studies by Shaik and Tolosa( 2014) confirmed that timely disbursement of loan increases the
borrowers loan repayment probability.
Accordingly, the null hypothesis H2 is fail to rejected. That is the null hypothesis that expected positive
relationship between delayed service and defaulting is accepted. Because the result of the regression proves
that as the time of delivering service elongated the likelihood of defaulting increases and the probability of
performing decreases.

Educational qualification; Table 4.13 and 4.14, shows that educational level is significant at 5% and
positively related to borrowers ability to repay their loans. An increasing the level of education has the
effect of decreasing the likelihood of defaulting by 14% ceteris paribus. This figure reveals that the
borrowers whose educational is at tertiary level have the probability of decreasing default rate by 14
percent than the borrowers who is at elementary education level/ illiterates table 4.13. This implies that
borrowers that were more educated may have access to business information, use their personal knowledge,
skill and experience to properly manage their loan and repay timely. This result was consistent with
preceding descriptive analysis of this study and these results are resembled with the output of Michael
(2006) and Olomola (2009) described that default rate decreased with education level of the borrower
increased.
The null hypothesis H1 is fail to reject. The null hypothesis stated that there is positive relationship
between education and loan repayment performances, which found true under the regression result. So, the
null hypothesis is accepted.

Family size: As shown under the logit regression table above, family sign showed significance to the
dependent variable. It was hypothesized that a borrower having larger family number is likely to default
than a borrower having small family number, and vice versa. The coefficient of family size is negatively
related to the dependent variable, loan repayment performances and is strongly significant at 5% level.
Increasing borrower’s family size by one person decreases the likelihood of being able to repay one’s loan.
This means that the smaller the size of the borrower family, the higher the probability that borrowers will
be able to repay their loans and vice versa. This result and conclusion is similar with the results under
descriptive analysis is part of the study and the reason of such may resulted from the fact that large
household sizes increased the household head’s domestic responsibilities and thereby constituted leakage to
the household’s income stream. As household income depleted, liability of the household increased and
there would be greater tendency to divert loans meant for borrower production resulting in default in loan

79
repayment. The odds ratio indicates as family number increases the likelihood of defaulting increases by 60
percent compared to small family holders. This result is similar with (Abbafita, 2005 and Berhanu, (2005).
The null hypothesis H1 is accepted. The null hypothesis stated that the increase in number of
dependents/family increases the probability of defaulting and decreases the likely hood of performing the
loan. The logistic regression result indicated this same result; hence accepted.

Other business: The variable other business has a positive sign as expected and is statistically significant
at 5%. The result shows that as the borrower have other sources of income his/her capacity to repay his
loan increase. Under the descriptive analysis part of the study, having other business positively contributed
to repayment performances and this result is almost the same with the results of this econometric logistic
regression result. This implies that incases income from the project under consideration fail to meet their
debt obligation income from such other source could help to settle their repayments. The studies made by
(Kibrom, 2002) and (Abraham 2002) supports this result.
The logistic regression and chi square of fail to reject hypothesis H3. The null hypothesis which stated
having other business in addition to the current one is better for repayment and there is positive relationship
with repayment performances. The regression result showed this assumption true and the hypothesis is
accepted.

Credit experience: The coefficient of this variable was expected to influence the repayment capacity
positively and the result of logit regression shows the same as expected. P-value of the credit factor is
statistically significant at 5% (0.050) and has a positive influence on the dependent variable, which is in
line with the research hypothesis (there is a positive relationship between credit experience and repayment
performances). The coefficient value of the variable (i.e.4.007) indicated a percentage rise/decline in years
of experience resulted performing/ nonperforming of the loans. The coefficient value tells us there is a
strong positive relationship between credit experience and repayment performances. The implication of this
result is that those who had long credit experience have good knowledge of managing and handling the
financial aspects of their business and at better position than those who never had such exposure. This
result is the same with results presented under descriptive part of this research. This result agreed with
(Firafis Haile, 2003).
The logistic regression and chi square of fail to reject hypothesis H1. The null hypothesis which stated
having credit experience better for repayment and there is positive relationship with repayment
performances. The regression result showed this assumption true and the hypothesis is accepted.

80
Loan Diversion: The coefficient sign of loan diversion shows that there is a negative relationship between
loan diversion and loan repayment performances. This variable adversely and significantly influences loan
repayment performances at 10% significance level, borrowers who diverted the loan other than the
intended purpose are found to be defaulters. An application of entire loan for intended and productive
business lessens the probability of defaulting by 0.060 (table 4.13). It is obvious that diverted loans miss
their intended target and out of the sight of lending institutions. Hence, unless the responsible borrowers
willingly pay their loan from such other business it would be difficult to repay the loan according to the
terms of the contract. This result is the same with (Jemal, 2003), and results under descriptive analysis part
of this research in preceding section. The null hypothesis H2 is failed to reject. The null hypothesis stated
that there is negative relationship between loan diversion and loan repayment performances, which found
true under the regression result. So, the null hypothesis is accepted.

Loan size: this variable also was found to influence borrowers’ loan repayment performance positively and
significantly at 10% significance level. Keeping the other factors constant, having sufficient loan size and
operating business with adequate amount of capital decreases the probability of being defaulter by 0.088
(table 4.13). Large amount of money creates huge capacity to performance with full capacity and effective
manner. Accordingly, loan size showed positive relationship with loan repayment performances indicating
that the increase in the loan size likely increases the loan repayment capacity of the borrowers. This same
result was found by Olomola (2009), Nawai and Shariff (2013), Abafita (2003). Shaik Abdul Majeeb
PASHA (2014). The result here is almost presented in descriptive part of this study too. The null
hypothesis H2 is accepted. The null hypothesis expected the positive relationship between loan repayment
performances and loan size, which found to be true under logistic regression result. Hence, the null
hypothesis is accepted.

Follow up: This variable was to have positive and significant association with the dependent variable. It is
significant predictor loan repayment performance at 10% significance level. If other variables held
constant, continuous follow up and visit of respondents reduces their probability of being defaulter by
0.088. The importance of follow up is unquestionable in every credit monitoring and loan collection. The
odds ratio of the variable indicated that a project that follow up is properly made is 29.95% more likely to
repay its loan than that never proper follow up is made. This result is the same with finding of
Wongnaa(2013).

81
The null hypothesis H2 expected the follow-up to have positive relationship with repayment. The result of
the logistic regression showed that there is strong positive relationship between loan follow-up and
repayment of loans. Hence, the null hypothesis is accepted.

82
 CHAPTER FIVE

CONCLUSION AND RECOMMENDATION

The preceding chapter presented results and discussion of the study, while this chapter deals with
conclusion and recommendation of the study based on the findings. Accordingly this chapter is organized
into two sub-sections. The first section of this chapter discusses the conclusions part briefly and the second
section presents recommendation for the findings.

5.1. Conclusion

The objective of this research is to identify and determine factors affecting loan repayment performances at
DBE Jimma District. To achieve this broad objective, the study used both qualitative and quantitative data
and the primary data was collected from 150 borrowers, nine senior expertise and three managers at
different level of the bank using semi structured open ended and close ended questionnaire. For data
analysis purpose both descriptive statistics and binary logistic model were employed.
Therefore, this study was intended to identify and discuss factors which affect borrowers’ loan repayment
performance and finally concludes that low repayment performance was one of the main problems of the
District as compared to its plan and other performances such as loan appraisal, loan approval and
disbursements. The descriptive statistics findings shows that there were significant association between
dependent variable with respect to time horizon, level of education, family size, experience, other business,
follow-up, loan size and loan diversion variables were significantly influenced the repayment performances
of the loans. On the other hand, twenty one explanatory variables were entered in to binary logistic model
and out of which eight variables were found significant to determine loan repayment performance of
borrowers.

The results of this study revealed that the time horizon negatively and significantly affected the loan
repayment performance of borrowers. Time lag between loan application and disbursement should be
reduced to increase repayment rate. The complicated loan processing procedures, which might lead to
delay in disbursement, further, it will increase default rate. When the bank deliver its services timely, the
probabilities of paying loan and in the reverse if the bank fail to provide services after a long time of
waiting and after time of utilizing opportunity is lapsed, the probabilities of defaulting is very high. Time

83
of loan disbursement was also another significant variable with default loan negatively. Thus, unless the
bank faces strange problems, the risk of being default most probably decreases when disbursements
performed on time. Therefore disbursing the loan on time, we can expect high loan repayment
performance.

The education qualification level determines loan repayment positively and significantly. The borrowers
who attained higher education level able to pay better than the borrowers who were in lower level
schooling and/or illiterates. Therefore, institution should motivate educated people and also easy to provide
training. The selection of educated borrowers decreases the probability of being default. This is the fact that
the literates can easily grasp knowledge, information, capable to manage their business, adopt new
technologies and workable strategy for their business than the illiterates.

Family size also influenced the repayment performances of loans significantly and shows negative sign.
This indicates that increasing in the number of family size increases the probabilities of default and vice
versa. Borrowers who have small number of or no dependents in the household perform better in loan
repayment. The borrowers who support large number of dependents face difficulties of repayment. The
logic behind is that the borrower having larger family size as compared to those having smaller family size
have tremendous challenges to administer the demands of his/her family and run the business
simultaneously. The larger family sizes have different needs and high consumption, while the small size
borrowers can focus on administering their business without much challenges and difficulties

Loan size; is the other variable showing positive relationship with loan repayment performances and
statistically significant. Repayment capacity of borrowers depends on the capacity of investments and the
profit they generate from the business itself. A project, In order to operate with full capacity and without
any financial constraints needs to have full financial support, including investment cost and working
capitals. So, when huge capacity is created, the probability of defaulting is low and vice versa.

Loan diversion was also found as essential and significant factors of loan repayment rate negatively. Loan
diversion is negatively affecting the loan repayment capacity. It is clear that diverted loans miss their target
and cannot repay the loan according to the duty. This means, diverting loan into non-income generating

84
activities increases default rate. Therefore, it is recommended that the institution should give attention to
continuous follow-up on proper loan utilization

Credit experience is also another significant variable. A borrower having credit experience is at better
position to repay its loan than those new comers. The reason is that while experienced borrowers use their
skill, knowledge and familiarity in carrying out their duty, new borrowers faces new environment to begin
from the scratch.

Having another business is another significant variable influencing the dependent variable positively. The
help of having addition business is to use the experience such other business in running the current one and
financial support in case of default. But this should be handled very carefully because the existence of
another business may also be the cause of failure if diversion of money to such other business occurred.
The positive sign and the significance is from the support and help it gives to the current business while the
other side should be considered with caution.

The other significant variable was follow-up. This variable influence borrower’s loan repayment
performance positively and significantly. Giving Projects proper follow up, the probability of default
decrease since problems will be tackled immediately and utilize their loan effectively, generate revenue,
and then make loan repayment. The follow-up and supervision made by the loan officers and concerned
bankers should be increased and it leads to increase repayment performances.

Generally, the finding of the study failed to reject two research hypotheses that indicate the relationship
between loan repayment performances and borrower related factors, specifically education level, family
size of borrowers and experience and bank related factors like having other business, follow-up, loan size,
time horizon and loan diversion whereas the remaining were insignificant.

85
 5.2. Recommendation
It is apparent that DBE has to work to avert the loan repayment problems. The source of loan repayment
performances as indicated under this research is from four different areas and the bank is required to work
on the solution to bring about better performances. Financial performances and wellness is one criteria of
measuring financial institution healthiness, which in cases of DBE is possible through loan approval
collection and disbursement.
Now, Based on the analysis and findings of this study, the researcher therefore recommends that:

The study revealed that among personal or borrower characters, educational level, credit experience and
family size were the main and significant factors of loan collection performance which was unattractive in
the past consecutive years. From bank specific factors, time horizon, follow up, loan diversion and loan
size were found significant variables and from business specific factors having another business is found
significant variable.

Therefore, the bank is recommended to select and screened out those customers who are more educated and
have credit experience in running related business. Proper due diligence should be conducted in screening
customers with better educational level and credit experience. The major activities of screening know the
personal traits and history of the borrower and the feasibility and viability of the business. Hence, customer
with better educational qualification and experience should be selected.

The researcher also recommends timely disbursement of loan. Since projects are sensitive to season
(production, market, and implementation) for these hold proper amount and disburse when the need arises.
Disbursement dalliance was the problem, where the main challenge of the bank in the last four consecutive
years. The main justification behind such dalliance was less number of contact officers and engineers as
compared with financed projects which could unable to make necessary follow up and progress report.

Follow-up being one significant variable by itself, when properly implemented solve other related
problems too. The bank has to increase the number of officers and engineers who has responsibility of
taking full-fledged follow up and revision as well as progress report, respectively. Thus, follow up also as
being one of significant factor, increasing the number of contact officers and give more attention on follow
up can increase good performance of projects hence loan collection performance of the bank.

86
The other important recommendation is regarding loan diversion. The bank is highly recommended to
follow the money released for project development and avert the diversion of loan. The main cause of loan
diversion is lack or loose of following the money and progress supervision.

The other important recommendation is regarding loan size. In order to implement the project with full
potential and capacity necessary capital should be allocated. Such loan size shouldn’t be more than what is
needed or less than what is required. So, the bank is recommended to conduct critical feasibility

Finally, the researcher recommends other researchers to do by including the other Districts & head office,
and the determinants of other variables like loan repayment performance, outreach, using innovative
features of the bank and the other variable

Generally, internal factors can be easily controlled while external factors can be a threat to the viability of
banks. Banks have to be vigilant in their lending decisions so as to avoid loan losses and the accumulation
of non-performing loans. Banks need to concentrate on sectors that are performing well and avoid lending
to those sectors which have already recorded a significant amount of non-performing loans. One thing to
note is that, this result can be generalized to the whole banking sector in Ethiopia as almost all the banks
have been affected by non-performing loans. Therefore, the recommendations generated are a prescription
for all banks engaged on similar investment activities in Ethiopia.

87
 REFERENCE
Abafita, J. (2003). Microfinance and loan Repayment Performance, A Case Study of the Oromia Credit and
Savings Share.

Abebe Mijena (2011).Determinants of Credit Repayment and Fertilizer Use by Cooperative Members in
Ada District, East Shoa Zone, Oromia Region, Haramaya University, and un- published thesis.

Abdullah- Al- Mamun, S. A. (2011). Empirical Investigation on Repayment Performance of Amanah,

Ikhtiar Malaysia’s Hardcore Poor Clients. International Journal of Business and Management, Vol.
6.

Abereham, G. (2002). Loan Repayment and Its Determinant in Small Scale Financing in Ethiopia.
unpuplished.

Addisu, M. (2006). Micro-finance repayment problems in the informal sector in Addis Ababa, Ethiopian
Journal of Business & Development.

Amare, B. (2005). Determinants of Formal Source of Credit Loan Repayment Performance of Smallholder
Farmers. Unpublished, 110.

Assefa, A. (2004). A Review of the Performance of Agricultural Finance in Ethiopia Pre-and Post-Reform
Periods. Unpublished.

Belay, K. (1998). Factors affecting loan repayment performance of smallholders in the Central Highlands
of Ethiopia. Ethiopian Journal Agricultural Economics, 2 (2), 6 1-89.

DBE annual report. (2009). Development Bank of Ethiopia annual report 2009. Unpublished.

DBE annual report. (2015). Development Bank of Ethiopia annual report 2015. Unpublished.

Development Bank of Ethiopia. (2014, 2015, and 2016). DBE Annual Reports, Fourth Quarter and Annual
Report. Unpublished.

Development Bank of Ethiopia. (2015). Development Bank of Ethiopia Procedural Manual. Unpublished.

Development Bank of Ethiopia. (2015). Revised credit policy of Development Bank of Ethiopia.
Unpublished.

Firafis Haile, (2015) Determinants of loan repayment performances: case study of Harari micro finance
institutions, journal of Agricultural Extension and rural development, January 22, 2015.

88
Garsone, D. (2008). Factor Analysis: Statnotes. Retrieved March 22, from North Carolina State, University
Public Administration Program.

Gebeyehu, B. (1987). Currency and Banking in Ethiopia, Addis Ababa.

Ghatak M. and Guinnane w. (1999). The Economics of Lending with Joint Liablity. Journal of Development
Economics, Vol. 60, pp.1 95-228.

Gujarati, D. N. (2004). Basic econometrics, 4th edition, 4th edition, The McGraw-Hill Companies.

Itana, A. (2003). Credit Policy, Financial Policy and Privet Investment in Ethiopia. Unpublished, 237-252.

Kahansal, M. R., & Mansoori, H. (2009). Factors affecting on loan Repayment Performance of Farmers in
Khorasan-Razavi Province of Iran. Conference on International Research on Food Security, Natural
Resource Management and Rural Development.

Khandker, S. K. (1995). Grameen Bank: performance and sustainability, World Bank, Washington DC.
Discussion paper- 306.

Kibrom, T. (2010). Determinants of successful loan repayment performance of private borrowers in

Development Bank of Ethiopia, North region. Unpublished.

Maddala, G. (1998.). Introduction to Econometrics, 5th edition. New York: Macmillan Publishing
Company.

Mehmood, Y., Ahmad, M., & Anjum, M. (2012). Factors Affecting Delay in Repayments of Agricultural
Credit. World Applied Sciences Journal 17 (4), 447-451.

Micha'el, A. (2006). Micro-finance Repayment Problems in the Informal Sector in Addis Ababa. Ethiopian
Journal of Business &Development, Volume 1.

Million Sileshi, R. N. (2012). Factors Affecting Loan Repayment Performance of Smallholder

Farmers in East Harerghe, Ethiopia. Developing Country Studies, Vol 2, No.11.

Ministry of Finance and Economic Development. (2014). Annual Report of Ministry of Finance and
Economic Development. Unpublished.

Nawai, N. B., & Sharif, M. N. (2013). Determinants of loan Repayment Performance in Microfinance
Programs in Malaysia. Labuan Bulletin of International Business & Finance, Volume 11, PP 14-29.

89
Nwagbo, E. C. (1989). Impact of institiutional credit on agricultur in Funa local government area of Kattsina
state, Nigeria. Samanu journal of agricultural research vol.6, 78-89.

Nwankwo, O. (2013). Agricultural Financing in Nigeria,an Empirical Study of Nigerian Agricultural Co-
operative and Rural Development Bank (NACRDB): 1990-2010 . Journal of Management
Research, Volume 5, No.2.

Oboh, G. B. (2006). A model of publice expenditure for agricultur in the budget. Journal of economic
development.

Oke, J. A. (2007). An Empirical Analysis of MicrocreditRepayment in Southwestern Nigeria. . Humanity &

Social Sciences Journal 2 (1):63-74.

Olomola, A. S. (1998). Determinants of Smallholder loan Repayment Performance.

Rahaman, S. u., Hussien, A., & Taqi, M. (2014). Impact of Agricultural Credit on Agricultural Productivity
in Pakistan. International Journal of Advanced Research in Management and Social Science,
Volume 3, No.4.

Raji, M. (2000). An analysis of the determinants of Agricultural credit approval/loan size by commercial
Banks in South Western Nigeria. Journal of Nigerian Development Studies, 1(1), 6-26.

RK, G. (2012). Determinants of Loan Repayment in Commercial Banks A case of CRDB Bank Plc (Moshi
Branch).

Singh, Gurbachan, J.S., S., & Balawant, S. (1985). A Study on Repayment Performance of Borrowers in
Punjab.

Stigiltz, J. (1989). Money, Credit and Business Fluctuations. National Bureau of Economic Research,
Working Paper No. 2823.

Stigiltz, J. (1999). The Financial System, Business Cycles and Growth. National Bureau of Economic
Research.

Tenaye, A. (2003). Factors influencing repayment performance of agricultural credit in Southern Ethiopia.
unpublish.

Tesfaye Ginbare Gutu, T. S. (2014). Determinants of Group Loan Repayment Performance of MSEs
Manufacturing Sector. Research Journal of Finance and Accounting, Vol, 5.

90
Tesfaye, A. (1993). Rural Credit in Ethiopia. Addis Ababa, Ethiopia.

Welday, A. (2003). Micro finance in Ethiopia: Performance, Challenges and Role in Poverty Reduction.
AEMFI, Occasional Paper, 7.

William E. Wilcox, MCE (n.d). What qualifies you for credit? The six “c” of credit, CBM Credit
Education foundation, inc.

Wongnaa, C. A. and Awunyo-Vitor, D. (2013). Factors Affecting Loan Repayment Performance among
Yam Farmers in the Sene District, Ghana. Agris on-line Papers in Economics and Informatics, 5, 111-122.

World Bank. (2015). Annual Report of World Bank.

Yarnold, L. G. (1995). Principal components analysis and exploratory and confirmatory factor analysis. In
L. G. Grimm & R R. Yarnold (Eds.), Reading and understanding multivariale statistics (pp. 99-136).
Washington, DC: American Psycholog.

Zelalem, G., Hassen, B., & Jemal, H. (2013). Determinants of Loan Repayment Performance of Smallholder
Farmers. 431- 446.

Yacob, G. (2014). Socio-Economic Determinants of Eritrea‟s Savings and Micro Credit Program Loan
Repayment Performance: a case of the Dekemhare Suz-Zone, International Journal of Development and
Economic Sustainability Vol.2, No.2, pp. 48-63, June 2014.

Yasir, M., Mukhtar, A.,and Muhammad ,B.(2012). Factors Affecting Delay in Repayments of Agricultural
Credit; A Case Study of District Kasur of Punjab Province, World Applied Sciences Journal, 17 (4): 447-
451.

Zena Lemat Bank (2009). “A short history of Development Bank of Ethiopia”, No.45.

91
 APPENDIX

Appendix 1. Questionnaire paper

JIMMA UNIVERSITY
COLLEGE OF BUSINESS AND ECONOMICS

DEPARTMENT OF MANAGEMENT

MBA RESEARCH QUESTIONARY

This Research Questionnaire is for academic purpose only!

Dear respondent, this questionnaire is prepared to collect data on loan repayment performances, for the
purpose of MBA research to be conducted under a title Factors affecting loan repayment performances in
Development Bank of Ethiopia Jimma District. Get relaxed and feel free to respond the questions and focus
on providing the required information to help the researcher do his/her job rightly. Hence, I kindly request
you to fill the questionnaire very carefully and provide genuine information so as to help me find the actual
reason for the identified problems. Advance thanks for your patience and time
Factors of loan repayment performance
A. Borrower’s related information
B. Name of borrower(optional)_____________________________________________
C. Gender 1) Male 0) Female
D. Age: 1. 15-30 2. 31-50 3. Above 51
E. Marital status 1. Single 2. Married 3. Divorced/widowed
F. What is your family size? ---------------------------
1. Small (1-3) 2. Medium (4-5) 3. Large (above 6)
G. Educational Background: 1. No formal education 2. Primary school completed 3. High School
completed 4. College/University graduate
H. Do you have any credit experience in running similar project? 1) Yes 0) No.
If yes, did it help you for current business? Explain how ___________________
I. Business related questions;

92
 A. What is the current status of your business? 1. Performing good/successfully operating 2. Not
good/defaulted(substandard, doubtful and loss)
B. What is your business form? 1. Sole proprietor 2. PLC 3. SHC and Others
C. What is your business sector? 1. Agriculture 2. Service 3. Industry
D. Have you gained sufficient income compared to your plan? 0) No 1. No
If no, why (list reasons) …………………………………………………………………..…
…………………………………………………………………………………………………………
…………………………………………………………………………………………………………
…………………………………………
E. Do you have other business? 1)yes 0) No
II. Institution related questions;
A. What is a Loan size permitted for the project? ----------------------- birr
B. Do you believe such amount is sufficient for your project as compared to feasibility study? 0) No
1) Yes
If no, explain how it affected your repayment performances------------------------------------------------
------------------------------------------------------------------------------------------
C. Have you used any amount of money from the loan to operate some other business or used for your
personal consumption? 0) No 1) yes
If yes, explain the amount, ----------------------------------------
D. Amount of equity contribution? ------------------ birr
Is such amount 1) Exceed 30% of total investment 2. 30% only
E. Do sufficient grace period granted to begin repayments? 0) No 1) Yes
If No, explain how it affected your repayment capacity? ---------------------------------------------------
--------------------------------------------------------------------------------------------
F. Do you think the bank has made a proper follow up to the project? 0) No 1) Yes
G. Do you think the bank has secured its loan with enough/sufficient collateral in cases of default?
0) No 1) Yes
H. What is the collateral of the bank for the loan? Is there any property other than the project itself?
0) No 1) Yes
If yes, explain the amount-------------------------------------------------------------------------
I. What is the ratio of debt to collateral value? -----------------------------------------------

93
 J. How do you evaluate the change (increase) in interest rate, do you think it affected your repayment
performances? 0) No 1) Yes
If yes, is that positively or negatively? Explain ---------------------------------------------------------------
--------------------------------------------------------------------------------------------
K. Do you believe the KYC (know your customer) assessment was performed duly according to policy
and procedures of the bank? 0. Yes 1. No
L. How do you evaluate the service period (time to conduct KYC, appraisal and approval)? 1. Timely
2. Elongated 3. Too late
If such time has affected your repayment capacity, explain your reason. ---------------------------------
---------------------------------------------------------------------------------------------------------
III. External factors
A. Do you think International/national market fluctuation affected your repayment performances?
0) Yes 1) No
If yes, what is the reason, how? ---------------------------------------------------------------------------------
---------------------------------------------------------------------------------------.
B. Were the project attacked by pest and weed problem?
0) Yes 1) No
If yes, what are the causes? --------------------------------------------------------------------------------------
---------------------------------------------------------------------------------
C. Were the project faced bad weather condition problem like flood, too less or too much rainfall? 0)
No 1) No
If yes, what are the causes, explain -----------------------------------------------------------------------------
------------------------------------------------------------------------------------------
Other factors
Elaborate other major challenges and factors that challenged the repayment of bank credit and over all
performances of your business. -------------------------------------------------------------------------------------------
---------------------------------------------------------------------------------------------------------------------------------
---------------------------------------------------------------------------------------------------------------------------------
-----------------------------------------------------------
Thank you for your cooperation!

94
Appendix.2 Interview questions
JIMMA UNIVERSITY

COLLEGE OF BUSINESS AND ECONOMICS

DEPARTMENT OF MANAGEMENT

MBA RESEARCH INTERVIEW

This interview questions is for academic purpose only! (For Bank staffs & officials)
This interview questions are prepared to collect relevant data for my MBA research under a title ‘Factors
affecting loan repayment performances, the case of DBE Jimma District’ and your answers thereby will be
utilized for the same purpose. Thank you in advance for your willingness and cooperation and please help
me in providing a genuine information. The confidentiality of the information you provide will be kept.

1. How do you analyses the current performances of the bank and what do you think contributed for such
performances?

2. Do you believe the screening process and due diligence are duly conducted to select credit worthy
borrowers? What are the limitations seen during the due diligence assessment?

3. Do you think the bank conducted the appraisal and approval activities duly as per the policy of the
bank? What drawbacks do you observe regarding appraisal and approval process?

4. What do you think about the lengthy and time taking process in all screening, appraisal, approval and
disbursement process of the bank? What do you think the bank should do to solve such problems?

5. Do you believe that the bank has done fledged follow up for its customers /loan? What are the results
after project follow up?

6. How do you explain the impact of due diligence, loan appraisal, and project follow up with loan
repayment?

7. What are the crucial confronting factors for loan repayment in the District?

8. What alternative measures were taken on the side of the bank to improve the repayment Situation?

9. Were the measures taken brought an improvement in repayment status of the project?

Thank you!

95
 Annex 3. Correlation matrix of coefficients of regress model
e(V) Gender Age Marriage Educat~n Househ~e Experi~e Otherb~s Busine~m Busine~r Income Loanam~t

Gender 1.0000
Age 0.0123 1.0000
Marriage 0.0119 0.1132 1.0000
Education -0.1016 0.0805 0.1678 1.0000
Households~e -0.1766 -0.2098 -0.1919 -0.0241 1.0000
Experience 0.0499 0.0796 -0.0379 0.0028 0.0430 1.0000
Otherbusin~s -0.0107 -0.1843 -0.0094 -0.1512 0.0063 0.0688 1.0000
Businessform -0.1868 0.0715 -0.1180 0.0908 0.0261 0.1947 0.0404 1.0000
Businessse~r 0.0417 -0.0780 -0.0571 -0.2016 0.0607 -0.0116 -0.0415 -0.0690 1.0000
Income 0.0888 -0.1700 -0.0591 -0.2683 0.0464 0.0007 0.0406 -0.1400 0.1151 1.0000
Loanamount -0.0001 0.0459 -0.2583 -0.0890 -0.0358 -0.0142 -0.0166 0.1724 -0.3131 -0.1162 1.0000
Diversion -0.0680 -0.1369 -0.0075 0.1029 -0.0013 -0.0530 -0.0022 0.0074 0.0265 -0.0277 0.0207
Equity 0.0886 -0.0742 -0.0693 -0.1436 0.1023 -0.0128 -0.0264 -0.1039 0.0921 0.1862 -0.0282
Graceperiod 0.0291 0.1231 -0.1207 -0.0873 0.0859 0.1042 -0.2372 0.0207 0.0983 -0.0490 -0.1448
Followup -0.0379 0.0030 0.0315 -0.0572 -0.0324 -0.0139 0.0240 0.1050 -0.0874 -0.1106 -0.0017
Collateral -0.0424 -0.0762 -0.0517 -0.0296 0.0273 -0.0512 -0.0909 0.0981 0.0643 0.0451 0.0303
Interest -0.0943 -0.0327 0.1511 -0.1199 0.0053 -0.0717 0.0976 -0.0510 0.0308 -0.0127 -0.2071
Kyc 0.0648 -0.0523 -0.0289 -0.2337 -0.0726 -0.1079 -0.1155 0.0374 0.0033 0.0883 0.0891
Timehorizon -0.1318 -0.0355 0.0771 0.3210 -0.0571 0.0919 0.0993 -0.0321 -0.0122 0.0936 -0.1239
Market 0.1758 -0.0265 -0.1309 -0.0591 0.0704 0.0601 0.0460 -0.0052 0.0902 0.0317 0.0671
Whether 0.1610 0.0824 0.0861 0.0529 -0.1941 0.0710 0.0632 -0.0234 -0.0334 0.0452 0.0125
_cons -0.2784 -0.2898 -0.2912 -0.4279 -0.0671 -0.2707 -0.0251 -0.3133 -0.0877 0.0123 -0.1038

e(V) Divers~n Equity Gracep~d Followup Collat~l Interest Kyc Timeho~n Market Whether _cons

Diversion 1.0000
Equity -0.0204 1.0000
Graceperiod 0.0586 0.0188 1.0000
Followup 0.1182 -0.1482 -0.0211 1.0000
Collateral -0.0273 -0.0439 0.0848 -0.0563 1.0000
Interest -0.0292 0.0409 0.0072 -0.0076 -0.0034 1.0000
Kyc -0.0235 0.0079 -0.1584 -0.0304 0.3433 0.1955 1.0000
Timehorizon -0.0044 -0.1111 0.0417 0.0936 0.0722 -0.0777 -0.1131 1.0000
Market -0.0649 0.0547 -0.0336 -0.0709 -0.0230 0.0606 0.0034 -0.1829 1.0000
Whether -0.0097 0.1174 -0.0986 0.0323 -0.0531 0.0927 0.0748 0.1305 -0.0049 1.0000
_cons 0.0029 0.0038 -0.0223 -0.0419 -0.1741 -0.0382 -0.0546 -0.4358 -0.1285 -0.3015 1.0000

96
 Annex 4. VIF test result

Variable VIF 1/VIF

Education 1.71 0.583593

Kyc 1.44 0.693435
Loanamount 1.44 0.695333
Timehorizon 1.41 0.711450
Graceperiod 1.34 0.747682
Otherbusin~s 1.29 0.772964
Businessse~r 1.26 0.791117
Income 1.26 0.791368
Marriage 1.24 0.804159
Collateral 1.23 0.813179
Gender 1.22 0.817818
Interest 1.22 0.818259
Businessform 1.22 0.822914
Age 1.20 0.830964
Households~e 1.19 0.839626
Whether 1.16 0.859426
Followup 1.13 0.883061
Equity 1.13 0.884816
Market 1.13 0.887345
Experience 1.12 0.891935
Diversion 1.09 0.913764

Mean VIF 1.26

97
 Annex 5. Logit regression result

Logistic regression Number of obs = 150

LR chi2(21) = 141.26
Prob > chi2 = 0.0000
Log likelihood = -16.359487 Pseudo R2 = 0.8119

Repayment Coef. Std. Err. z P>|z| [95% Conf. Interval]

Gender -2.234012 64.67539 -0.03 0.972 -128.9954 124.5274

Age 1.498252 .938601 1.60 0.110 -.3413725 3.337876
Marriage .1965257 1.236763 0.16 0.874 -2.227485 2.620537
Education 4.994262 2.29616 2.18 0.030 .4938714 9.494653
Householdsize -2.783878 1.394197 -2.00 0.046 -5.516454 -.0513033
Experience 4.997728 2.54503 1.96 0.050 .0095619 9.985895
Otherbusiness 5.08024 2.396175 2.12 0.034 .3838236 9.776657
Businessform -.664479 .9807302 -0.68 0.498 -2.586675 1.257717
Businesssector -1.532271 1.395945 -1.10 0.272 -4.268272 1.203731
Income 1.793644 1.533704 1.17 0.242 -1.21236 4.799648
Loanamount 2.806922 1.503764 1.87 0.062 -.1404005 5.754245
Diversion -8.941219 4.74869 -1.88 0.060 -18.24848 .3660423
Equity 1.589673 1.459438 1.09 0.276 -1.270772 4.450119
Graceperiod 4.900312 31.71605 0.15 0.877 -57.262 67.06263
Followup 3.399658 1.992316 1.71 0.088 -.5052096 7.304525
Collateral 2.710324 1.934826 1.40 0.161 -1.081866 6.502513
Interest -2.105947 1.666111 -1.26 0.206 -5.371464 1.15957
Kyc .8590335 1.453748 0.59 0.555 -1.990261 3.708328
Timehorizon -6.440273 2.561279 -2.51 0.012 -11.46029 -1.420258
Market 2.231671 1.952738 1.14 0.253 -1.595625 6.058967
Whether -.0583181 1.189738 -0.05 0.961 -2.390161 2.273525
_cons -1.443807 64.80903 -0.02 0.982 -128.4672 125.5796

98
 Annex 6. Logistic regression of categorical variables
• -------------------------------------------------------------------------------
• Repayment | Odds Ratio Std. Err. z P>|z| [95% Conf. Interval]
• --------------+----------------------------------------------------------------
• _IAge_2 | 1.125774 .8320763 0.16 0.873 .2644331 4.792773
• _IAge_3 | 9.261878 9.800595 2.10 0.035 1.164092 73.69037
• _IMarriage_2 | 1.309394 .8651349 0.41 0.683 .358649 4.780475
• _IMarriage_3 | 11.06442 16.85735 1.58 0.115 .5585586 219.1738
• _IEducation_2 | 6.165825 7.782631 1.44 0.150 .519502 73.18048
• _IEducation_3 | 282.0967 401.4662 3.96 0.000* 17.33844 4589.72
• _IEducation_4 | 1379.739 2426.282 4.11 0.000* 43.94939 43315.3
• _IBusinessf_2 | .0496303 .0442256 -3.37 0.001* .0086545 .2846121
• _IBusinessf_3 | .9602172 1.019277 -0.04 0.969 .1198983 7.68999
• _IBusinesss_2 | 4.847324 5.179998 1.48 0.140 .5968732 39.36606
• _IBusinesss_3 | .4423924 .6805392 -0.53 0.596 .0216966 9.020371
• _IHousehold_2 | .5290126 .459665 -0.73 0.464 .096349 2.904589
• _IHousehold_3 | .0805831 .0743908 -2.73 0.006* .0131963 .49208
• _ILoanamoun_2 | .4663687 .3681641 -0.97 0.334 .0992585 2.191246
• _ILoanamoun_3 | 1.423455 1.377071 0.36 0.715 .2137368 9.480002
• _cons | .3954776 .5345418 -0.69 0.493 .0279645 5.592896
• -------------------------------------------------------------------------------

99
Annex 7. The odds ratio of logistic regression

Logistic regression Number of obs = 150

LR chi2(21) = 141.26
Prob > chi2 = 0.0000
Log likelihood = -16.359487 Pseudo R2 = 0.8119

Repayment Odds Ratio Std. Err. z P>|z| [95% Conf. Interval]

Gender .1070979 6.926597 -0.03 0.972 9.51e-57 1.21e+54

Age 4.47386 4.199169 1.60 0.110 .7107941 28.15924
Marriage 1.217167 1.505347 0.16 0.874 .1077992 13.7431
Education 147.564 338.8306 2.18 0.030 1.638648 13288.48
Householdsize .0617984 .0861591 -2.00 0.046 .0040201 .9499905
Experience 148.0764 376.8588 1.96 0.050 1.009608 21717.96
Otherbusiness 160.8127 385.3354 2.12 0.034 1.467887 17617.66
Businessform .5145415 .5046264 -0.68 0.498 .0752699 3.517381
Businesssector .2160446 .3015863 -1.10 0.272 .014006 3.332528
Income 6.011317 9.219578 1.17 0.242 .2974944 121.4676
Loanamount 16.55887 24.90063 1.87 0.062 .8690101 315.5271
Diversion .0001309 .0006215 -1.88 0.060 1.19e-08 1.442016
Equity 4.902148 7.154379 1.09 0.276 .2806149 85.63712
Graceperiod 134.3317 4260.473 0.15 0.877 1.35e-25 1.33e+29
Followup 29.95385 59.67753 1.71 0.088 .6033791 1487.014
Collateral 15.03414 29.08845 1.40 0.161 .3389625 666.8155
Interest .1217303 .2028162 -1.26 0.206 .0046473 3.188561
Kyc 2.360878 3.432122 0.59 0.555 .1366598 40.78555
Timehorizon .001596 .0040877 -2.51 0.012 .0000105 .2416515
Market 9.315421 18.19058 1.14 0.253 .2027818 427.9332
Whether .9433498 1.122339 -0.05 0.961 .0916149 9.713582
_cons .2360275 15.29671 -0.02 0.982 1.61e-56 3.46e+54

100
 Annex 8. Hetroschedasticity tests

….imtest, white

White's test for Ho: homoskedasticity against Ha: unrestricted heteroskedasticity

chi2(100) = 108.81

Prob> chi2 = 0.2571

Source chi2 Df p
Heteroskedasticity 108.81 100 0.2571
Skewness 66.82 20 0
Kurtosis 0.1 1 0.7565

Total 175.73 121 0.0009

….hettest

Breusch-Pagan / Cook-Weisberg test for heteroskedasticity

Ho: Constant variance

Variables: fitted values of Repayment

chi2(1) = 3.4 Prob> chi2 = 0.0526

………. rvfplot
1
.5
Residuals

0
-.5
-1

0 .5 1 1.5
Fitted values

101
 Appendex 10. Kernel density normality test

Kernel density estimate

1.5
1
Density

.5
0

-.5 0 .5 1 1.5
Fitted values

Kernel density estimate

Normal density
kernel = epanechnikov, bandwidth = 0.1115

--------------------------------------------------------------End--------------------------------------------------------------
---------------------------------------------------------------------------------------------------------------------------------
---------------------------------------------------------------------------------------------------------------------------------

102