ST.

MARY’S UNIVERSITY SCHOOL

OF GRADUATE STUDIES

FACTORS AFFECTING NON-PERFORMING LOAN IN THE

CASE OF ETHIOPIAN COMMERCIAL BANKS

BY
MESERET AMSALU KASSA

June 2018
ADDIS ABABA, ETHIOPIA
A THESIS SUBMITTED TO ST. MARY’S UNIVERSITY, SCHOOL
OF GRADUATE STUDIES

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE

MBA – ACCOUNTING AND FINANCE

June 2018
ADDIS ABABA, ETHIOPIA

i
 ST.MARY’S UNIVERSITY SCHOOL OF GRADUATE STUDIES
FACULTY OF ACCOUNTING AND FINANCE

FACTORS AFFECTING NON-PERFORMING LOAN IN THE

CASE OF ETHIOPIAN COMMERCIAL BANKS

BY
MESERET AMSALU KASSA

APPROVED BY BOARD OF EXAMINERS

Dean, Graduate Studies Signature & Date

Advisor Signature & Date

External Examiner Signature & Date

Internal Examiner Signature & Date

ii
 Table of Contents
ACKNOWLEDGEMENTS ............................................................................................................ v
ACRONYMS AND ABBRIVATIONS ........................................................................................ vi
LIST OF TABLE .......................................................................................................................... vii
LIST OF FIGURE........................................................................................................................ viii
ABSTRACT ................................................................................................................................... ix
CHAPTER ONE ............................................................................................................................. 1
1. Introduction ............................................................................................................................. 1
1.1. Background of the Study ......................................................................................................... 1
1.2 Overview of banking history in Ethiopia .................................................................................. 2
Table 1.1 Lists of public and private Commercial Banks in Ethiopia ............................................ 5
1.3 Statement of the problem .......................................................................................................... 6
1.4 Research Questions ................................................................................................................... 7
1.5 Objectives of the study.............................................................................................................. 7
1.5.1 General Objective of the Study .......................................................................................... 7
1.5.2 Specific objectives of the Study ......................................................................................... 7
1.6 Research Hypothesis ................................................................................................................. 8
1.7 Significance of the study........................................................................................................... 8
1.8 Scope of the Study .................................................................................................................... 9
1.9 Limitation of the Study ............................................................................................................. 9
1.10 Organization of the paper........................................................................................................ 9
CHAPTER TWO .......................................................................................................................... 10
Review Related Literature ......................................................................................................... 10
2.1 Introduction ............................................................................................................................. 10
2.2 Basic concept of loan .............................................................................................................. 10
2.3 Non-Performing Loans .................................................................................................... 10
2.4 Measurement of non-performing loans ........................................................................... 11
2.5 Banks’ internal factors causing non-performing loans .................................................... 11

iii
 2.6 Customers related factors ................................................................................................ 14
2.7 Macroeconomic factors ................................................................................................... 15
2.8 Empirical Studies ............................................................................................................. 17
2.9 Knowledge Gap Analysis ................................................................................................ 20
2.10 Conceptual Framework.................................................................................................. 21
CHAPTER THREE ...................................................................................................................... 22
Research Methodology and Design .............................................................................................. 22
3. Introduction ........................................................................................................................... 22
3.1. Research Design ............................................................................................................. 22
3.2 Data sources and types of data ........................................................................................ 22
3.3 Population Sampling Technique of the Study ................................................................. 22
Table 2 List of selected Banks ...................................................................................................... 23
3.4 Methods of Data Analysis ............................................................................................... 24
3.5 Description of Variables .................................................................................................. 24
3.6 Model Specification ......................................................................................................... 27
CHAPTERFOUR .......................................................................................................................... 28
Data Analysis and Interpretation ............................................................................................... 28
4. Introduction .................................................................................................................... 28
4.1. Descriptive statistics ....................................................................................................... 28
4.1.2 Correlation Analysis ..................................................................................................... 40
Table 3 Correlation Analysis ................................................................................................. 41
4.1.3 Regression Model Assumption and Diagnostic Test .................................................... 41
CHAPTER FIVE .......................................................................................................................... 49
Summary, Conclusion and Recommendation of the study ........................................................... 49
5. Introduction .................................................................................................................... 49
5.1 Summary ............................................................................................................................. 49
5.2 Conclusion........................................................................................................................... 51
5.3 Recommendation ................................................................................................................. 51
References ..................................................................................................................................... 53
Appendixes ................................................................................................................................... 59

iv
 ACKNOWLEDGEMENTS

All praise and glory belongs to the Almighty God for the tremendous power and courage he gave
me in the course of my whole education endeavor. I have no words to thank him enough for all
his mercy, guidance and unlimited favors and blessings.

I would like to express my sincere gratitude to Associate. Professor. Dejene Mamo, for his
guidance, consistent encouragement and support.

I am grateful to all my family and pals for their non-stop endurance and assist in helping me
dedicate all my time to end this work.

v
ACRONYMS AND ABBRIVATIONS
NBE – National Bank of Ethiopia
CBE – Commercial Bank of Ethiopia

DB – Dashen Bank S.C.

AIB – Awash International Bank S.C.

BOA – Bank of Abyssiniya S.C.

UB – United Bank S.C.

NIB – Nib international Bank S.C.

BS – Bank Size

EXR – Exchange Rate

GDP – Gross Domestic Product

INFR – Inflation Rate

LG – Loan Growth

LIQ – Liquidity

LR – Lending Rate

ROA – Return on Asset

vi
 List of Tables
Table 1 List of selected Banks .................................................................................................................... 23

Table 2 Correlation Analysis ...................................................................................................................... 41

Table 3 Heteroscedasticity Test .................................................................................................................. 42

Table 4 Multicollinearity Test..................................................................................................................... 43

Table 5 Normality Test ............................................................................................................................... 44

Table 6 Regression analysis result between variables ................................................................................ 45

vii
 List of Figures
Figure 1 Conceptual framework ................................................................................................................. 21

Figure 2 Non-Performing Loan Trend Analysis of Commercial banks in Ethiopia ................................... 29

Figure 3 Average natural logarithm of total asset Trend of the Industry .................................................... 30

Figure 4 Average natural logarithm of total asset Trend each banks .......................................................... 31

Figure 5 Average Liquidity growth rate and trend of the studied banks from 2002 – 2016 ....................... 32

Figure 6 Average Liquidity growth rate of each banks 2002 – 2016.......................................................... 33

Figure 7 Average Loan growth trend of the studied banks ......................................................................... 34

Figure 8 Average LG ratio of each banks performance from 2002 - 2016 ................................................. 34

Figure 9 Trend of profitability of the industry ............................................................................................ 35

Figure 10 Trend of each banks performance of Profitability ...................................................................... 36

Figure 11 Inflation Trend of the country and its effect on NPLs ................................................................ 37

Figure 12 Trend of GDP and Its effect on NPLs ........................................................................................ 38

Figure 13 Average lending rate trend and its effect on NPLs of the studied banks .................................... 39

Figure 14 Trend of exchange rate and its effect on NPL ............................................................................ 40

viii
 ABSTRACT

The main objectives of the study were to examine factors affecting non-performing loans in
Ethiopian commercial banks in the period 2002 – 2016. The study used secondary data which is
audited annual financial reports of the seven selected banks. Descriptive and inferential statistics
are used to analyze the data collected. The descriptive statistic shows the trend analysis of the
dependent and independent variables by using graphical methods, while panel regression analysis
was used to identify the relative importance of each independent variables influence NPLs of
Ethiopian banks by using E-views 9 software. The finding of the study shows that the trend analysis
of dependent and independent variables are downward sloping and NPLs level indicates above
the threshold of NBE rules. The regression result shows the determinant variables are a significant
relationship with NPLs. Based on the findings the researcher forward subsequent
recommendation: each banks improve the inspection techniques and bankers must understand how
the risks of individual loans and portfolios are interrelated and the bank managers should be
ensure the credit department adequately resourced to support for monitoring activities and follow
up the borrowed fund are being used the intended purpose and timely monitor the loan is being
disbursed.

Key words: Non-performing loan, Micro-economic and internal bank factors

ix
 CHAPTER ONE
1. Introduction

Banks are very important constituents in the financial system of countries and play a
fundamental role in the global economy. Therefore, if the financial system does not work
properly, its problems have a strong impact on the whole economy (Rodriguez-Moreno,
Pena, 2013). For this reason, this study is to assessing factors affecting non-performing
loan in commercial bank of Ethiopia through selecting some commercial banks activities.
To achieve the overall objective of the study will use several techniques in the entire
chapter, in this regards in chapter one the study briefly provide the introductory parts such
as, background of the study, organizational background, statement of the problem, basic
research questions, objective of the study, scope of the study and significance of the study.

1.1. Background of the Study

Lending is the principal business activity for commercial banks. The loan portfolio is typically the
largest asset and the predominant sources of revenue. As such, it is one of the greatest sources of
risk to a bank’s safety and soundness (Richard, 2009). According to directives of national bank
of Ethiopia “loan” or advances” means any financial asset of the bank arising from the direct or
indirect advances by a bank to a person that are conditioned on the obligation of the person to
repay the fund, either on a specified date or dates usually within interest (Tsinghua, 2008) .While
performing one of its main functions granting loan, the bank is exposed to credit risk i.e. non-
performing loan; A loan that is not earning income or full payment of principal and interest is no
longer anticipated i.e. Principal or interest is 90 days or more delinquent.

The issue of non-performing loans has gained increasing attentions in because the immediate
consequence of large amount of NPL in the banking system is bank failure (Holger, 2008).
Literature reveals there are a lot of challenges that can tackle loan management system of banks
in Ethiopia such as, due to lax credit standard, poor portfolio risk management, or weakness in the
economy; loan portfolio problems have historically been the major cause of bank loss and failures
(Wondimu, 2007). Effective management of loan portfolio and the credit function is fundamental
to bank’s safety and soundness. Loan portfolio management (LPM) is the process by which risks

1
that are inherent in the credit process are managed and controlled. Healthy loan portfolios are vital
assets for banks in view of their positive impact on the performance of banks. Unfortunately, some
of these loans usually do not perform and eventually result in nonperforming which affect banks’
earnings on such loans (Fofack, 2005).

These non-performing loans become cost to banks in terms of their implications on the quality of
their assets portfolio and profitability. This is because in accordance with banking regulations,
banks make provisions for non-performing loans and charge for NPLs which reduce their loan
portfolio and income (Bloem & Gorter, 2001).

The rise of non-performing loan portfolios in banks significantly contributed to financial distress
in the banking sector. Non-performing loans are the main contributor to liquidity risk, which
exposes banks to insufficient funds for operations. As loans & advances are the major portion of
bank’s asset, when they become non-performing, it will affect both profitability and liquidity of
the bank.

The minimization of NPL is a necessary condition for improving economic growth. When NPL
retained permanently, these will have an impact on the resources that are enclosed in unprofitable
areas. Thus, NPL are likely to hamper economic growth and reduce the economic efficiency (Hou,
2007). The shocks to the financial system can arise from factors specific to the company
(idiosyncratic shocks) or macroeconomic imbalances (systemic shocks). In general, the researches
adopted in Ethiopia commercial banks affected by several types of factors whether through internal
or external factors , therefore, investigating and studying factors affecting commercial banks non-
performing loan is important.

1.2 Overview of banking history in Ethiopia

The agreement that was reached in 1905 between Emperor Minilik II and Mr. Ma Gillivray,
representative of the British owned National Bank of Egypt marked the introduction of modern
banking in Ethiopia. Following the agreement, the first bank called Bank of Abyssinia was
inaugurated in Feb.16, 1906 by the Emperor. The Bank was totally managed by the Egyptian

National Bank.

Generally, in its short period of existence, Bank of Abyssinia had been carrying out limited
business such as keeping government accounts, some export financing and undertaking various

2
tasks for the government. Moreover, the Bank faced enormous pressure for being inefficient and
purely profit motivated and reached an agreement to abandon its operation and be liquidated in
order to disengage banking from foreign control and to make the institution responsible to
Ethiopia’s credit needs. Thus by 1931 Bank of Abyssinia was legally replaced by Bank of

Ethiopia shortly after Emperor Haile Selassie came to power.

The new Bank, Bank of Ethiopia, was a purely Ethiopian institution and was the first indigenous
bank in Africa (NBE 2009/2010) and established by an official decree on August 29, 1931 with
capital of £750,000. Bank of Egypt was willing to abandon it’s on cessionary rights in return for a
payment of Pound Sterling 40,000 and the transfer of ownership took place very smoothly and the
offices and personnel of the Bank of Abyssinia including its manager, Mr. Collier, being retained
by the new Bank. Ethiopian government owned 60 percent of the total shares of the

Bank and all transactions were subject to scrutiny by its Minister of Finance.

Bank of Ethiopia took over the commercial activities of the Bank of Abyssinia and was authorized
to issue notes and coins. The Bank with branches in Dire Dawa, Gore, Dessie, Debre

Tabor, Harar, agency in Gambella and a transit office in Djibouti continued successfully until the

Italian invasion in 1935. During the invasion, the Italians established branches of their main

Banks namely Banco di Italia, Banco di Roma, Banco di Napoli and Banco Nazionale del lavoro
and started operation in the main towns of Ethiopia. However, they all ceased operation soon after
liberation except Banco di Roma and Banco di Napoli which remained in Asmara. In 1941 another
foreign bank, Barclays Bank, came to Ethiopia with the British troops and organized banking
services in Addis Ababa, until its withdrawal in 1943. Then on 15th April 1943, the

State Bank of Ethiopia commenced full operation after 8 months of preparatory activities. It acted
as the central Bank of Ethiopia and had a power to issue bank notes and coins as the agent of the
Ministry of Finance. In 1945 and 1949 the Bank was granted the sole right of issuing currency and
deal in foreign currency. The Bank also functioned as the principal commercial bank in the country
and engaged in all commercial banking activities.

The State Bank of Ethiopia had established 21 branches including a branch in Khartoum, Sudan
and a transit office on Djibouti until it ceased to exist by bank proclamation issued on December,

3
1963. Then the Ethiopian Monetary and Banking law that came into force in 1963 separated the
function of commercial and central banking creating National Bank of Ethiopia (NBE) and
commercial Bank of Ethiopia (CBE). Moreover it allowed foreign banks to operate in Ethiopia
limiting their maximum ownership to be 49 percent while the remaining balance should be owned
by Ethiopians.

There were two other banks in operation namely Banco di Roma S. C. and Bank of di Napoli

S.C. that later reapplied for license according to the new proclamation each having a paid up capital
of Eth. Birr 2 million. The first privately owned bank, Addis Ababa Bank S.C., was established on
Ethiopians initiative and started operation in 1964 with a capital of 2 million in association with
National and Grindlay Bank, London which had 40 percent of the total share. In 1968, the original
capital of the Bank rose to 5.0 million and until it ceased operation, it had 300 staff at 26 branches.

There were other financial institutions operating in the country like the Imperial Savings and

Home Ownership Public Association (ISHOPA) and Saving and Mortgage Corporation of

Ethiopia (SMCE). But following the declaration of socialism in 1974 the government extended its
control over the whole economy and nationalized all large corporations. Organizational setups
were taken in order to create stronger institutions by merging those that perform similar functions.
Accordingly, the three private owned banks, Addis Ababa Bank, Banco di Roma and

Banco di Napoli Merged in 1976 to form the second largest Bank in Ethiopia called Addis Bank
with a capital of Eth. birr 20 million and had a staff of 480 and 34 branches. Before the merger,
the foreign participation of these banks was first nationalized in early 1975. Then Addis Bank

S.C. and Commercial Bank of Ethiopia were merged by proclamation No.184 of August 2, 1980

to form the sole commercial bank in the country till the establishment of private commercial banks
in 1994. The Commercial Bank of Ethiopia commenced its operation with a capital of Birr

65 million, 128 branches and 3,633 employees. The Savings and Mortgage Corporation S. C and

Imperial Saving and Home Ownership Public Association were also merged to form the Housing
and Saving Bank with working capital of Birr 6 million and all rights, privileges, assets and
liabilities were transferred by proclamation No.60, 1975 to the new bank.

4
Following the fall of the Dergue regime in 1991 that ruled the country for 17 years under the rule
of command economy, the EPRDF declared a liberal economy system. In line with this,

Monetary and Banking proclamation of 1994 established the National Bank of Ethiopia as a
judicial entity, separated from the government and outlined its main function.

Monetary and Banking proclamation No.83/1994 and the Licensing and Supervision of Banking

Business No.84/1994 laid down the legal basis for investment in the banking sector.

Consequently after the proclamation issued private equity holders began to join the Ethiopian
banking industry and as of today seventeen commercial banks are operated and out of this sixteen
are private owned.

Table 1.1 Lists of public and private Commercial Banks in Ethiopia

S.N Name of bank Year of esta Ownership

1 Commercial Bank of Ethiopia (CBE) 1963 Public

2 Awash International Bank S.C (AIB) 1994 Private

3 Dashen Bank S.C (DB) 1995 Private

4 Bank of Abyssinia S.C (BoA) 1996 Private

5 Wegagen Bank S.C (WB) 1997 Private

6 United Bank S.C (UB) 1998 Private

7 Nib International Bank S.C (NIB) 1999 Private

8 Cooperative Bank of Oromia S.C (CBO) 2005 Private

9 Lion International Bank S.C (LIB) 2006 Private

10 Oromia International Bank S.C (OIB) 2008 Private

11 Zemen Bank S.C (ZB) 2009 Private

12 Bunna International Bank S.C (BIB) 2009 Private

13 Berhan International Bank S.C (BBI) 2010 Private

5
 14 Abay Bank S.C. (AB) 2010 Private

15 Addis international Bank SC. (AdIB) 2011 Private

16 Debub Global Bank S.C. (DGB) 2012 Private

17 Enat Bank S.C. (EB) 2013 Private

Source NBE annual report 2015/16

1.3 Statement of the problem

According to directives of national bank of Ethiopia “loan” or advances” means any financial asset
of the bank arising from the direct or indirect advances by a bank to a person that are conditioned
on the obligation of the person to repay the fund, either on a specified date or dates usually within
interest (Directives of National Bank of Ethiopia 2008) .While reforming one of its main functions
granting loan, the bank is exposed to credit risk i.e. non-performing loan; A loan that is not earning
income or full payment of principal and interest is no longer anticipated i.e. Principal or interest is
90 days or more delinquent.

The rise of non-performing loan portfolios in banks significantly contributed to financial distress
in the banking sector. Non-performing loans are the main contributor to liquidity risk, which
exposes banks to insufficient funds for operations. As loans & advances are the major portion of
bank’s asset, when they become non-performing, it will affect both profitability and liquidity of
the bank (Hou, 2007)

There are two factors affecting the occurrence of non-performing loans, namely macroeconomic
factors and bank specific factors. Macroeconomic factors include government policy, inflation,
currency change and GDP and the bank specific factors include interest rate charged by banks,
bank size, ownership (state owned and private), integrity problem, credit follow up weakness, and
others related problems (Ricardas, 2014).

According to National Bank of Ethiopia’s (NBE), report in 2015/16 there is NPL performance in
commercial banks of Ethiopia even their threshold average is varies year to year but, the result of
NBE only indicate average result of each banks based on the standard of the NBE, however, it was
not indicated factors that affect each banks of non-performing loan practice. These is one of the
knowledge gap that motivated the researcher to assess factor affecting nonperformance loan of
commercial banks. In addition to NPL result of NBE, the study also assess several research results
6
to reduced similarity, accordingly there are several research results related with NPL, however,
there were no exact similar research results except the study took place by, Wondimu (2012)
conducted a study on determinants of nonperforming loans and found as poor credit assessment
,failed loan monitoring, underdeveloped credit culture, lenient credit terms and conditions,
aggressive lending, compromised integrity, weak institutional capacity, unfair competition among
banks, and fund diversion for un expected purposes and overdue financing had an effect on the
occurrence of NPLs. Even though as to the knowledge of the researcher, there is only a single
study made by Wondimu (2012) in Ethiopia which is related with this title and in addition there is
time gap between the researches mad by Wondimu and the proposed study of this paper.

Thus, given the unique features of banking sector and environment in which they operate and also
rapid expansion of banking institutions in Ethiopia, there is strong wishes to conduct a study on
the identification of factors affecting NPLs of commercial banks in Ethiopia.

1.4 Research Questions

The following specific research questions were formulated to fill the above gaps

1. What are the internal factors that affect NPLs of commercial Banks of Ethiopia?
2. What are the External (Macroeconomic) factors affecting NPLs of commercial Banks of
Ethiopia
3. What seem strands of Commercial banks of Ethiopia regarding NPLs performance?

1.5 Objectives of the study

1.5.1 General Objective of the Study
Objectives of the study is assessing factors affecting NPLs of commercial banks of Ethiopia. Non-
performing loans proportion is one of the determinant factors that depict soundness of the banking
sector. Thus, identifying and investigating the determinants of nonperforming loans is the central
objective of this study.

1.5.2 Specific objectives of the Study

The following are the specific objective of the study:

1. To assess internal factors that affect NPLs of commercial Banks of Ethiopia

7
 2. To assess the External (Macroeconomic) factors affecting NPLs of commercial Banks of
Ethiopia
3. To examine strands of Commercial banks of Ethiopia regarding NPLs performance

1.6 Research Hypothesis

The purpose of this study is to examine the determinants of nonperforming loans (NPLs) of
commercial banks in Ethiopia. The empirical studies made around the world demonstrate various
outcomes on determinants of nonperforming loans of the financial sectors divided in to internal
and external factors as discussed in the major research questions part of the study. Accordingly the
study hypothesized the following major areas as a determinate variables (Null hypothesis).

H1.Bank size (BS) has positive relation with NPLs of commercial banks in Ethiopia.
H2.Exchange rate (EXR) has negative relation with NPLs of commercial banks in Ethiopia.
H3.Gross domestic product (GDP) has a positive/negative relation with NPLs of
commercial banks in Ethiopia
H4.Inflation rate (INF) has positive/negative relation with NPLs of commercial banks in
Ethiopia
H5. Loan growth (LG) has positive relation with NPLs of commercial banks in Ethiopia.
H6.Liquidity (LIQ) has positive/negative relation with NPLs of commercial banks in
Ethiopia.
H7. Lending rate (LR) has positive relation with NPLs of commercial banks in Ethiopia.
H8.Return on asset (ROA) has positive relation with NPLs of commercial banks in
Ethiopia

1.7 Significance of the study

The finding of this study which details with the determinants of nonperforming loan of
Commercial banks in Ethiopia are beneficial for different stakeholders such as, for academicians,
bank sectors, as well as for the researcher to develop technical knowhow of how academic
research’s developed technically. In addition, since such investigation has policy implication, the
finding of this study might be used as a directive input in developing regulatory standards
regarding the lending policies of commercial banks of Ethiopia. This study initiate the commercial
Bank management to give due emphasis on the management of the identified variables and

8
provides them with understanding of activities to enhance their loan performance indicate which
factors more affecting the environment.

1.8 Scope of the Study

This study delimited in scope of delimited commercial banks, issues that had been discussed as
well as panel data. Accordingly, the study limit this study the commercial banks found in Ethiopia
namely commercial bank of Ethiopia, Awash international bank, Dashen bank, bank of Abyssinia,
United bank, and Nib International bank and Wegagen bank that, this is because the banks have
similar lending experience in Ethiopia as they joined early in the market. The study also specified
in the issued that had been discussed such as, internal (specific) factors, that could arise from the
banks strategies, capacity, competitiveness and others related factors determined by each banks
practices and the macroeconomics factors that can affect the banks activities such as, inflation and
GDP. The study also assess past trend of each banks performance of NPLs (2002 – 2016 G.C)
collecting from each banks Annual report as well as NBE report.

1.9 Limitation of the Study

Due to the confidential policy of banks, only access officially disclosed financial information. The
study was also limited to financial constraints and shortage of time forced the researcher to
minimize scope and reduce sample size which lessened richness of the outcome and in turn affects
the generality of the findings into the whole banking practice.

1.10 Organization of the paper

This thesis is organized into five chapters. The first chapter starts with presenting background of
the study, statement of the problem, objective of the study, significance of the study, scope and
limitation of the study. The second chapter focuses on both theoretical and empirical review of
related literature. The third chapter deals with the research methodology. Chapter four deals with
the data analysis and presentation and the fifth chapter contain the conclusion and recommendation
of the study including the direction for further study

9
 CHAPTER TWO
Review Related Literature
2.1 Introduction
This chapter starts with presenting the overview of banking system in Ethiopia. Besides, bank
loans including it determinant factors were presented. Furthermore, concepts relating to
nonperforming loans are discussed. Following this, empirical studies (cross countries and single
country) are reviewed by focusing on determinants of NPLs are presented.

2.2 Basic concept of loan

Lending Bank loans finance different corporate groups in the economy. Manufacturers,
distributors, service firms, farmers, builders, homebuyers, commercial real estate developers,
retailers, and others all depend on bank credit (Wondimu, 2012). The ways in which banks allocate
their funds strongly influences the economic development of the community and nation. Every
bank bears a degree of risk in its granting of credit, and, without exception, every bank experiences
some loan losses when certain borrowers fail to repay their loans. Whatever the degree of risk
taken loan, losses can be minimized through highly professional organization and management of
the lending functions. The composition and quality of a bank’s loans should be reflected in its loan
policy. The policy sets out the bank’s lending philosophy and specifies procedures and means of
monitoring lending activity (Holger, 2008).

2.3 Non-Performing Loans

The term ‘’bad loans’’ as described by Basu (1998), is used interchangeably with nonperforming
and impaired loans as identified in Fofack (2005). (Berger and De Young, 1997) also considers
these types of loans as “problem loans”. Thus these descriptions are used interchangeably
throughout the study. Generally, loans that are outstanding in both principal and interest for a long
time contrary to the terms and conditions contained in the loan contract are considered as non-
performing loans. This is because going by the description of performing loans above, it follows
that any loan facility that is not up to date in terms of payment of both principal and interest
contrary to the terms of the loan agreement, is non-performing. Available literature gives different
descriptions of bad loans. Some researchers noted that certain countries use quantitative criteria
for example number of days overdue scheduled payments while other countries rely on qualitative

10
norms like information about the customer’s financial status and management judgment about
future payments (Bloem and Gorter, 2001).

Alton and Hazen (2001) described non-performing loans as loans that are ninety days or more past
due or no longer accruing interest (Caprio and Klingebiel 1999), cited in Fofack (2005), consider
non-performing loans as 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. A
non-performing loan may also refer to one that is not earning income and full payment of principal
and interest is no longer anticipated, principal or interest is ninety days or more delinquent or the
maturity date has passed and payment in full has not been made (Fofack, 2005). A critical appraisal
of the foregoing definitions of bad loans points to the fact that loans for which both principal and
interest have not been paid for at least ninety days are considered non-performing.. Therefore any
loan that is outstanding for ninety days or more is considered a non-performing loan. According
to (Berger and De Young 1997), such loans could be injurious to the financial performance of
banking institutions.

2.4 Measurement of non-performing loans

In recent years the global financial crisis and the subsequent recession in many developed countries
have increased households’ and firm defaults, causing significant losses for banks Khon and Best
(2007). In this study the non-performing loans will be measured based on banks internal factors
and customer related factors.

2.5 Banks’ internal factors causing non-performing loans

These internal factors affect lending behavior of the bank. Many literature review have examined
the connection between these factors and NPLs in KCB Bank Kenya Limited. Literature on banks
internal factors that affects non-performing loans are reviewed in the following Bank’s loan
supervision capacity

The impact of bank’s loan supervision capacity on NPLs is extensively documented in the
literature. In fact, several studies report that bank’s loan supervision capacity is positively related
to NPLs (Abafita, 2003, Aballey; 2009, Kagimba; 2010). According to these studies the
relationship means that a good supervision capacity contributes to lower non- performing loans
and bad supervision capacity increases non- performing loans

11
Asset Quality
According to Grier (2007), “poor asset quality is the major cause of most bank failures”. A most
important asset category is the loan portfolio; the greatest risk facing the bank is the risk of loan
losses derived from the delinquent loans. The credit analyst should carry out the asset quality
assessment by performing the credit risk management and evaluating the quality of loan portfolio
using trend analysis and peer comparison. Measuring the asset quality is difficult because it is
mostly derived from the analyst’s subjectivity.

Management Quality
Management quality is basically the capability of the board of directors and management, to
identify, measure, and control the risks of an institution’s activities and to ensure the safe, sound,
and efficient operation in compliance with applicable laws and regulations William F. Caton,
(1997) The top management with good quality and experience has preferably excellent reputation
in the local communication. Management relates to the competency of the bank’s managers, using
their expertise’s to make subjective judgments, create a strategic vision, and other relevant
qualities. Management is the key variable which determines a banks‟ success. The evaluation of
the management is the hardest one to be measured and it is the most unpredictable (Golin, 2001).

Liquidity
There should be adequacy of liquidity sources compared to present and future needs, and
availability of assets readily convertible to cash without undue loss. The fund management
practices should ensure an institution is able to maintain a level of liquidity sufficient to meet its
financial obligations in a timely manner; and capable of quickly liquidating assets with minimal
loss. The liquidity ratio expresses the degree to which a bank is capable of fulfilling its respective
obligations. Banks makes money by mobilizing short-term deposits at lower interest rate, and
lending or investing these funds in long-term at higher rates, so it is hazardous for banks
mismatching their lending interest rate (Holger, 2008).

12
Return on Asset
It is an important indicator of the performance of a bank since it determines the profitability of the
bank based on its assets. Growing NPLs slowdown interest earning capacity due to their non-
recognition of interest and, on the other hand, provision for NPLs increases interest suspense but
reduces realized profits. In the context of emerging market economics, the findings of Godlewski
(2004) indicated that there is a negative impact of return on assets on the level of non-performing
loans.

Capital Adequacy
One of the examined financial factors is the capital adequacy ratio. It measures the risk that a bank
can undertake. Capital adequacy ratio is calculated by adding tier 1 capital to tier 2 capital and
dividing by risk weighted assets which is guided by Basel accord. Generally, capital adequacy
ratios widely used in similar studies are not clear whether they affect positively or negatively to
the aggregate NPLs (Sinkey and Greenawlat 1991). According to Mukherjee (2003), the presence
of large amount of NPLs is responsible for the decline in the profit margin of many banks.

Bank size
Rajan & Dahl (2003) in their study of commercial banks in India they used panel regression
analysis. Their study also indicates that bank size have significance on occurrence of NPLs. Sala
& Saurina (2003) indicated that bank size is among the factors that explained variations in NPLs
for Spanish banks. Besides, Bikker & Hu (2002) also shows that bank size is significantly related
rate of occurrence of loan default.

Loan Growth
Many studies indicate that loan delinquencies are associated with rapid credit growth. Keeton
(1999) used data from commercial banks in the United States from 1982 to 1996 and a vector auto
regression model indicates this association between loan and rapid credit growth. Sinkey and
Greenwalt (1991) also studied large commercial banks in the US and found out that excessive
lending explains loan loss rate. Study of Bercoff et al (2002) shows that asset growth explains
NPLs.

Interest rate
The commercial banks that charge high interest rate would relatively face a high loan default rate.
A study by Waweru & Kalini (2009) on commercial banks in Kenya using statistical analysis

13
indicates that high interest rate charged by the banks is one of the internal factors that leads to
incidence of non-performing loans. Bikker & Hu (2002) on 29 OECD countries, banks profit
margin demonstrated by high interest rate affects occurrence of NPLs. In fact several studies report
that high interest rate and non- performing loans are positively related (Sinkey and Greenwalt,
1999, Ewert, schenk and Szczesny, 2000, Fofack, 2005, Jimenez and Saurina, 2005, Mwakoba,
2011). The explanation provided by the literature is that banks charge high interest rates when they
perceive higher risk of default. This situation attracts bad borrowers to borrow, therefore,
increasing chances of loan default. Conversely, banks charge low interest rate when they perceive
low risk of default. This causes more borrowers to borrow money from banks.

Lending rates
Lending rates are one of the essential financial determinant of nonperforming loans/bad credits.
According to Glen and Mondragon-Velez (2011), changes of lending rate
will influence the capacity of borrowers to continue paying interest for the loan borrowed. When
economies develop strongly, bank will not anticipate abnormal deterioration in
their credit portfolio execution. This is because only a small portion of loans will go
default. However, in the event that the recession happens, borrowers may not be able to pay for
the interest of the loan borrowed, so at this time the probability of default in loan increases. In this
way, they accept that loan default positively related to lending rate.

2.6 Customers related factors

These are factors affecting loan repayment behavior of the borrowers. Customer failure to disclose
vital information during the application process leads to occurrence of non-performing loans
(Brown Bridge 1998). The following are some of the customer specific factors; diversion of funds
by the borrower from the intended purpose, death of the borrower, loss of a job, age and gender
among other factors contributes to loan.

Credit Information
Adequate and timely information that enables a satisfactory assessment of the creditworthiness of
borrowers applying for a bank loan is crucial for making prudent lending decisions. Prudent
lending decisions made on the basis of adequate information on the creditworthiness of borrowers
are one of the principal factors in ensuring the financial soundness of banks (Wondimu, 2007).
But, there has been serious difficulty in Ethiopia of getting accurate and timely information on

14
prospective borrowers that facilitates the making of such prudent lending decisions. One of the
means for alleviating this difficulty of getting accurate and timely information on prospective
borrowers is the establishment of a Credit Information Center (CIC) where relevant information
on borrowers is assumed to be pooled and made available to lending banks (Abdu, 2004).
According to article 36 of the Licensing and Supervision of Banking Business Proclamation No.
84/2002, the National Bank Ethiopia (NBE) has issued these directives to establish such a Credit
Information Center (CIC). Though there is still serious limitations in the accuracy of the credit
information extracted the summary of the directive is as follows:

Credit Process
The fundamental objective of commercial and consumer lending is to make profitable loans with
minimal risk. Management should target specific industries or markets in which lending officers
have expertise. The credit process relies on each bank’s systems and controls to allow management
and credit officers to evaluate risk and return tradeoffs (Charles Smithson, 2003).

Credit Approval and Implementation

The individual steps in the credit approval process and their implementation have a considerable
impact on the risks associated with credit approval. The quality of credit approval processes
depends on two factors, i.e. a transparent and comprehensive presentation of the risks when
granting the loan on the one hand, and an adequate assessment of these risks on the other.
Furthermore, the level of efficiency of the credit approval processes is an important rating element.
Due to the considerable differences in the nature of various borrowers and the assets to be financed
as well as the large number of products and their complexity, there cannot be a uniform process to
assess credit risks (Holger, 2008).

2.7 Macroeconomic factors

Large number of the literatures indicates the linkage between the phases of the business cycle with
banking stability. Macroeconomic stability and banking soundness are inexorably linked.
Economic theory and other evidences strongly indicate that instability in the macroeconomic is
associated with instability in banking and financial markets and vice versa. The relation between
the macroeconomic environment and loan quality has been investigated in the literature linking
the phase of the business cycle with banking stability. In this line of research the hypothesis is
formulated that the expansion phase of the economy is characterized by a relatively low number

15
of NPLs, as both consumers and firms face a sufficient stream of income and revenues to service
their debts. However as the booming period continues, credit is extended to lower-quality debtors
and subsequently, when the recession phase sets in, NPLs increase (Fisher 1999).

Gross Domestic Product (GDP)

According to Salas and Saurina (2002) there is a significant negative concurrent effect of GDP
growth on the NPL ratio and infer a quick transmission of macroeconomic developments to the
ability of economic agents to service their loans. The clarification given by the writing for this
relationship is that, Changes in business cycle affect the credit value of borrowers in terms of
reimbursement capacity. Consequently, solid positive development in genuine GDP as a rule
interprets into more pay which makes strides the obligation overhauling capacity of borrower
which in turn contributes to lower NPLs. Then again, when there is moderate down in the economy
low or negative GDP development), the financial exercises in common are diminishing and the
volume of cash held for either businesses or families is diminishing. These conditions contribute
in falling apart the capacity of borrowers to reimburse the advances, which lead to increment the
probability of delays their budgetary commitments and hence banks‟ introduction to credit hazard
increment. In this respect, Hou (2006) famous that, each NPL in the monetary division is seen as
a front-side reflect picture of a sickly unbeneficial venture.

Inflation:
Like GDP and exchange rates, inflation influences borrower’s obligation overhauling capacity
through diverse channels and its affect on NPL can be positive or negative (Fofack 2005, Pasha
and Khemraj (2009) and Nkusu 2011). The clarification given by the writing for this relationship
is that, higher expansion can make obligation overhauling less demanding by decreasing the
genuine esteem of extraordinary advances especially when the credit rates are settled (banks do
not adjust rates in understanding to the inflation alter to preserve their genuine returns). However,
it can additionally weaken some borrower’s potential to provider debt by means of reducing real
income. Besides, when advance rates are variable(adjusted in understanding to the inflation alter),
inflation is likely to diminish borrower’s advance overhauling capacity as lenders alter rates to
preserve their genuine returns or essentially to pass on increments in arrangement rates coming
about from financial approach activities to combat expansion. Against this foundation, the
relationship between NPL and inflation can be positive or negative.

16
Exchange Rates
Exchange rate are influences borrower’s obligation overhauling capacity through diverse channels
and its affect on NPL can be positive or negative (Nkusu 2011). As famous in Pasha and Khemraj
(2009), deterioration of the trade rate can have blended suggestions on borrower’s obligation
overhauling capacity. On the one hand, it can progress the competitiveness of export-oriented
firms. As long as the esteem of household money deteriorated (lower), export-oriented firms can
rule the worldwide showcase at lower cost (since their production fetched is secured in household
money which has lower esteem than foreign currency and their income is collected in foreign cash
which has higher esteem as compared to the residential cash. Subsequently, devaluation of trade
rate can move forward the debt-servicing capacity of export-oriented borrowers. On the other hand,
it can unfavorably influence the debt-servicing capacity of borrowers who borrow in outside cash
(import-oriented firms).

2.8 Empirical Studies

In this part the study will discussed an empirical studies conducted by several authors both in
Ethiopia and outside of Ethiopia. Shingjergji (2013) studied the impact of different bank specific
factors on non-performing loans of Albanian banks by taking quarterly data from 2002-2012.
Dependent variable used in the study is non-performing loans (NPLs) while independent variables
include capital adequacy ratio (CAR), loan to asset ratio (LTA), return on equity (ROE), natural
log of total loans, and natural log of net interest margin (NIM). Regression results obtained by
using ordinary least square revealed negative insignificant relation of CAR with NPLs. Relation
of loan to asset ratio has been found negative but total loans level is positively influencing the
NPLs means increased loans level will result in increased level of NPLs. On the other hand, NIM
and ROE are negatively linked with NPLs depicting that high NPLs deteriorate the performance
of banks (Kirui, 2014).

Louzis et al., (2010) conduct study to examine the determinants of NPLs in the Greek financial
sector using fixed effect model from 2003-2009 periods. The variables included were ROA, ROE,
solvency ratio, loan to deposit ratio, inefficiency, credit growth, lending rate and size, GDP growth
rate, unemployment rate and lending rates. The finding reveals that loan to deposit ratio, solvency
ratio and credit growth has no significant effect on NPLs. However, ROA and ROE has negative
significant effect whereas inflation and lending rate has positive significant effect on NPLs. It

17
justifies that performance and inefficiency measures may serve as proxies of management quality
(Gadise, 2014).

Ranjan & Chandra (2003) analyze the determinants of NPLs of commercial banks’ in Indian in
2002. The objective of the study was to evaluate how NPLs influenced by financial and economic
factors and macroeconomic shocks. In the study, they utilized panel regression model and found
that lending rate also have positive impact on the NPLs justifying that the expectation of higher
interest rate induced the changes in cost conditions to fuel and further increase in NPLs. Besides,
loan to deposit ratio had negative significant effect on NPLs justifying that relatively more
customer friendly bank is most likely face lower defaults as the borrower will have the expectation
of turning to bank for the financial requirements (Gadise, 2014).

Salas and saurian (2002) investigated the determinates of problem loans of Spanish commercial
and saving banks using a dynamic model and panel dataset covering the period 1985-1997. The
finding of the study was that real growth in GDP, rapid credit expansions, bank size, capital ratio
and market power all explain variation in non-performing loans with a panel dataset covering the
period 1996-1999, used a regression analysis and analyzed the relationship between NPLs and
ownership structure of commercial banks in Taiwan. The study showed that banks with higher
government ownership recorded lower non-performing loans. The finding of the study showed
that bank size is negatively related to NPLs while diversification may not be determinant. This
study was only limited to commercial banks of Taiwan.

Tomak (2013) conducted study on the “Determinants of Bank’s Lending Behavior of commercial
banks in Turkish” for a sample of eighteen from 25 banks. The main objective of the study was to
identify the determinants of bank`s lending behavior. The data was covered 2003 to 2012 periods.
The variables used were size, access to long term funds, interest rates, GDP growth rate and
inflation rate. The finding reveals that bank size, access to long term loan and inflation rate have
significant positive impact on the bank`s lending behavior but, interest rates and GDP are
insignificant (Gadise, 2014).

Wambugu (2010) sought to determine the relationship between non-performing loans management
practices and financial performance of commercial banks in Kenya and he use a causal design, and
population of all 43 commercial banks in Kenya. The study concluded that there is need for
commercial bank to adopt non-performing loans management practices. Such practices include;

18
ensuring sufficient collaterals, limiting lending to various kinds of businesses, loan securitization,
ensuring clear assessment framework of lending facilities and use of procedures in solving on
problematic loans among others. This study used causal effect design to study the relationship
between non-performing loans which was the independent variable and financial performance
which was the dependent variable. However, the study did not determine the effect of non-
performing loans on financial performance.

Hong and Sung (1995) have tried to analyze Korean banks’ performance which was reflected on
their financial statements and to provide some comments to improve their banking business. The
study was carried out by comparing the eight Korean banks’ past five years performance results
with other banks in the State of California, other banks include Asian banks other than Korean
banks owned by such Asians (e.g., Chinese and Japanese) and American banks owned by other
ethnic groups of Americans (e.g., “white” American). The comparative financial analysis indicated
that Korean banks were relatively conservative in managing operations and lending and were more
actively involved in their services for international business and sales activities. The analyses also
indicated that the Korean banks’ loan quality was relatively low and their loan market appears to
have been saturated. They recommend on the basis of the analysis that the Korean banks should
adopt a more active marketing strategy to expand and create their own market, consider tighter
control for their operations with understanding banking regulations (e.g., Financial Institutions
Reform, Recovery, and Enforcement Act) and adopt the loan policy in a way that they can make a
loan decision with more reliable cash flow analysis.

Abdus (2004) has examined empirically the performance of Bahrain's commercial banks with
respect to credit (loan), liquidity and profitability during the period 1994-2001. Nine financial
ratios (Return on Asset, Return on Equity, Cost to Revenue, Net Loans to Total Asset, Net Loans
to Deposit, Liquid Asset to Deposit, Equity to Asset, Equity to Loan and Non-performing loans to
Gross Loan) were selected for measuring credit, liquidity and profitability performances. By
applying these financial measures, this paper found that commercial banks' liquidity performance
was not at par with the Bahrain banking industry. Commercial banks are relatively less profitable
and less liquid and, are exposed to risk as compared to banking industry. With regard to asset
quality or credit performance, this paper found no conclusive result. Non-performing loans to gross
loans (NPLGL) indicates that there was no difference in performance between the commercial
banks and the banking industry in Bahrain. Chowdhury and Ahmed (2007) have tried to analyze
19
the development and growth of selected private Commercial Banks of Bangladesh. It was observed
that all the selected private commercial banks were able to achieve a stable growth of branches,
employees, deposits, loans and advances, net income and earnings per share during the period of
2002-2006. Seven trend equations have been tested for different activities (growth in branch,
employees, deposits, loans and advances, net income and earnings per share) of the private
commercial banks. Among them the trend value of branches, employees, deposits and net income,
were positive in case of all the selected banks. The above empirical review of literature emphasizes
that all the studies so far conducted are mainly discussing the loan recovery problems, determinant
factors for default of borrowers in financial institutions in general at Macro-level (Bloem, and
Gorter ,2001).

2.9 Knowledge Gap Analysis

The knowledge gap analysis of this study was done based on the studies conducted in Ethiopia.
Accordingly, the study were assessed several published and unpublished research result related to
NPL. There some few studied conducted in Ethiopia regarding the studied areas such as, a study
conducted by Negera (2012) sought to find out the determinants of non-performing loans in the
case of Ethiopian banks, using a causal design and a population of all banks in Ethiopia were
include in the study. The findings of the study shows that poor credit assessment, failed loan
monitoring, under developed credit culture, lenient credit terms and conditions, aggressive lending,
compromised integrity, weak institutional capacity, unfair competition among banks, willful
default by borrowers and their knowledge limitation, fund diversion for unintended purpose,
over/under financing by banks ascribe to the causes of loan default. However the study did not
consider the relationship between non-performing loans, indicate determinates of NPL using
empirical data rather the study were depend on primary data like questioner and interview.

Habtamu (2015) sought to find out bank specific factors affecting occurrence of NPLs in Ethiopian
private banks. A survey study research design of six private Banks was employed in his study.
Accordingly the findings of the study shows that the major factors affecting NPLs were poor credit
assessment, poor loan follow up, undeveloped credit culture, lenient credit terms and conditions,
knowledge limitation, compromised integrity, unfair competition among banks, fund diversion for
unintended purpose, shareholders influences are bank specific factors ascribed to the occurrence
of loan default. On the other hand the finding of the document does not support that Bank size,

20
credit growth, and interest rate charged by banks have relationship with the occurrence of non-
performing loans. Based on the opinion of the respondents and interviewees‟ argument, the
findings shows that occurrence of NPLs had high influence on the profitability of banks and it
scarce the existence of credit of banks to the needy customers. The study focused on bank specific
factors affecting occurrence of NPLs in Ethiopian private banks. However the study did not
consider the relationship between non-performing loans and major internal and external factors
that determine non- performing loan.

2.10 Conceptual Framework

The conceptual frame work which describes the relationship between NPL with internal
bank factors and macroeconomic factors based on the theoretical and empirical
perspectives was formulated as follows:-

Figure 1 conceptual framework

Dependent variable Independent variables

GDP
Macro-economic
Exchange Rate
Factors
Inflation Rate

None Performing Bank Size

Loan
Loan Growth

Liquidity Bank Factors

Lending Rate

ROA

21
 CHAPTER THREE
Research Methodology and Design
3. Introduction
This study aims to examine the determinants of NPLs in the commercial banks found in Ethiopia.
Accordingly, this chapter discussed the research procedure that is used to carry out this study. In
case, it starts by discussing research design followed by the nature and instruments of data
collection and sampling design. The subsequent section presents and discusses method of data
process and analysis.

3.1. Research Design

The essential point of this examination was to examine the determinants of NPLs in Ethiopian
commercial banks with a specific end goal to achieve the objective, explanatory and descriptive
type of research design was employed. This type of research design helps to identify and evaluate
the causal relationships between the different variables under consideration (Creswell, 2009). So,
the explanatory and descriptive research design was employed to examine the relationship of the
dependent and independent variables and also the present study enabled to describe the trend of
variables from the years 2002 to 2016.

3.2 Data sources and types of data

The types of data that used in this study are panel data and Quantitative in nature. Balanced panel
data meaning that each cross sectional units have same number of time series observations. The
study has highly focused on secondary data source from the audited annual financial report (2002
– 2016) from National Bank of Ethiopia and MOFEC, journals, articles, internet and books, online
information which is relevant to explain the factors affecting bank’s NPL.

3.3 Population Sampling Technique of the Study

In this research, the target population is the banking sector in Ethiopia. According to NBE annual
report (2015/16), Ethiopia consists of 17 Commercial banks. Commercial Bank of Ethiopia (CBE),
Dashen Bank S.C (DB), Awash International Bank S.C (AIB), Wogagen Bank S.C (WB), United
Bank S.C (UB), Nib International Bank S.C (NIB), Bank of Abyssinia S.C (BOA), Lion
International Bank S.C (LIB), Cooperative Bank of Oromia S.C (CBO), Berehan International
Bank S.C (BIB), Buna International Bank S.C (BUIB), Oromia International Bank S.C (OIB),

22
Zemen Bank S.C (ZB), Abay Bank(AB),Addis International Bank(ADIB), Debub Global
Bank(DGB) and Enat Bank (EB).

Commercial Banks of Ethiopia categorized into three peer groups. It is based on the establishment
period and asset sizes of the banks. A large bank is the first category, there is only one banks that
is Commercial Bank of Ethiopia (CBE), The second peer group is middle banks, under this
category there is five medium banks which are Awash, Dashen, Abyssinia, Wegagen, United and
Nib Banks. The final peer group is small banks; this group is relatively small in asset size, which
is Cooperative bank of Oromia, Oromia International Bank, Lion, Zemen, Bunna, Berhan, Abay,
Addis, Enate and Debub Global Banks. The study were consider large and medium banks, to get
fair output of the industry, accordingly, the researcher select one large bank and from 16 total
private commercial banks 6 of them were selected from medium peer groups.

The study also used purposive sampling technique, because purposely select one government bank
and six private commercial banks of Ethiopia according to their prior experience in Ethiopia, this
is because, the banks have better experience in lending and having lot of customers than the rest
banks. In such manner the study was include, Commercial Bank of Ethiopia, Awash Bank, Dashen
Bank, Bank of Abyssinia, United Bank, Nib International Bank, and Wegagen Bank, The
researcher believed that the sample size is adequate to make sound conclusion. The selected banks
establishment period are shows in the below table:

Table 2 List of selected Banks

Name Of Banks Year of Establishment

Commercial Bank of Ethiopia 1963

Awash International Bank 1994

Dashen Bank 2003

Bank of Abyssinia 1996

United Bank 1998

Nib International Bank 1999

23
 Wegagen Bank 1997

From Wikipedia, the free encyclopedia

3.4 Methods of Data Analysis

After collecting the relevant data through the data gathering methods that will use in this study,
the researcher will categorize the data appropriately for interpretation to achieve the stated
objectives. In this study two type of statistical analysis were used to test the proposed hypotheses.
These are descriptive statistics and inferential statistics to see the cause and effect relationship
between the dependent and independent variables. The descriptive statistics of both dependent and
independent variables were calculated over the sampled periods. This helps to convert the raw data
in to a more meaning full form which enables the researcher to understand the ideas clearly. Then,
correlation analyses between dependent and independent variables were made and finally a
multiple linear regression analysis and diagnostic test was used to determine the relative
importance of each independent variable in influencing NPLs of Ethiopian commercial banks by
using E-views 9 software.

3.5 Description of Variables

The studies had both dependent and independent variables and explain character before, designing
model specification of variables.

3.5.1 Dependent Variables

Non-performing loans (NPLs) means loans & advances whose credit quality has deteriorated such
that full collection of principal and/or interest in accordance with the contractual repayment term
of the loan or advance is in question (NBE directive No SBB/43/2008). The rise of non-performing
loan portfolios in banks significantly contributed to financial distress in the banking sector. NPLs
can be determined both Macro and Internal determinant variables of the bank. Below the study
tried to explain both macro and internal bank factors or independent variables.

3.5.2 Independent Variables

Bank size (BS): Too big to fail hypothesis assumes that large banks take excessive risks by
increasing their leverage too much and extend loans to lower quality borrowers, and therefore have
more NPLs. Some researchers such as (Salas and Saurina, 2002) found a negative relation between
bank size and NPLs and argued that bigger size allows for more diversification opportunities. We

24
expect a positive effect of size on NPLs. In order to emphasize this possible non-linear relationship,
as a proxy the study use the logarithm of banks total assets.

H1: There is positive relationship between bank size and NPLs:

Exchange rate (EXR): No one can predict what the exchange rate will be in the next period, it
can move in either upward or downward direction regardless of what the estimates and predictions
were. An appreciation of exchange rate can have mixed effects. It may weaken the competitiveness
of export-oriented firms and adversely affect their ability to pay their debts (Fofack, 2005)
.However; it may improve the debt servicing capacity of borrowers whose loans are in foreign
currencies. So, the relationship between EXR and NPL may be mixed. An increase in the EXR is
expected to decrease nonperforming loan ratio.

H2: There is a negative relation between EXR and NPLs

Economic growth (GDP): There is a significant empirical evidence of negative association

between economic growth and non-performing loans (Farhan et al. 2012). Carey (1998) argues
that the state of the economy is the most important factor affecting diversified debt portfolio loss
rates. Salas and Saurina (2002) found a significant negative effect of GDP growth on NPLs.
Economic growth usually increases the income which ultimately enhances the loan payment
capacity of the borrower which in turn contributes to lower bad loan and vice versa (Khemraj and
Pasha, 2009). Accordingly we expect a negative effect of economic growth on NPLs.

H3: There is negative relationship between GDP and NPL

Inflation rate (INF): many researchers such as (Khemraj and Pasha, 2009) and (Fofack, 2005)
found a positive relationship between the inflation and NPLs. While Nkusu, (2011) argued that
inflation can affects the borrowers loan payment capacity positively or negatively, higher inflation
can enhance the loan payment capacity of borrower by reducing the real value of outstanding debt;
moreover increased inflation can also weaken the loan payment capacity of the borrowers by
reducing the real income when salaries are sticky. So according to literature relationship between
inflation and nonperforming loans can be positive or negative depending on the economy of
operations (Farhan et al. 2012).

H4: There is a positive/Negative relationship between INF and NPL

Loan Growth: The loan is typically the largest asset and the prevail source of revenue.

25
Since loans are illiquid assets, increase in the amount of loans means increase in illiquid assets in
the asset portfolio of a bank. As it was made by various empirical studies this study expected
positive relationship between banks loan growth and NPLs. (Keeton, 2003) showed a strong
relationship between credit growth and damaged assets.

H5: There is positive relationship between loan growth and NPLs.

Liquidity (LIQ): High ratio of liquidity may send a positive signal to the depositors that the bank
is liquid; hence, higher ratio is the depositors' confidence. However, a lower value of this ratio
may signal that a bank is not in a good situation. On the other hand, higher liquidity may also
imply the inefficient utilization of resources therefore may be associated with a high probability
of failure. A higher ratio of liquid assets to total assets implies a greater capacity to discharge
liabilities, and is therefore associated with a higher survival time. Thus the study hypothesized that

H6: There is significant positive / negative relationship between Liquidity and NPLs of commercial
banks

Lending rate (LR): Lending rates denote the weighted average interest rates on loans and
advances. Many empirical evidence such as (Nkusu, 2011), (Adebola te al., 2011) and
(Berge and Boye, 2007) found a positive correlation between lending rate and NPLs. An
increase in interest rate weakens loan payment capacity of the borrower therefore non-
performing loans and bad loans are positively correlated with the interest rates (Nkusu,
2011). (Farhan et al. 2012) argued that banks with aggressive lending policies charging
high interest rates from the borrowers incur greater non-performing loans. We expect a
positive effect of lending rates on NPLs.

H7: there is a Positive Relationship between LR and NPLs

Return on Asset (ROA): The ROA reflects the ability of a bank’s management to generate
profits from the bank’s assets. It shows the profits earned per birr of assets and indicates how
effectively the bank’s assets are managed to generate revenues. This is probably the most
important single ratio in comparing the efficiency and operating performance of banks as it
indicates the returns generated from the assets that bank owns.(Getahun, 2015). Thus the study
hypothesized that

26
H8: There is significant positive relationship between NPLs and ROA

3.6 Model Specification

In establishment of the relationship between study variable comprising of independent variables
including size of the banks, exchange rate, GDP, Inflation rate, Loan growth, Liquidity, Lending
rate and Return on Asset. The regression model was as follows based on this to analyze the cause
effect relation the study were developed the following model;

Yᵢ= β0 +β₁ᵢt +β₂ᵢt + β₃ᵢt + β₄it + β5it + β 6it+ β 7it +β 8it +ɛᵢ

Where: - Yit is the dependent variable for firm ‘i’ in year ‘t’, β0 is the constant term, β is the
coefficient of the independent variables of the study, X it is the independent variable for firm ‘i’
in year ‘t’ and εit the normal error term.

Thus, this study is based on the conceptual model adopted from Fawad and Taqadus (2013).
Accordingly, the estimated models used in this study are modified and presented as follow;

NPL= β0+β1(SIZE)it +β2(EXR)it+β3(GDP)it+β4(INFR)t+β5(LG)it+ β6(LIQ) it+ β7(LR)it+ β8(ROA)it+εit

Where;

 β0 is an intercept
 β1, β2, β3, β4, β5, β6, and β7represent estimated coefficient for specific bank i at time t ,
 BS, EXR, GDP, INFR, LG, GDP, LIQ, LR and ROA represent Size of banks, exchange
rate, gross domestic products, inflation rate, loan growth, liquidity, lending rate and
return on asset, respectively εit represents error terms for intentionally/unintentionally
omitted or added variables.

27
 CHAPTERFOUR
Data Analysis and Interpretation
4. Introduction
This core chapter deals with the discussion and analysis of data collected from the sampled banks
annual publications of the national bank of Ethiopia (NBE) and each commercial banks audited
annual financial reports. The audited financial statements of the banks over the study period has
been obtained from National Bank of Ethiopia, (which is responsible for maintaining the audited
financial statements of all banks operating in the country and regulate their operating activities),
the country’s central bank. Basically, the balance sheet and income statements were the main
sources of the relevant data to address the stated objectives of the study. Based on this the study
were analyzed in two major sections. The first section describes determinates of commercial banks
non – performing loan using percentage ratio and the second section presented the correlation and
regression analysis to determine cause effect relationship between dependent and independent
variables.

4.1. Descriptive statistics

In this part the study discussed both dependent and independent variables. The dependent variable
of the study is NPLs of Ethiopian commercial banks in Ethiopia. On the other hand the independent
variable of the study consider both Macro – economic determinates such as GDP, Inflation rate,
and Exchange rate, while, the internal determinate factors considered in the study were Bank size,
Loan growth, Liquidity, Lending rate and Return on asset.

4.1.1 Trend of Non – Performing Loan (NPLs)

In this study, NPL is measured by the share of non-performing loans from the total loans &
advances of the bank. The National Bank of Ethiopia has provided direction to all commercial
banks to maintain the NPL ratio below 5%. The below table implied trend of non –performing loan
at each bank level.

28
 Figure 2 Non-Performing Loan Trend Analysis of Commercial banks in Ethiopia
0.350
0.300
0.250
Percentage

0.200
0.150
0.100
0.050
0.000
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Avg
CBE 0.246 0.290 0.244 0.212 0.177 0.142 0.061 0.031 0.018 0.025 0.022 0.026 0.028 0.024 0.027 0.105
AB 0.038 0.055 0.077 0.062 0.049 0.043 0.046 0.055 0.047 0.036 0.027 0.023 0.023 0.017 0.015 0.032
DB 0.031 0.039 0.037 0.032 0.027 0.025 0.023 0.023 0.022 0.020 0.021 0.022 0.019 0.017 0.017 0.02
BOA 0.057 0.077 0.076 0.049 0.031 0.047 0.089 0.098 0.074 0.033 0.026 0.020 0.018 0.015 0.014 0.043
WB 0.049 0.051 0.058 0.051 0.048 0.044 0.059 0.061 0.040 0.045 0.024 0.022 0.017 0.016 0.016 0.033
UB 0.012 0.024 0.039 0.039 0.029 0.030 0.027 0.031 0.036 0.028 0.023 0.019 0.014 0.012 0.013 0.023
NB 0.012 0.040 0.038 0.041 0.039 0.034 0.038 0.046 0.039 0.041 0.027 0.025 0.021 0.015 0.018 0.047
Avg 0.063 0.082 0.081 0.069 0.057 0.052 0.049 0.049 0.039 0.032 0.024 0.022 0.020 0.016 0.017 0.032

Source Each Banks Annual report

The trend analysis implied on the above figure that, NPLs performance of the studied banks were
decrease from year to years, as implied by the average non performing rate was high in 2003 and
2004, 8.22% and 8.12% respectively in 2015 and 2016 the average performance was decreased
relatively with the previous years. This implied that performance of commercial banks increase in
collecting the loan from borrowers. However, trend of NPLs performance of each banks were not
the same some of the banks perform well while the others not well. Accordingly, at individual
bank level the highest percentage ratio of NPLs observed 4.7% in NB, while the lowest were 2.0%
in DB. Generally, from the stated data one can deduced that, significant decline of NPLs might
imply either improvement in the levels of loan quality or being escaping of banks from providing
loan and advances. Even though, the trend implied on the studied years improvement, however,

29
the average NPLs of the studied banks for the past fifteen years at 3.2% implied still there is a
challenge on commercial banks to administrator loan effectively.

Bank Size (SIZE)

Bank size is what the bank possesses and it is useful to measure the banks general capability to
undertake its intermediary function. In this study, the proxy used to measure bank size was the
natural logarithm of the total asset. Larger banks have the advantage of better access to additional
financing, dealing with liquidity problems and diversifying risk. This is probably due to the fact
that larger banks benefit from a “too large to fail” policy and are believed to be more likely to
survive than smaller banks. The study implied below in the fig. both the average trend of the
industries and each of the studied banks size growing trends and their effect on commercial banks
NPLs.

. Figure 3 Average natural logarithm of total asset Trend of the Industry

Industry Asset growuth

5
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
AVG trend 3.09 3.2 3.32 3.44 3.55 3.8 3.8 3.91 4 4.1 4.2 4.16 4.35 4.37 4.46

Source Each banks annual Report

The average total assets of Ethiopian commercial banks have shown consistent growth throughout
the studied period. Accordingly the maximum total asset was registered in 2016 (4.46%) while the
minimum were registered in 2002 which is 3.09%. Hence, the larger bank size induces economy
of scale there by making larger banks more profitable and will reduce the cost of gathering and
processing information. Since larger banks are more able to solve problems of information

30
asymmetry in comparison to their smaller counterparts. Skilled employees and quality information
bases, larger banks are more effective in credit analysis and monitoring their debtors. Therefore
larger banks have the positive impact of the banks and the country’s economy. Regarding
individual banks asset level the growth rate were not consistent some of the studied banks average
total asset were grow at a fastest rate while the others were not similarly grown below the graph
implied each of the studied banks trend of total asset growth between the studied years.

Figure 4 Average natural logarithm of total asset Trend each banks

Each Banks avg asset growth 2002 -2016

6
5
4
3
2
1
0
CBE AB DB BOA WB UB NB AVG
AVG Asset 4.84 3.78 3.92 3.78 3.52 3.52 3.59 3.87

Source Each banks annual Report

The average ratio of total asset growth of each bank for the last fifteen years has similarities except
slight asset growth rate score in each bank. Accordingly, maximum average total asset growth rate
scored by CBE (4.84%), DB (3.92%), and AB (3.78%), while the minimum average asset growth
was observed at UB (3.52%) and NB (3.52%). Totally the studied commercial bank’s total asset
grows for the past 15 consecutive years 3.87% average growth rate. Therefore, banks asset growth
can enhance the NPLs performance of each bank, this is because when asset increase profitability
also increase and capacity of the banks increase interims of increasing number of employee,
branches, in turn, it increases profitability and the chance of a non-performing loan.

Liquidity Position of studied Banks

The Liquidity position of the studied banks was measured based one Liquid asset/net deposit ratio
which indicates the extent to which the bank’s total liquid assets are composed of deposits from
customers and other financial institutions. The measure implied that liquid assets are cash on hand,

31
deposits with local and foreign banks and treasury bills and other items compared with liquid
assets. On the other hand, the net deposit is composed of demand deposits, savings deposits and
time deposits which are liabilities for the bank. One of the liquidity measures of this study is liquid
asset-to-deposit and other short-term borrowings ratio. The National Bank of Ethiopia also uses
this ratio as the measurement of banks liquidity level and the liquidity requirement directive is
based on this ratio. As per NBE directive number SBB/57/2014 issued by the National Bank of
Ethiopia, any licensed commercial banks are required to maintain liquid asset not less than 15%
of its net current liabilities (which includes the sum of demand deposits, saving deposits, time
deposits and similar liabilities with less than one-month maturity).Commercial banks may confront
with liquidity deficit when they face a problem of meeting a large amount of demand
(withdrawals). A high ratio of liquidity may send a positive signal to the depositors that the bank
is liquid; hence, higher is the depositors' confidence. However, a lower value of this ratio may
signal that a bank is not in a good situation. On the other hand, higher liquidity may also imply the
inefficient utilization of resources, therefore, may be associated with a high probability of failure.
A higher ratio of liquid assets to total assets implies a greater capacity to discharge liabilities and
is, therefore, associated with a higher survival time. Liquidity is a prime concern for banks and the
shortage of liquidity can trigger bank failure. Banking regulators also view liquidity as a major
concern. This is because banks without sufficient liquidity to meet demands of their depositors
risk experiencing bank run. Holding assets in a highly liquid form tends to reduce income as liquid
assets are associated with lower rates of return. Below the study implied both trends of the studied
banks average liquid position as well as each banks position of liquidity for the past 15 years

Figure 5 Average Liquidity growth rate and trend of the studied banks from 2002 – 2016

Avg. Liquidity Growth Trend

50
40
30
20
10
0
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 AVG
Liq 39.05 40.05 41.88 38.37 33.04 35.09 38.22 47.49 45.92 42.49 29.7 23.46 22.09 22.78 20.11 34.65

Source Each banks annual Report

32
The above fig indicated the average liquidity position of the industry were grow from the year
2009 – 2011, however, starting 2012 – 2016 it implied constant declining; it may relate in addition
to increasing loan facilities for several sectors the application of 27% NBE bill Purchase
regulatory in 2011 may also have its own impact. Generally, average minimum liquidity position
of the industry observed in 2016 (20.11%) while the maximum was 47.49% in 2009. The study
also assessed each of the studied bank’s liquidity position and their growth trend. Accordingly,
below the figure implied each of the studied bank’s liquidity positions between the studied years.

Figure 6 Average Liquidity growth rate of each banks 2002 – 2016

Each Banks avg growth rate of liquidity 2002 -2016

38
36
34
32
30
CBE AB DB BOA WB UB NB AVG
LIQ 35.13 33.93 32.84 34.6 35.33 37.16 33.55 34.65

Source Each banks annual Report

As implied from the above figure liquidity position of each bank were not the same accordingly,
the average growth rate CBE at 35.13% was the highest of government commercial banks while
Dashen bank at 32.84% of average liquidity growth rate recorded the lest among the studied private
banks for the last 15 years. Therefore from the result the study deduced that, even though trend of
liquidity position becoming to decrease, however, the position of each bank and the total average
of the studied banks have engaged bay far NBE requirement of 15%, this implies that the
inefficient utilization of resources and loan service by the banks is decline and also not disburse
additional loans to the prominent clients to strength the borrower’s capacity. The banks and this
can affect the performance of loan granting of each bank.

Loan Growth

Hence, the loan is the principal business activity for all commercial banks in Ethiopia and the loan
portfolio is the largest asset and the predominant source of revenue. The higher the loan growth

33
has probability the higher profit. The graph implied the trend of the average loan growth rate on
the studied banks.

Figure 7 Average Loan growth trend of the studied banks

Loan Growth Rate Trend

45
40
35
Loan Gro2wth Rate %

30
25
20
15
10
5
0
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
Loan Growth Rate 19.24 38.93 24.54 36.28 41.11 28.33 23.24 4.45 16.2 25.44 37.32 24.77 13.03 22.66 20.91

Source Each banks annual Report

The trend shows an increase in loans granted to borrowers throughout the studied years except in
the year 2009 decline to 4.45% and which is the minimum average growth rate of the loan, while
the maximum loan growth rate was observed in 2006 (41.11%) and 2003 (38.93%). Regarding
each of the studied bank’s performance the fig. below implied its trend

Figure 8Average LG ratio of each banks performance from 2002 - 2016

35 33.19

30 28.08
25.39 25.39
23.96
25 21.82
20 17.86

15
10
5
0
CBE AB DB BOA WB UB NB

Source Each banks annual Report

34
Regarding individual banks, average growth rate of loan are UB (33.19%) and NB (28.08%) while
the lowest was CBE (17.86%). Generally, from the trend of LG, one can understand that, as the
main income of commercial banks depends from an interest of loan and advance the loan growth
of the studied banks implied that, all of the studied banks are profitable. This implies that
commercial Banks of Ethiopia have been utilized the liquid assets effectively and efficiently.

The Effect of ROA on NPLs

Profitability is the likelihood of a business earning the desired level of income within a specific
period of time under certain prevailing business conditions. ROA measured by the ratio of net
profit before tax to total asset.Net profit before tax was used in order to avoid the impact of
different period’s tax rate on the net profit of the bank. The Figure indicated both average growth
rate of the studied banks and each banks position. Below the table indicated profitability trend of
the studied commercial banks

Figure 9 Trend of profitability of the industry

Averege ROA from 2002 - 2016

4

0
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
ROA 0.88 1.44 2.23 2.8 3.17 3.24 2.95 2.98 3.27 3.56 3.69 3.17 3.12 2.54 2.58 2.775

Source Each banks annual Report

As shown in above fig. the average growth rate of Return on Asset (ROA) of the studied banks
constantly increased from the year 2010 to 2012, however, starting to 2015 and 2016 highly
decreased. Accordingly, the minimum return on asset of 0.88% was registered in the year 2002
and the maximum return on asset of 3.69% was registered on the year 2012. On the other hand,
the ROA of each bank also not similar, accordingly below the fig implied each of the studied
commercial banks potential of ROA during the last 9 years.

35
 Figure 10 Trend of each banks performance of Profitability

Each Banks Avg. growth rate ROA 2002 -2016

3.5
3
2.5
2
1.5
1
0.5
0
CBE AB DB BOA WB UB NB
ROA 2.34 2.87 2.94 2.23 3.26 2.58 3.20

Source Each banks annual Report

Regarding the studied banks level the average growth rate of WB 3.26%, NB (3.20%), DB (2.94%)
and AB (2.87%) were the highest while, UB (2.58%), CBE (2.34%), and BOA (2.23%) was the
lowest of all banks considered in this study. Though the net profit of older banks was higher in
magnitude than newly opened banks, equivalently the total asset of the older banks was higher and
as a result, the ratio of ROA has not shown a significant difference between the studied banks.
Generally, as the rate of profitability declining throughout the years it affects NPLs performance
of commercial banks negatively.

Inflation rate of the country (INF)

It is a situation in which the economies overall price level is rising. It represents sustained and
pervasive increment in aggregate price of goods and services resulting decline in purchasing power
of money. Accordingly, when inflation is high and unexpected, it can be very costly to an economy.
At the same time, inflation generally transfers resources from lender and savers to borrowers since
borrowers can repay their loans with birr that are worthless. It is determined as the general
consumer price index. This indicates that, as inflation increase, the cost of borrowing gets more
expensive and deteriorates the quality of loan portfolio. Below the fig. implied the effect of
inflation over the studied banks

36
 Figure 11 Inflation Trend of the country and its effect on NPLs

INFLATION
36.40
34.10

25.30
18.10
15.82
13.50
10.92 10.58 9.69
7.35 6.13 8.10 7.70
2.80
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
-10.57

AB DB BOA WB UB NB

Source NBE Report

The maximum inflation rate was recorded in the year 2009 (36.40%) and in 2012 (i.e. 34.10%)
followed by the year 2008 (25.300%) and the minimum inflation rate which was recorded in 2002
(-10.57%) and 2010 (2.8%). High inflation may pass through to nominal interest rates, reducing
borrowers’ capacity to repay their debt. Through its attraction with the tax system, it can increase
tax burden by artificially increasing income and profits. Besides, inflation cause firms to increase
their costs of changing prices. Finally, it made individuals hold less cash and make more trips to
banks since inflation lowers the real value of money holdings.

Gross Domestic Product (GDP)

Gross Domestic Product (GDP) is an indicator of the economic health of a country as well as the
gauge of a country's standard of living. It is the measurement of the level of economic activity of
a country. For the purpose of this study, GDP is measured by the annual real growth rate of the
gross domestic product. Below the figure indicates the country GDP for the following 15
consecutive years and its effect on NPLs of Commercial Banks

37
 Figure 12 Trend of GDP and Its effect on NPLs

15.00
GDP
GDP Growth Rate

10.00

5.00

0.00

-5.00
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
GDP 1.63 -2.10 11.73 12.64 11.54 11.79 11.19 10.04 10.57 11.40 8.70 9.90 10.35 10.40 8.00

Source NBE Report

As indicated on the above figure the minimum GDP growth rate was scored in 2003(-2.10%), 2002
(1.63%), 2016 (8%), 2012(8.70%) and 2013 (9.90%) otherwise the country GDP was Growth in
Double-digit. According to the study results the GDP of the country specifically, in the year 2016
decline this is because of the political instability of the country. Scholar analyzed the relation
between GDP and NPLs in several ways accordingly, GDP growth and employment are negatively
associated with the NPL. Conversely, unemployment is positively related to the NPL. Several
empirical studies have found a negative association between NPL and real GDP low growth rate
(Salas and Saurina 2002; Fofack, 2005; Jimenez and Saurina, 2006; Khemraj and Pasha, 2009;
Dash and Kabra, 2010). The justification provided in the empirical literature of this association is
that higher positive level of real GDP growth habitually entails a higher level of income. This
improves the capacity of the borrower to pay its debts and contributes to reducing bad debts. When
there is a downturn in the economy (slowed or negative growth of GDP) the level of bad debts will
increase.

Lending Rate
Lending Rate/interest rate: Lending rates are one of the primary economic determinants of non-
performing loans. As far as interest rate policy is concerned it plays a very important role in NPLs
growth rate in a country/economy, Hoque and Hossain (2008) examined this issue and according
to them, non-performing loans are highly correlated with the lending rate. Below the fig. Implied
lending rate of the studied banks

38
 Figure 13 Average lending rate trend and its effect on NPLs of the studied banks

Lending Rate
14
Lending arte of the country

12
10
8
6
4
2
0
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
Lending Rate 10.75 10.75 10.75 10.5 10.5 10.5 11.50 12.25 12.25 11.88 11.88 11.88 11.88 11.88 12.75

Source Each banks annual Report

The last independent macro variable is the lending rate. The mean value of lending rate was
10.50% observed in 2005 to 2007 and the highest was 12, 75% in 2016. The lending rate of the
country over the past 15 years was the stable that implies the bank’s profit increased and has a
positive impact on the country’s economic growth.

Exchange Rate
No one can predict what the exchange rate will be in the next period, it can move in either upward
or downward direction regardless of what the estimates and predictions were. An appreciation of
exchange rate can have mixed effects. It may weaken the competitiveness of export-oriented firms
and adversely affect their ability to pay their debts (Fofack, 2005). However; it may improve the
debt servicing capacity of borrowers whose loans are in foreign currencies. So, the relationship
between EXR and NPL may be mixed. An increase in the ER is expected to decrease
nonperforming loan ratio. The results of the study trend analysis also implied this, accordingly
the fig. below portray the increment trend of foreign exchanges.

39
 Figure 14 Trend of exchange rate and its effect on NPL

Exchange Rate
25.00
Excenge Rate in Percentage

20.00
15.00
10.00
5.00
0.00
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
Exchange Rate 8.57 8.60 8.63 8.66 8.69 9.03 9.61 11.30 13.53 16.91 17.73 18.64 19.58 20.57 21.80

Source Each banks annual Report

As implied by the above the exchange rate of the banks consistently increases from the year 2011
to 2016. Accordingly, the minimum exchange rate observed in 2002 (8.57%) while the maxim
was 2016 (21.80%). This implies that the foreign exchange rate in Ethiopia during the study period
remains high. Since the country’s currency highly devaluated and during the period the bank’s
client especially importers are highly disputed and failed to repay the required bank loan
repayments.

4.1.2 Correlation Analysis

In this section, the correlation between the dependent variables and the independent variables have
been presented and analyzed. According to Brooks (2008), the correlation between two variables
measures the degree of linear association between them. To find the relationship between variables
I had used the most broadly applied correlation statistics of Pearson correlation which was once
used in this study. Values of Pearson’s correlation coefficient are always between -1 and +1. A
correlation coefficient of +1 indicates that two variables are perfectly related in a positive sense; a
correlation coefficient of -1 indicates that two variables are perfectly related in a negative sense,
and a correlation coefficient of 0 indicates that there is no linear relationship between the two
variables. A low correlation coefficient; 0.1 - 0.29 suggests that the relationship between two items
is weak or non-existent. If r is between 0.3 and 0.49 the relationship is moderate. A high correlation
coefficient i.e. >0.5 indicates a strong relationship between variables. The direction of the
dependent variable's change depends on the sign of the coefficient. If the coefficient is a positive
number, then the dependent variable will move in the same direction as the independent variable;

40
if the coefficient is negative, then the dependent variable will move in the opposite direction of the
independent variable. Hence in this study both the direction and the level of relationship between
the dependent and independent variables conducted using the Pearson’s correlation coefficient.
The table below presents the result of the correlation analysis.

Table 3 Correlation Analysis

NPL BS EXR GDP INFR LG LIQ LR ROA
NPL 1
BS 0.11576 1
EXR -0.63541 0.65756 1
GDP -0.06288 0.31946 0.15322 1
INFR -0.00378 0.25559 0.10690 0.25612 1
LG -0.18999 -0.31178 -0.15003 -0.0807 -0.0309 1
LIQ -0.53948 -0.38816 -0.57222 -0.0464 0.12463 -0.1846 1
LR -0.47760 -0.60465 -0.59953 0.13159 0.32387 -0.3174 -0.2441 1
ROA -0.19542 0.1642 0.26194 -0.53346 0.41081 0.17567 -0.0256 0.26235 1

Source: Author Own computation (2018)

As implied in the above correlation matrix between independent variables, there were fairly low
data correlations among the independent variables. These low correlation coefficients indicate that
there is no problem of multicollinearity in the study. Moreover, Kennedy (2008) stated that
multicollinearity problem exists when the correlation coefficient among the variables is greater
than 0.70, but in this study, there is no correlation coefficient that exceeds or even close to 0.70
except some variables such as EXR and BS. Accordingly, in this study there is no problem of
multicollinearity which enhanced the reliability for regression analysis.

4.1.3 Regression Model Assumption and Diagnostic Test

After different tests were run to make the data prepared for analysis and to get reliable output from
the research classical linear regression model is better to satisfy basic assumption. As noted by
Brooks (2008), once these assumptions are satisfied, it is considered as all available information
is employed within the model. However, if these assumptions are violated, there will be a data that
left out of the model and also the researcher choose fixed effects model. Accordingly, before
applying the model for testing the significance of the slopes and analyzing the regressed result,
heteroscedasticity, autocorrelation, multicollinearity and normality tests are made for identifying

41
misspecification of information if any thus on fulfill analysis quality. After the diagnostic test
finalized

4.1.3.1 Results of Regression Analysis and Diagnostics test

Different tests were run to form the data ready for analysis and to get reliable output from the
study. These tests were expecting to check whether the OLS assumptions, are fulfilled when the
explanatory variables are regressed against the dependent variables.

Heteroscedasticity Test
In the classical linear regression model, one of the assumptions is Homoscedasticity. When the
scatter of the errors is different, varying depending on the value of one or more of the independent
variables, the error terms are heteroskedastic (Gujarati & Porter, 2009).

Heteroscedasticity white test is used to test the heteroscedasticity problem in this research. This
test is very important because if the model consists of heteroscedasticity problem, the OLS
estimator no longer BEST and error variances are incorrect, therefore the hypothesis testing,
standard error and confident level will be invalid. If the p-value is less than significant level we
reject the null hypotheses otherwise, do not reject the null.

Table 4 Heteroscedasticity Test

Heteroscedasticity Test: White

F-statistic 3.17472 Prob. F(44,60) 0.15399

Obs*R-squared 73.4508 Prob. Chi-Square(44) 0.26011
Scaled explained SS 77.2539 Prob. Chi-Square(44) 0.48432

The p-value of this model is 0.48432 which is more than the significant level 0.05 (5%), so the
model doesn’t have heteroscedasticity problem.

Autocorrelation Test
The most commonly used test is "Durbin-Watson test for autocorrelation'' is based on the
assumption that the errors in the regression model are generated by a first-order autoregressive
process observed at equally spaced time periods. The Durbin-Watson statistic ranges in value from
0 to 4. A value near 2 indicates non-autocorrelation; a value toward 0 indicates positive
autocorrelation; a value toward 4 indicates negative autocorrelation. The result of this study was
1.662771, so the value indicates non-autocorrelation.

42
Multicollinearity
According to the Gujarati (2009), multicollinearity will occurs if some or all of the independent
variables are highly correlated with one another. It will causes the regression model has difficulty
telling which independent variables are affecting the dependent variable. If an independent
variable is an exact linear combination of the other independent variables, then we say the model
suffers from perfect collinearity, and it cannot be estimated by OLS Brooks (2008).

If multicollinearity problem too serious in model, this study have to take action to add in other
important independent variable or drop unimportant independent variables.

Table 5 Multicollinearity Test

BS EXR GDP INFR LG LIQ LR ROA

BS 1

EXR 0.657557 1

GDP 0.319456 0.153219 1

INFR 0.255589 0.106898 0.256125 1

LG -0.31178 -0.15003 -0.08069 -0.03087 1

LIQ -0.38816 -0.57222 -0.04642 0.124634 -0.1846 1

- -
LR -0.60465 0.599533 0.131594 0.323874 -0.31745 0.24406 1
- -
ROA 0.164197 0.261941 0.533455 0.410815 0.175671 0.02557 0.26235 1

The result of correlation matrix indicates that there were low data correlations among the
independent variables. Kennedy (2008) stated that multicollinearity problem exists when the
correlation coefficient among the variables are greater than 0.70, but in this study there is no
correlation coefficient that exceeds or even near to 0.70.

Normality
Normality test is used to determine whether the error term is normally distributed or not. Jarque
Bera test is to ensure that the data set is well-modeled by a normal distribution. The hypothesis for
the Normality Test is stated as follow:

43
H0: The error term is normally distributed
H1: The error term is not normally distributed
If P-value of JB is less than significant level of 5% we reject the H0. Otherwise, do not reject H0.

Table 6 Normality Test

Probability (P-Value)

Jarque Bera Test 1.397736

Source: Developed by researcher

P-Value = 1.3977 means do not reject H0 the error term is normally distributed.

Result of Regression Analysis

The section covers the empirical regression model used in this study and the results of the
regression analysis. Empirical model: As presented in the methodological part of the study, the
empirical model used in the study in order to identify the factors that can affect Ethiopian
commercial banks NPLs provided as follows:

Yᵢ= (β₀ +β₁ BSᵢt +β₂ EXRᵢt + β₃ GDPᵢt + β₄ INFRit + β5LGit + β 6LIQit+ β 7LRit +β 8ROAit) +ɛᵢ

According to Chris brooks (2008), the p-value of Hausman test is less than 1% fixed effect
model is appropriate. Due to this, the researcher use fixed effect model. The Fixed Effect model
assumes that the marginal effects of the explanatory variables on the dependent unit are the same
for all units. Constant term is allowed to vary among the banks to account for the differences
between units. These constant terms capture all unobserved characteristics that differentiate the
units from each other.

44
 Table 1 Regression analysis result between variables

Variable Coefficient Std. Error t-Statistic Prob.

C -0.014029 0.013328 -1.052598 0.2952

BS 0.002377 0.00037 6.430936 0.0000
EXR -0.000826 0.000187 -4.422019 0.0000
GDP 0.025293 0.012902 1.96039 0.0528
INFR -0.001427 0.004268 -0.334385 0.0388
LG -0.006013 0.002593 -2.3189 0.0225
LIQ 0.000538 0.000152 3.527652 0.0006
LR -0.275964 0.109567 -2.51867 0.0134
ROA 0.051112 0.053626 0.95311 0.0342

Effects Specification

Cross-section fixed (dummy variables)

R-squared 0.793398 Mean dependent var 0.012294

Adjusted R-squared 0.769534 S.D. dependent var 0.006628
S.E. of regression 0.004002 Akaike info criterion -3.122067
Sum squared resid 0.001538 Schwarz criterion -2.894585
Log likelihood 135.4085 Hannan-Quinn criter. -3.029887
F-statistic 21.6504 Durbin-Watson stat 1.662771
Prob(F-statistic) 0.00000

R-squared

The R-squared (R2) statistic measures the success of the regression in predicting the values of the
dependent variable within the sample. In standard settings, R2 may be interpreted as the fraction
of the variance of the dependent variable explained by the independent variables. The statistic will
equal one if the regression fits perfectly, and zero if it fits no better than the simple mean of the
dependent variable. It can be negative for a number of reasons. For example, if the regression does

45
not have an interceptor constant, if the regression contains coefficient restrictions, or if the
estimation method is two-stage least squares.

In this study the R-squared statistics of the model was 0.793398 . This indicates that the
changes in the independent variables collectively explain 79.3398% of the changes in the
dependent variable and the remaining 20.6602% of changes is explained by other factors which
are not included in the model. Thus these variables collectively, are good explanatory variables.
One of the problem using R2 is every time when add an independent variable to the model the R2
never decreases.

Adjusted R-Squared

Adjusted R2 is a corrected goodness-of-fit (model accuracy) measure for linear models. It

identifies the percentage of variance in the target field that is explained by the inputs.

Adjusted R2 is always less than or equal to R2. A value of 1 indicates a model that perfectly predicts
values in the target field. A value that is less than or equal to 0 indicates a model that has no
predictive value. In the real world, adjusted R2 lies between these values. In our model the adjusted
R2 result is 0.769534 it is less than the R2 result and the value indicates the model was perfectly
predicts values in the target field

Probability (F-statistic)

The probability of (F-statistic) test is 0.000 indicates strong statistical significance, which
enhanced the reliability and validity of the model means all selected explanatory variables can
affect the level of NPLs in common.

Following the result obtained from the regression analysis as depicted in the above table, the next
section tries to present the analysis concurrently with respect to each NPLs determent factors.

Bank Size

With regards to Size of the bank and its relation with NPLs, Size of the banks has a positive
relationship with NPLs at 1% significance level with a P-value of 0.0000.

This finding suggests that diversification and effective monitoring measures increase bank size to
cover unexpected events and reduce the chances of insolvency (see also Baradwaj et al. 2014;
Marijana Ćurak et al. 2013).

46
Exchange Rate

Negative and significant relationship between the EXR rate and NPLs. Ethiopia had high level of
foreign currency loans, and it is expected that the NPLs ratio reacts strongly to exchange rate
volatility. This result suggests that a depreciation of the domestic currency would lead to an
increase in the NPLs rate, to the decline of credit worthiness of private debtors and the fact that
export-oriented companies do not use the positive effects of depreciation of the national currency
on export, due to low competitiveness of their products. This result is in contrast with the analysis
conducted by Beck et al. (2015);

Gross Domestic Products

GDP coefficient implied positively relate with NPLs at a significance level of 10% with a
p-value (0.0528); which implies when GDP goes up by 1 NPLs also goes up by 0.025293.
Economic growth usually improves income which ultimately enhances the loan payment
capacity of the borrower which in turn contributes to lower bad loan and vice versa.

GDP growth positively affects loan demand and supply of deposits hence the positive impact on
bank profitability. The positive relationship is supported by Pervan et al., (2015), Sufian and
Habibullah, (2009) and Kosmidou, (2008).The results did not confirm the findings by Tan and
Floros, 2012) that with economic growth, business environment is improved and barriers to entry
are lowered leading to high competition which reduces profitability.

Inflation

The indirect and significant relation between NPLs and inflation suggests that recessionary period
deteriorate underwriting standards and reduces the ability of borrowers to repay loans because of
the higher prices of goods. The Table shows that the coefficient of inflation is negative and
significant at 5% significant level.

Loan Growth

The study result with related to explanatory variables of Loan Growth (LA) has a negative
relationship with NPLs at 1% significance level with a P-value 0.0225. The result implied that
when loan growth is goes up by one unit non- performing loan is goes down by 0.006013.

47
Liquidity

The explanatory variable liquidity (LIQ), have a positive relationship with NPLs and at 1%
statically significance level with (P-value = 0.0006). The result is consistent with theory Richard
(2011), liquidity ratio has a negative influence on bank NPLs such as inefficient utilization of
resources and loan services by the banks is decline and also not disburse additional loans to the
prominent clients to strength the borrower’s capacity.

Lending Rate

The study variable Lending rate (LR) had a negative associate with NPLs at 5% significance level
with a p-value 0.0134.Hence, When lending rate is going up by one unit non-performing loan is
goes down by 0.275964 provided other independent variables are constant.

Return on Asset

There is a positive and significant relationship between ROA and NPLs. This result points out that
When ROA of the ratio increases NPLs also increase and they are positively correlated at 5%
significance level with a P-value of 0.0342.This implies when profitability goes up by one unit
NPLs also improved by 0.005112.

48
 CHAPTER FIVE
Summary, Conclusion and Recommendation of the study
5. Introduction
This study aims to identify factors affecting NPLs in some selected commercial banks of Ethiopia.
In doing so, previous studies on bank NPLs have been reviewed and NPLs determent factors are
identified. Therefore, this study specified an empirical framework to investigate the determinants
of Ethiopian commercial banks NPLs from 2002 to 2016. The NPLs determinant factors that were
used in this study include variables such as, bank size, exchange rate, GDP, inflation, loan growth,
liquidity, lending rate and return on asset based on this, the major findings of the study
summarized, concluded and recommended as follows:

5.1 Summary
The main objectives of the study were to determine factors affecting non-performing loans
in Ethiopian commercial banks during the year 2002 – 2016. Non-performing loans used
as a dependent variable and bank size, loan growth, liquidity, lending rate, ROA, GDP,
exchange rate and inflation rate were used as independent variables. The study used both
descriptive and inferential statistics technique.

The descriptive statistics result showed that the NPLs of Ethiopian commercial banks and
the independent variables of BS, LG, LIQ, LR, ROA, GDP, EXR AND INFR trend
analysis major findings of the study summarized as follow:

Regarding NPLs the trend implied that performance of the studied banks was improved from year
to years as implied by the average non-performing rate, it was improved from 0.063% and 0.082
in 2002 and 2003, to 0.016% and 0.017% in 2015 and 2016 respectively. This implied performance
of commercial banks increases in collecting the loan from borrowers. However, the trend of NPLs
performance of each bank was not the same some of the banks perform well while the others not
well.

The average total assets of Ethiopian commercial banks have shown consistent growth throughout
the studied period. Accordingly, the maximum total asset was registered in 2016 (4.46%) while
the minimum was registered in 2002 which is 3.09%. Hence, the larger bank size induces economy

49
of scale thereby making larger banks more profitable and will reduce the cost of gathering and
processing information.

The average liquidity position of the industry was to grow from the year 2002 – 2011, however,
starting 2012 – 2016 it implied constant declining; it may relate in addition to increasing loan
facilities for several sectors the application of 27% NBE bill Purchase regulatory. Generally,
average minimum liquidity position of the industry observed in 2016 (20.11%) while the
maximum was 47.49% in 2009.

Regarding loan growth, the trend shows fluctuate the rate throughout the studied years in year
2009 decline to 4.45% and which is the minimum average growth rate of the loan, while the
maximum loan growth rate was observed in 2003 (38.93%) and 2012 (37.32%).

In relation to ROA, the average growth rate implied consistent increment from the year 2002 to
2012, however, starting in 2013 – 2016 slightly decreased. Accordingly, the minimum return on
asset of 0.88% was registered in the year 2002 and the maximum return on asset of 3.69% was
registered on the year 2012.

In relation to INF, the maximum inflation rate was recorded in the year 2009 (36.400%) and in
2012 (34.10%) and the minimum inflation rate which was recorded in 2002 (-10.57%). Therefore;
a positive and significant relationship between INFR and NPLs. This discovering points to the
conclusion that the effect of higher interest rates due to inflation and declining economic
conditions which are commonly associated with rising inflation, succeed over the tremendous
impact that inflation would possibly have on borrowers debt servicing capacity

Regarding GDP, the study found that the minimum GDP growth rate was scored in 2003 (-2.10%),
2002(1.63%), 2012 (8.70%), 2013 (9.90%) and 2016 (8.80%) otherwise the country GDP was
Growth in Double- digit. According to the study results the GDP of the country specifically, in the
year 2016 decline this is because of the political instability of the country. GDP growth has a
positive and significant influence on bank performance. This suggests an improvement in the
widespread profits in the economic system is profit-enhancing. GDP growth positively affect the
loan demand and supply of deposits hence the positive impact on bank profitability. The positive
relationship is supported by Pervan et al., (2015), Sufian and Habibullah, (2009) and Kosmidou,
(2008).

50
The minimum value of lending rate was 10.5% observed in 2005 to 2007 and the highest were 12,
75% in 2016. The lending rate of the country over the past 15 years was stable lending rate that
implies the banks profit increased and has a positive impact for the country’s economic growth.

In relation to exchange rate of the banks consistently increases from the year 2002 to 2016.
Accordingly the minimum exchange rate observed in 2002 (8.57%) while the maxim was 2016
(21.80%). This implies that the foreign exchange rate in Ethiopia during the study period remains
high.

5.2 Conclusion
The main objective of the study was to identify the main banks internal factors and macro-
economic factors that can affect NPLs of Ethiopian banks

Regarding the trend analysis of the eight commercial banks of Ethiopia had downward
sloping of NPLs for the period 2002-2016.

The descriptive statistics indicate the levels of NPLs of commercial banks in Ethiopia are
above the threshold i.e. more than 5%, this means when NPLs increase above the threshold,
they start to cause negative effect on lending.

The regression analysis result showed that the determinant variables BS, GDP, LIQ and
ROA are positive and significant relationship with NPLs, and also EXR, INFR, LG and
LR are a negative and significant relationship with NPLs.

5.3 Recommendation
The study implied that an increase in nonperforming loans increases credit risk. Based on the
conclusion this study recommends that management of each bank improve their inspection
techniques and loan application methodologies in screening potential borrowers because the
existing credit risk trend may bring a series collapse against the sector as well as the national
economy in general.

It is necessary to oversight of the risk in individual loans. The Prudent risk in selections is vital to
maintaining favorable loan quality. To manage loans properly, bankers must understand not only
the risk posed by each credit but also how the risks of individual loans and portfolios are
interrelated.

51
Understanding the credit culture and the risk profile of the banks is central to successful loan
management. Because of the significance of a bank’s lending activities, the influence of the credit
culture frequently extends to other banks activities. Staff members throughout the bank should
understand the bank’s credit culture and risk profile. The knowledge should pass from chief credit
policy officer to account officers to administrative support. Directors and senior management
officers should not only publicly endorse the credit standards that are a credit culture’s backbone,
but should also employ them when formulating strategic plans overseeing portfolio management.

To ensure effective monitoring, it is recommended that management should ensure that credit
offices of the branch should be adequately resourced in terms of staff, vehicles and other logistics,
to support monitoring activities and follow up the borrowed fund are being used the intended
purpose and timely monitor the loan is being disbursed. It enables the lender assesses borrowers’
current financial conditions, ensure the adequacy of collaterals, ensure that loans are in compliance
with the terms and conditions of the facility, and identify potential problem loans for action to be
taken.

52
References
Abafita,J. (2003).Micro Finance and Loan Repayment Performance: A case study of the Oromia

credit and savings share Company in Kuyu, MBA thesis, University of South Africa.

Aballey, F.B. (2009).Bad Loans Portfolio: A case of ADB, MBA Dissertation,

University of Pretoria.
Adebola, S.S, Sulaiman, W., Yusoff, W., Dahalan, J. (2011), An ARDL Approach to the
Determinants of nonperforming loans in Islamic banking system in Malaysia. Arabian
Journal of Business and Management Review, 1(2).
Abdus.F (2004) Bahrain’s Commercial Banks Performance; Credit and financial performance.
New Delhi: Vikas Publishing House.
Basu, A. (1998). Banking system reforms in sub-Saharan Africa: Status and challenges.

Published Working paper International Monetary Fund, Washington D.C.

Berge, T., and Boye, K., (2007). An analysis of bank's problem loans. Norges Bank Economic

Bulletin Vol. 78.

Berger, N. A. and De Young R. (1997), Problem Loans and Cost Efficiency in Commercial

Banks, Washington DC, Journal of Banking and Finance, Vol. 21.

Bercoff, Jose. J., Julian di, Govanni and Franque, Grimard (2002), Argentinean Banks, Credit

Growth and the Lequila Crisis: A Duration Analysis.

Bikker, J.A., Hu, H. (2002), cyclical patterns in profits, provisioning and lending of banks and
procyclicality of the new Basel capital requirements. BNL Quarterly Review, 55.
Bloem, A. M. and Gorters, C. N. (2001).The Macroeconomic Statistical Treatment of,
Nonperforming Loans: Discussion Paper Statistics Department of the International
Monetary Fund WP/01/209.
Brooks C 2008, Introductory Econometrics for Finance, 2nd edn, Cambridge University Press,
New York
Brown Bridge, M.C., Harvey, A and Fritz, D; (1998). Financial Distress in Local banks in
Kenya, Nigeria, Uganda and Zambia: Causes and Implications for Regulatory Policy.
Development Policy Review, 16.

53
Capiro Gerard and Daniela Klingebiel 1999,”Bank insolvencies: cross country experience”.
Policy research working paper 1620, World Bank Washington D.C.
Charles Smithson, (2003) Credit Portfolio Management, Canada: John Wiley & Sons.

Creswell (2009): Research Design: Qualitative, quantitative and mixed method approaches
3rd Edition.
Damodar N Gujarati; Dawn C Porter (2009) Basic Econometrics, Department of Financial

Services, Ministry of Finance November 2014. Financial inclusion, regulation and

inclusive growth in Ethiopia working paper 408.

Fan, L., Shaffer, S. (2004). “Efficiency versus Risk in Large Domestic US Banks.” Managerial

Finance, Vol. 30, pp. 1-19.

Fofack, H. (2005), Non-performing loans in sub-Saharan Africa: Causal Analysis and

Macroeconomic Implications. World Bank Policy Research Working Paper n° 3769.

Fisher, I. (1999).The Debt- Deflation Theory of Great Depressions. Econometrical.

Fawad Ahmad and 2Taqadus Bashir, 2013, Explanatory Power of Macroeconomic Variables as
Determinants of Non-Performing Loans: Evidence from Pakistan, World Applied
Sciences Journal 22 (2):

Gadise Gezu (2014). “Determinants of Non-Performing Loans.” Empirical in the Case of the
Ethiopian Commercial Banks. Jimma University
Grier, Waymond A. (2007). “Credit Analysis of Financial Institutions”. 2nd ed. Euro money
Institution Investor PLC.
Habtamu Gemechu. E (2015), Assessment of Factors Affecting Non-Performing Loans the case
of Ethiopian Private Banks. A.A University
Holger M. Mueller (2008), Bank capital structure and credit decisions, University of Frankfurt
(IMFS), p. 295–314. J. Financial Intermediation.
Hong and Sung (1995), performance of Korean banks.

Hou, Y. (2007), The Non-performing Loans: Some Bank-level Evidences. The 4th Advances in

Applied Financial Economics, the Quantitative and Qualitative Analysis in Social Sciences
conferences.

54
J. Glen and C. Mondragón-Vélez: “Business Cycle Effects on Commercial Bank Loan Portfolio
Performance in Developing Economies”, International Finance Corporation, World Bank
Group, 2011.
Jimenez, G., Saurina J. (2005), Credit cycles, credit risk, and prudential regulation. International
Journal of Central Banking 2.
Kangimba, E.J. 2010. Assessment of causes for non-performing loans in commercial Banks:
Case of standard charted bank, MBA thesis, University of Dodoma
Keeton, William and Morris, S. (2003).Why Do Banks loan losses Differ? Federal Reserve Bank
of Kansas City, Economic Review.
Kennedy P 2008, Guide to econometrics, 6th edn, Blackwell Publishing, Malden.

Khemraj, T.,Pashas.(2009);The determinant of NPL, An econometric case study of Guyana The

Caribbean Center for Banking and Finance Bi- Annual Conference on banking and finance,
St. Augustine Trinidad.
Khon .T.F and Best. K (2007), Performance Evaluation of selected private commercial
banks in Bangladesh, New Delhi: Vikas Publishing House.
Kirui, S. (2014), The effects of Non-performing loans on profitability of commercial Banks in
Kenya. MBA project, University of Nairobi.
Louzis, D.P., Vouldis, A.T., Metaxas, V.L. (2010), Macroeconomic and bank‐specific
Determinants of non‐performing loans in Greece: a comparative study of mortgage,
business and consumer loan portfolios. Bank of Greece, Working Paper, n°118.

Marijana Curak , Sandra Pepur , Klime Poposki (2013) “Determinants of non-performing loans
evidence from Southeastern European banking systems” Banks and Bank Systems,
Volume 8, Issue 1
Mwakoba,L. H. (2011). Credit Risk Management in Financial Institutions in Tanzania,
un published MBA Dissertation, University of Dodoma.
NBE, Licensing and Supervision of Banking Business, Directives No. SBB/43/2008

NBE 2012/13, “Annual report”, Domestic Economic Analysis and Publication

Directorate National Bank of Ethiopia

NBE 2015/16, “Annual report”, Domestic Economic Analysis and Publication

Directorate National Bank of Ethiopia

55
NBE 2016/17, “Annual report”, Domestic Economic Analysis and Publication
Directorate National Bank of Ethiopia

Negera, W. (2012).Determinants of Non-Performing Loans the Case of Ethiopian Banks.

Unpublished Research Report, Graduate School of Business Leadership University of
South Africa
Nkusu, M. (2011), Nonperforming Loans and Macro-financial Vulnerabilities in Advanced
Economies, IMF Working Paper WP/11/161.
Nikola Radivojevic, 1 Jelena Jovovic, 2017, Examining of Determinants of Non-performing
Loans High Technical School at Applied Studies, Kragujevac, Serbia
Pervan, M., Pelivan, I., and Arnerić, J. (2015). Profit persistence and determinants of bank
profitability in Croatia. Econ. Res.-Ekon. Istraživanja28, 284–298.
Rajan, R., Dhal, S.C. (2003), Non-performing loans and terms of credit of public sector banks in
India: An empirical assessment. Reserve Bank of India Occasional Papers, 24(3).
Rajiv Ranjan and Sarat Chandra Dhal (2003), Non-Performing Loans and Terms of Credit of
Public Sector Banks in India: An Empirical Assessment
Ralf Ewert, Gerald Schenk & Andrea Szczesny (2000), Determinants of Bank Lending
Performance in Germany. Business Review.
Ricardas Mileris, 2014, Macro-economic Factors of Non-performing loans in Commercial Banks,
Issn 1392-1258 Ekonomika Vol. 93
Rodriguez-Moreno, M., Pena, J. I. (2013). Systemic risk measures: the simpler the better? Journal
of Banking & Finance, Vol. 37.
Salas, V., Saurina, J. (2003), Credit risk in two institutional regimes: Spanish commercial and
savings banks. Journal of Financial Services Research, 22(3).
Shingjergji, A. (2013). The Impact of Bank Specific Variables on the Non-Performing Loans
Ratio in the Albanian Banking System. Research Journal of Finance & Accounting, 4(7).
Sinkey, J.F., Greenwalt, M. (1991), Loan-loss experience and risk-taking behavior at large
commercial banks. Journal of Financial Services Research, 5(1).
Sufian, F. and Habibullah, M. S. (2009). Determinants of bank profitability in a developing
economy: Empirical evidence from Bangladesh. J. Bus. Econ. Manag.10, 207–217.

56
Tan, Y. and Floros, C. (2012). Bank profitability and GDP growth in China: a note. J. Chin.
Econ. Bus. Stud.10, 267–273.
Tekalagn Getahun, Lu Anwaen & Md. Shafiqul Bari, (2015), Credit Risk Management and Its
Impact on Performance of Commercial Banks: In of Case Ethiopia, Research Journal of
Finance and Accounting, Vol. 6 No. 24
Tsango (2007), Tanzania Banking and Finance Sector Report,Southern Sun. Dar es Salaam.

Waweru, N. M. & Kalani, V. M. (2009). “Commercial Banking Crises in Kenya: Causes and
Remedies”, African Journal of Accounting. Vol. 4, No.
William F. Caton, 1997 Uniform Financial Institutions Rating System, Federal Deposit Insurance

Corporation p. 6).

Wondimu, (2007), Analysis of technical efficiency of banks in loan delivery. A comparative

study on commercial Bank of Ethiopia, United Bank and construction and Business Bank,
unpublished thesis.
Wondimagegnehu Negera (2012). “Determinants of nonperforming loans” .The case of

Ethiopian Banks.

57
58
Appendixes
Appendix I: Regression Result
Dependent Variable: NPL
Method: Panel Least Squares
Date: 03/22/18 Time: 11:52
Sample: 2002 2016
Periods included: 15
Cross-sections included: 7
Total panel (balanced) observations: 105

Variable Coefficient Std. Error t-Statistic Prob.

C -0.014029 0.013328 -1.052598 0.2952

BS 0.002377 0.00037 6.430936 0.0000
EXR -0.000826 0.000187 -4.422019 0.0000
GDP 0.025293 0.012902 1.960394 0.0528
INFR -0.001427 0.004268 -0.334385 0.0388
LG -0.006013 0.002593 -2.3189 0.0225
LIQ 0.000538 0.000152 3.527652 0.0006
LR -0.275964 0.109567 -2.51867 0.0134
ROA 0.051112 0.053626 0.95311 0.0343

Effects Specification

Cross-section fixed (dummy variables)

R-squared 0.793398 Mean dependent var 0.012294

Adjusted R-squared 0.769534 S.D. dependent var 0.006628
S.E. of regression 0.004002 Akaike info criterion -3.122067
Sum squared resid 0.001538 Schwarz criterion -2.894585

59
 Log likelihood 135.4085 Hannan-Quinn criter. -3.029887
F-statistic 21.6504 Durbin-Watson stat 1.662771
Prob(F-statistic) 0.00000
Appendix II: Heteroscedasticity White Test

F-statistic 3.174724 Prob. F(44,60) 0.15399

Obs*R-squared 73.45076 Prob. Chi-Square(44) 0.26011
Scaled explained SS 77.25394 Prob. Chi-Square(44) 0.48432

Test Equation:
Dependent Variable: RESID^2
Method: Least Squares
Date: 03/22/18 Time: 12:27
Sample: 1 105
Included observations: 105

Variable Coefficient Std. Error t-Statistic Prob.

C -0.005289 0.014668 -0.360609 0.7197
BS^2 5.99E-06 2.390000E-06 2.50938 0.0148
BS*EXR 8.82E-08 1.72E-06 0.051338 0.9592
BS*GDP 7.13E-05 8.82E-05 0.808317 0.4221
BS*INFR 1.16E-05 4.76E-05 0.243975 0.8081
BS*LG 6.58E-06 1.44E-05 0.458171 0.6485
BS*LIQ 1.11E-07 1.09E-06 0.101542 0.9195
BS*LR -0.001765 0.000819 -2.154921 0.0352
BS*ROA -0.00047 0.000405 -1.160758 0.2503
BS -6.63E-05 0.000111 -0.594375 0.5545
EXR^2 5.64E-07 1.74E-06 0.323484 0.7475
EXR*GDP 2.03E-05 0.000809 0.025099 0.9801
EXR*INFR -1.10E-05 4.52E-05 -0.244238 0.8079
EXR*LG 1.11E-05 9.56E-06 1.164683 0.2488
EXR*LIQ 9.30E-07 4.34E-07 2.143833 0.0361
EXR*LR -0.000573 0.001701 -0.336763 0.7375
EXR*ROA 0.000456 0.00018 2.531627 0.014
EXR -5.12E-06 0.000187 -0.027393 0.9782
GDP^2 0.004295 0.005326 0.806371 0.4232
GDP*INFR 1.98E-05 0.002689 0.007354 0.9942
GDP*LG -9.99E-05 0.000669 -0.149237 0.8819
GDP*LIQ -9.88E-05 4.87E-05 -2.029251 0.0469
GDP*LR -0.051175 0.62439 -0.08196 0.935

60
 GDP*ROA 0.004055 0.012588 0.322153 0.7485
GDP 0.007257 0.060823 0.119315 0.9054
INFR^2 0.000176 0.000929 0.189247 0.8505
INFR*LG 0.000342 0.000281 1.216749 0.2285
INFR*LIQ -4.62E-06 1.53E-05 -0.301969 0.7637
INFR*LR -0.00601 0.017543 -0.342587 0.7331
INFR*ROA 0.007608 0.004069 1.869515 0.0664
INFR 0.00036 0.002371 0.151731 0.8799
LG^2 -0.000129 8.31E-05 -1.554189 0.1254
LG*LIQ -7.71E-06 6.98E-06 -1.103684 0.2741
LG*LR -0.016985 0.007204 -2.357891 0.0217
LG*ROA 0.002807 0.003417 0.821429 0.4147
LG 0.001868 0.000774 2.414748 0.0188
LIQ^2 -2.47E-07 2.68E-07 -0.920322 0.3611
LIQ*LR -0.000601 0.000349 -1.724879 0.0897
LIQ*ROA 0.000279 0.000166 1.687688 0.0967
LIQ 7.44E-05 3.99E-05 1.86411 0.0672
LR^2 0.085737 0.827588 0.103599 0.9178
LR*ROA -0.175838 0.106287 -1.654377 0.1033
LR 0.066623 0.228823 0.291155 0.7719
ROA^2 -0.070496 0.038499 -1.831104 0.0721
ROA 0.014759 0.014919 0.989279 0.3265

R-squared 0.699531 Mean dependent var 1.46E-05

Adjusted R-squared 0.479187 S.D. dependent var 2.33E-05
S.E. of regression 1.68E-05 Akaike info criterion -18.8473
Sum squared resid 1.70E-08 Schwarz criterion -17.7099
Log likelihood 1034.482 Hannan-Quinn criter. -18.3864
F-statistic 3.174724 Durbin-Watson stat 1.79331
Prob(F-statistic) 0.000019

Appendix III: Normality Test

14
Series: Residuals
12 Sample 1 105
Observations 105
10
Mean -1.48e-16
Median 0.029858
8 Maximum 1.045535
Minimum -1.095375
6 Std. Dev. 0.382696
Skewness -0.193710
4 Kurtosis 3.411567

61 Jarque-Bera 1.397736
2
Probability 0.497148
0
-1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0
Appendix IV: Raw Data
Appendix IV: Choosing Random Vs Fixed Effect Model
Dependent Variable: NPL
Method: Panel Least Squares
Date: 08/31/11 Time: 00:51
Sample: 2002 2016
Periods included: 15
Cross-sections included: 7
Total panel (balanced) observations: 105

Variable Coefficient Std. Error t-Statistic Prob.

BS 0.319513 0.118914 2.686916 0.0086

EXR -0.0662 0.020954 -3.15935 0.0022
INFR 0.005517 0.00342 1.613041 0.1102
GDP 0.002942 0.011563 0.25445 0.7997
LG -0.00346 0.002218 -1.5608 0.1221
LIQ 1.812623 0.287478 6.305263 0
LR -0.35785 0.089203 -4.01163 0.0001
ROA -0.19748 0.04782 -4.12971 0.0001
C 3.112543 0.94238 3.302854 0.0014

Effects Specification

Cross-section fixed (dummy variables)

R-squared 0.815936 Mean dependent var 1.229421

Adjusted R-squared 0.787304 S.D. dependent var 0.662785
S.E. of regression 0.30567 Akaike info criterion 0.598942
Sum squared resid 8.409067 Schwarz criterion 0.978079
Log likelihood -16.4444 Hannan-Quinn criter. 0.752575
F-statistic 28.49715 Durbin-Watson stat 1.262905
Prob(F-statistic) 0.00000

62
 Bank Loan Return Inflation Lending Exchange
Year Banks NPLs Liquidity GDP
Size Growth on Asset Rate Rate Rate

2002 CBE 0.03203 23.03 37.56 -0.08510 -0.02159 -0.10572 0.01634 0.10750 8.56600

2003 CBE 0.03367 23.12 42.35 -0.12286 0.02352 0.10924 -0.02099 0.10750 8.60000

2004 CBE 0.03195 23.26 43.01 -0.02666 0.01280 0.07347 0.11729 0.10750 8.63000

2005 CBE 0.03054 23.43 42.32 0.14787 0.01871 0.06126 0.12644 0.10500 8.66000

2006 CBE 0.02874 23.50 43.17 -0.02721 0.02324 0.10577 0.11539 0.10500 8.69000

2007 CBE 0.02653 23.68 43.57 0.04981 0.02179 0.15823 0.11795 0.10500 9.03000

2008 CBE 0.01808 23.82 38.59 0.77667 0.02900 0.25300 0.11187 0.11500 9.61000

2009 CBE 0.01131 23.97 35.81 0.20576 0.03498 0.36400 0.10041 0.12250 11.30090

2010 CBE 0.00588 24.17 33.52 0.14880 0.02947 0.02800 0.10567 0.12250 13.53210

2011 CBE 0.00916 24.55 35.75 0.49816 0.03038 0.18100 0.11400 0.11880 16.90810

2012 CBE 0.00788 24.83 30.69 0.73186 0.03980 0.34100 0.08700 0.11875 17.73050

2013 CBE 0.00956 25.01 31.48 0.14812 0.03432 0.13500 0.09900 0.11880 18.64260

2014 CBE 0.01030 25.17 27.86 0.21968 0.03056 0.08100 0.10348 0.11880 19.57710

2015 CBE 0.00875 25.63 32.39 -0.30310 0.01591 0.07700 0.10400 0.11880 20.56590

2016 CBE 0.00993 26.07 32.73 0.31646 0.02880 0.09689 0.08000 0.12750 21.80040

2002 AB 0.01335 19.48 38.40 0.20000 0.02317 0.05356 0.03400 0.10750 8.56600

2003 AB 0.01705 19.80 38.65 0.25589 0.01114 0.10924 -0.02099 0.10750 8.60000

2004 AB 0.02041 20.12 39.29 0.18250 0.01640 0.07347 0.11729 0.10750 8.63000

2005 AB 0.01825 20.42 37.99 0.36364 0.01902 0.06126 0.12644 0.10500 8.66000

2006 AB 0.01589 20.78 35.89 0.45116 0.03012 0.10577 0.11539 0.10500 8.69000

2007 AB 0.01459 21.10 35.90 0.34188 0.04216 0.15823 0.11795 0.10500 9.03000

2008 AB 0.01526 21.38 38.64 0.08992 0.03302 0.25300 0.11187 0.11500 9.61000

2009 AB 0.01705 21.71 41.62 -0.00909 0.02543 0.36400 0.10041 0.12250 11.30090

2010 AB 0.01548 21.95 41.93 0.15949 0.03446 0.02800 0.10567 0.12250 13.53210

2011 AB 0.01281 22.22 39.57 0.26728 0.03994 0.18100 0.11400 0.11880 16.90810

2012 AB 0.00993 22.39 35.36 0.38083 0.03577 0.34100 0.08700 0.11875 17.73050

63
2013 AB 0.00833 22.62 33.49 0.40064 0.03788 0.13500 0.09900 0.11880 18.64260

2014 AB 0.00833 22.93 35.16 0.19019 0.03543 0.08100 0.10348 0.11880 19.57710

2015 AB 0.00531 23.11 30.43 0.36024 0.02940 0.07700 0.10400 0.11880 20.56590

2016 AB 0.00405 23.32 32.34 0.23784 0.02782 0.09689 0.08000 0.12750 21.80040

2002 BOA 0.01740 19.52 37.55 0.22129 0.01856 -0.10572 0.01634 0.10750 8.56600

2003 BOA 0.02041 19.73 36.90 0.45298 0.01553 0.10924 -0.02099 0.10750 8.60000

2004 BOA 0.02028 19.97 36.90 0.33386 0.02399 0.07347 0.11729 0.10750 8.63000

2005 BOA 0.01589 20.32 35.85 0.32071 0.02329 0.06126 0.12644 0.10500 8.66000

2006 BOA 0.01131 20.73 34.38 0.41756 0.03339 0.10577 0.11539 0.10500 8.69000

2007 BOA 0.01548 20.96 35.37 0.26043 0.03533 0.15823 0.11795 0.10500 9.03000

2008 BOA 0.02186 21.23 38.59 0.09880 0.03447 0.25300 0.11187 0.11500 9.61000

2009 BOA 0.02282 21.53 40.83 0.01586 0.02846 0.36400 0.10041 0.12250 11.30090

2010 BOA 0.02001 21.68 39.47 0.13419 0.02934 0.02800 0.10567 0.12250 13.53210

2011 BOA 0.01194 21.85 39.62 0.23148 0.03337 0.18100 0.11400 0.11880 16.90810

2012 BOA 0.00956 21.99 37.15 0.30660 0.04052 0.34100 0.08700 0.11875 17.73050

2013 BOA 0.00693 22.22 36.44 0.09091 0.03256 0.13500 0.09900 0.11880 18.64260

2014 BOA 0.00588 22.33 36.11 0.06401 0.03416 0.08100 0.10348 0.11880 19.57710

2015 BOA 0.00405 22.54 33.29 0.22242 0.03121 0.07700 0.10400 0.11880 20.56590

2016 BOA 0.00336 22.75 34.07 0.10134 0.02726 0.09689 0.08000 0.12750 21.80040

2002 DB 0.01131 19.88 38.68 0.13547 -0.00196 -0.10572 0.01634 0.10750 8.56600

2003 DB 0.01361 20.28 38.53 0.25589 0.00485 0.10924 -0.02099 0.10750 8.60000

2004 DB 0.01308 20.66 38.97 0.18250 0.02605 0.07347 0.11729 0.10750 8.63000

2005 DB 0.01163 20.96 38.43 0.36364 0.03350 0.06126 0.12644 0.10500 8.66000

2006 DB 0.00993 21.31 35.80 0.45116 0.03476 0.10577 0.11539 0.10500 8.69000

2007 DB 0.00916 21.64 36.26 0.34188 0.02151 0.15823 0.11795 0.10500 9.03000

2008 DB 0.00833 21.93 37.25 0.08992 0.00380 0.25300 0.11187 0.11500 9.61000

2009 DB 0.00833 22.17 40.94 -0.00909 0.02062 0.36400 0.10041 0.12250 11.30090

2010 DB 0.00788 22.43 40.54 0.15949 0.02392 0.02800 0.10567 0.12250 13.53210

64
2011 DB 0.00693 22.61 38.64 0.26728 0.02669 0.18100 0.11400 0.11880 16.90810

2012 DB 0.00742 22.79 36.18 0.38083 0.02788 0.34100 0.08700 0.11875 17.73050

2013 DB 0.00788 22.92 31.44 0.40064 0.02355 0.13500 0.09900 0.11880 18.64260

2014 DB 0.00642 23.02 34.08 0.19019 0.04180 0.08100 0.10348 0.11880 19.57710

2015 DB 0.00531 23.14 40.33 0.36024 0.02339 0.07700 0.10400 0.11880 20.56590

2016 DB 0.00531 23.28 31.25 0.23784 0.02365 0.09689 0.08000 0.12750 21.80040

2002 NB 0.00182 18.37 37.90 0.18023 0.00976 -0.10572 0.01634 0.10750 8.56600

2003 NB 0.01386 19.15 37.98 0.40640 0.01433 0.10924 -0.02099 0.10750 8.60000

2004 NB 0.01335 19.64 38.44 0.29247 0.03154 0.07347 0.11729 0.10750 8.63000

2005 NB 0.01411 20.09 38.74 0.35772 0.03483 0.06126 0.12644 0.10500 8.66000

2006 NB 0.01361 20.30 36.16 0.58982 0.03665 0.10577 0.11539 0.10500 8.69000

2007 NB 0.01224 20.63 38.81 0.35279 0.03903 0.15823 0.11795 0.10500 9.03000

2008 NB 0.01335 21.04 41.08 0.08899 0.03651 0.25300 0.11187 0.11500 9.61000

2009 NB 0.01526 21.37 43.59 -0.09988 0.03908 0.36400 0.10041 0.12250 11.30090

2010 NB 0.01361 21.63 43.49 0.17113 0.04113 0.02800 0.10567 0.12250 13.53210

2011 NB 0.01411 21.83 42.41 0.17631 0.04684 0.18100 0.11400 0.11880 16.90810

2012 NB 0.00993 22.00 38.81 0.22530 0.04099 0.34100 0.08700 0.11875 17.73050

2013 NB 0.00916 22.11 36.04 0.31536 0.03664 0.13500 0.09900 0.11880 18.64260

2014 NB 0.00742 22.28 30.61 -0.01828 0.02818 0.08100 0.10348 0.11880 19.57710

2015 NB 0.00405 22.50 32.10 0.31872 0.02825 0.07700 0.10400 0.11880 20.56590

2016 NB 0.00588 22.69 33.31 0.23622 0.02512 0.09689 0.08000 0.12750 21.80040

2002 UB 0.00182 17.49 43.19 0.21642 0.01515 -0.10572 0.01634 0.10750 8.56600

2003 UB 0.00875 18.17 40.99 0.77914 0.01277 0.10924 -0.02099 0.10750 8.60000

2004 UB 0.01361 18.74 39.98 0.32414 0.01225 0.07347 0.11729 0.10750 8.63000

2005 UB 0.01361 19.43 40.25 0.54427 0.03549 0.06126 0.12644 0.10500 8.66000

2006 UB 0.01065 19.98 38.84 0.69309 0.03293 0.10577 0.11539 0.10500 8.69000

2007 UB 0.01099 20.40 38.96 0.40438 0.03385 0.15823 0.11795 0.10500 9.03000

2008 UB 0.00993 20.90 40.38 0.31891 0.03352 0.25300 0.11187 0.11500 9.61000

65
2009 UB 0.01131 21.34 42.30 0.15733 0.02369 0.36400 0.10041 0.12250 11.30090

2010 UB 0.01281 21.61 42.39 0.21437 0.03308 0.02800 0.10567 0.12250 13.53210

2011 UB 0.01030 21.92 40.72 0.25381 0.03404 0.18100 0.11400 0.11880 16.90810

2012 UB 0.00833 22.06 37.46 0.24670 0.03608 0.34100 0.08700 0.11875 17.73050

2013 UB 0.00642 22.20 32.42 0.15308 0.02278 0.13500 0.09900 0.11880 18.64260

2014 UB 0.00336 22.39 36.38 0.07618 0.01814 0.08100 0.10348 0.11880 19.57710

2015 UB 0.00182 22.59 31.39 0.35317 0.02144 0.07700 0.10400 0.11880 20.56590

2016 UB 0.00262 22.78 31.08 0.24406 0.02144 0.09689 0.08000 0.12750 21.80040

2002 WB 0.01589 18.67 38.80 0.54286 0.02989 -0.10572 0.01634 0.10750 8.56600

2003 WB 0.01629 19.15 37.26 0.69753 0.01832 0.10924 -0.02099 0.10750 8.60000

2004 WB 0.01758 19.51 36.83 0.42909 0.03283 0.07347 0.11729 0.10750 8.63000

2005 WB 0.01629 20.00 36.36 0.44148 0.03088 0.06126 0.12644 0.10500 8.66000

2006 WB 0.01569 20.44 34.00 0.30185 0.03086 0.10577 0.11539 0.10500 8.69000

2007 WB 0.01482 20.99 36.12 0.23186 0.03280 0.15823 0.11795 0.10500 9.03000

2008 WB 0.01775 21.19 39.88 0.16335 0.03613 0.25300 0.11187 0.11500 9.61000

2009 WB 0.01808 21.45 42.60 0.05037 0.03634 0.36400 0.10041 0.12250 11.30090

2010 WB 0.01386 21.58 43.09 0.14676 0.03728 0.02800 0.10567 0.12250 13.53210

2011 WB 0.01504 21.97 42.58 0.08656 0.03768 0.18100 0.11400 0.11880 16.90810

2012 WB 0.00875 22.01 39.33 0.34064 0.03720 0.34100 0.08700 0.11875 17.73050

2013 WB 0.00788 22.25 35.23 0.22489 0.03437 0.13500 0.09900 0.11880 18.64260

2014 WB 0.00531 22.33 31.86 0.19035 0.02990 0.08100 0.10348 0.11880 19.57710

2015 WB 0.00470 22.54 29.12 0.27485 0.02809 0.07700 0.10400 0.11880 20.56590

2016 WB 0.00470 22.71 31.77 0.08963 0.02680 0.09689 0.08000 0.12750 21.80040

66
67