The Impacts of Liquidity on Credit risk of Commercial Banks in Ethiopia

Research Paper Submitted to the College of Business and Economics in Bahir

Dar University in Partial Fulfillment of the Requirements for the Degree of
Master of Science in Accounting and Finance

Bahir Dar University

College of Business and Economics

Department of Accounting and Finance

By: Abdulaziz Kassa

Advisor: Tilahun Aiemro (Asst. Prof.)

Bahir Dar, Ethiopia

February, 2018

1
 Declaration Statement

I, Abdul-Aziz Kassa, declare that this thesis work on the title “the impacts of Liquidity on Credit risk of
Commercial Banks in Ethiopia” is original, work of my own effort and study and Submitted in Partial
fulfillment of the Requirements for Degree of Master of Accounting and Finance of Bahir Dar
University. Besides, all sources of materials used for this study have been properly acknowledged. This
work has been neither submitted for any thesis in any university nor published previously to the best of
the researcher’s knowledge.

Declared by:
[
Confirmed by Advisor:
Name: Abdulaziz Kassa Endrise Name: Tilahun Aemiro (Asst.Prof.)
Signature_________________ Signature__________________
Date___________________ Date___________________

i
ACKNOWLEDGEMENT
Prays and truthful thanks to Allah who gave me the patience and the ability to accomplish this paper.
I would like to thank my advisor, Tilahun Aiemro, for his guidance through the year researcher spent
under his supervision is greatly appreciated. This work is better for his inputs and directions regarding
the impact of liquidity on credit risk of banking business in Ethiopia.
In addition, I extend my appreciations to my wife Ms Neima Yimam for her unreserved
encouragements and help for the accomplishment of this research paper.
I am indebted to express my gratitude to Bahir Dar University, specifically College of Business and
Economics, for giving me the chance to conduct this research paper.
In addition, I express my gratitude to the people who have helping and support me during the study
period in Bahir Dar University. Also, special thanks are furthermore given to my colleagues in the
college for all the time, feedback and discussion.

ii
ABSTRACT
An appropriate credit risk level is a crucial decision for all business organization to be taken by
business organization for maximization of firm value and continued their growth. The main objectives
this study was to examine the effects of liquidity and other controlled variables on credit risk of
commercial banks of Ethiopia. Accordingly, the main concern of this study was to examine empirical
evidence from firm specific factors such as, liquidity, loan growth, bank size, profitability, bank capital
and macro variable like Inflation on credit risk of Ethiopian banking industry. To accomplish this
objective the research used panel data analysis and only secondary data were used. All banking
business was included in the sample frame if they have ten years annual report. Statistical tests like
descriptive statistics, correlation, and specific classical linear regression model assumption was tested.
A relationship was established between firm specific factors and credit risk measured by provision for
loan loss ratio of the banks for a period of ten years. The random regression results show that firm
profitability, bank Size, and inflation has significant impact on credit risk of Ethiopian banking industry
for the study period. Whereas firm liquidity, loan growth and bank capital has insignificant impact on
credit risk of banks. From this finding the researcher recommended that the sample of Ethiopian
commercial banking industry should give attention for significant variable that will result higher credit
risk such as profitability and size of banks. At the same time care should be take other variable even if
they have insignificant impact on credit risk.

iii
Table of contents

Contents
Table of contents ..................................................................................................................................................... iv
Chapter One ............................................................................................................................................................. 1
1.1. Background of the Study .............................................................................................................................. 1
1.2. Statement of the problem................................................................................................................................... 3
1.3 The Objective of the study ................................................................................................................................. 5
1.3.1 General objective ......................................................................................................................................... 5
1.3.2. Specific objectives ...................................................................................................................................... 5
1.4. Research hypotheses.......................................................................................................................................... 5
1.5. Significance of the study ................................................................................................................................... 5
1.6. Scope and delimitation of the study ................................................................................................................. 6
1.7. Limitation of the Study ...................................................................................................................................... 6
1.8. Organization of the study .................................................................................................................................. 7
CHAPTER TWO ...................................................................................................................................................... 8
2. LITERATURE REVIEW ..................................................................................................................................... 8
2.1. Credit Risk ...................................................................................................................................................... 8
2.2 Credit Risk Management Practices................................................................................................................. 8
2.3 Liquidity ......................................................................................................................................................... 9
2.4. Credit Risk and Liquidity ............................................................................................................................ 10
2.5. Theoretical Framework ................................................................................................................................... 11
2.5.1 Credit Risk Theory.................................................................................................................................. 11
2.5.2 Liquidity Preference Theory.................................................................................................................. 12
2.5.3 Liquidity Theory of Credit .......................................................................................................................... 12
2.5.3 Credit Risk Modeling ............................................................................................................................ 13
2.6 Determinants of Credit Risk ....................................................................................................................... 13
2.7. Review of related empirical studies ................................................................................................................ 16
2.7.1 Determinants of banks credit risk-empirical studies................................................................................. 16
2.7.2 Empirical Studies Related to the Study Variables ..................................................................................... 19
2.8. Knowledge Gap of the Study ........................................................................................................................... 21
2.9. Conceptual Framework of the Study ............................................................................................................... 22

iv
Chapter Three ......................................................................................................................................................... 23
3.1 Research Methodologies .................................................................................................................................. 23
3.2 Research Approach and Design ........................................................................................................................ 23
3.3. Data types and Collection Methods ............................................................................................................... 23
3.3 .1 Study Population and Sampling Technique ............................................................................................. 24
3.4. Definition and Measurement of Variables ...................................................................................................... 25
3.4 .1. Credit Risk (dependent variable) ............................................................................................................. 25
3.4.2. Liquidity (independent variable) .............................................................................................................. 26
3.4.3. Controlled Variables ................................................................................................................................. 27
3.5 Model Specification .......................................................................................................................................... 29
3.6 Method of data analysis and Presentation ...................................................................................................... 30
Chapter Four ........................................................................................................................................................... 31
4. Data presentation and analysis .......................................................................................................................... 31
4.1. Descriptive Statistics of the Data..................................................................................................................... 32
4.2 Pearson Correlation Matrix: ............................................................................................................................. 34
4.3 Unit Root Test ................................................................................................................................................... 35
4.4 Econometric Model Estimation Procedures and Specification Tests ............................................................... 35
4.5. Diagnostic Tests for Classical Linear Regression Model (CLRM) Assumptions ................................................ 36
4.5.1. Test of the Mean Value of Error term is Zero (E (ut) = ............................................................................. 36
4.5.2. Test of Normality Assumption .................................................................................................................. 36
4.5.3. Test of Multicollinarity ............................................................................................................................. 36
4.5.4 Autocorrelation ......................................................................................................................................... 37
4.5.5 Test of Heteroscedasticity: ........................................................................................................................ 37
4.6. Regression Analysis and Discussion................................................................................................................. 38
Chapter Five............................................................................................................................................................ 44
5. Conclusions, Recommendation and implications of the study .......................................................................... 44
5.1. Conclusions of the study ............................................................................................................................. 44
5.2 Recommendation ......................................................................................................................................... 46
5.3 Implications of the study ............................................................................................................................ 46
Reference. .............................................................................................................................................................. 48
Appendix ................................................................................................................................................................. 53

v
List of Figure
Figure 1: Conceptual Framework for the relationship of liquidity and Controlled Variables……….22

List of tables

Table 3.1 Sample of commercial banks and year of establishment…………………..25

Table 3.2: Measurement and expected sign of the study variables …………………..29
Table 4.1 Descriptive statistics of dependent and independent variables…………….32
Table 4.2 Person correlation matrix of dependent and independent variables ……….34
Table 4.3: Regression Result…………………………………………………………..38
Table 4.4 Summary of expected and actual signs of explanatory variables on the
dependent variables……………………………………………………………………43

vi
ACRONYMS

PLLR- Provision for Loan Loss Ratio

CRRI- Credit Risk
NPL- Non- performing Loan
NPLR- Non Performing Loan Ratio
CLRM- Classical Leaner Regression Model
FEM- Fixed effect model
CRGR- Credit Growth
GDP- Gross Domestic Product
BIS- Bank for International Settlement
DTMs- Deposit Taking Micro financial Institution
MOFED- Minister of Finance and Economic Development
REM – Random Effect Model
ROA- Return on Asset
LIQ – Liquidity
BCBS – Basel Committee of Banking Supervision
VIF- Variance Inflation Factor
NBE- National Bank of Ethiopia
BSIZE – Bank Size
CB- Capital Adequacy Ratio
INF- Inflation
CPI- Consumer Index Price

vii
Chapter One

1.1. Background of the Study

Financial institutions such as banks play a significant role in the economic growth of a given country
through transforming financial resource from surplus to deficit economies. According to (NBE, 2010),
banks are inherently exposed for risk and to some extent profit is the reward for taking risk since the
more the risky the business the more the return from the investment. In other case, a very high and
unsuccessfully managed risk can lead the bank in to distresses and failure. Therefore, risk become
warranted if and only if the risk are understandable, measurable, controllable and within a bank’s
capacity to resist the adverse results. Hence, banks should be at health condition to successfully
perform their responsibilities of availing funds to customers. The continued existence and accuracy of
the banking business naturally exposed to different risks like credit, liquidity, market, interest rate and
operational risk.

Credit risk refers to the possibility of loss as a result of default from an individual or an enterprise to
repay the loan granted (BCBS, 2001). Credit risk has an effect on financial performance of banks and
for the purpose of mitigating on the losses arising from lending activities; the bank should ensure that
lending risks should not be excessive. Credit risk is without a doubt the most key risk faced by Banks
and the success of their business highly depends on measuring in accurate way and efficient
management of this risk to a greater extent than any other risk (Gieseche, 2004).

Managing of credit risk in efficient way is an important part of the risk management system and it is
very crucial to financial institutions profitability and ultimately their survival in the market (Swarens,
1990).The key risk in a bank has been credit risk as it is suggested in different literature elsewhere in
the world and without any doubt failure to collect loans granted to customers has been the major factor
behind the collapse of many banks around the world. Hence, banks should have to effectively manage
credit risk in the entire business or the risk in individual credits or transactions (BCBS, 2001).

In addition, Banks should also consider the relationships between credit risk and other risks and the
bank should aware that credit risk does not exist in separate form from other risks rather it is closely

1
related with other risks. According to (NBE, 2010), effective credit risk management is the process of
managing an institution’s activities which create credit risk exposures, in a manner that significantly
reduces the chance that such activities will impact negatively on a bank’s earnings and capital.
According to (BIS, 2008) liquidity is defined as “the ability of a bank to fund increases in assets and
meet obligations as they come due, without incurring unacceptable losses”. The existence of liquidity
risk is from the fundamental role of banks in the maturity transformation of short-term deposits into
long-term loans. When a bank becomes unable to pay its obligation at maturity; the bank will exposed
to liquidity risk.

Liquidity risk has two types: funding liquidity risk and market liquidity risk. Funding liquidity risk is
the risk that the bank unable to meet efficiently both the expected and unexpected current and future
cash flow and collateral needs without affecting either daily operations or the financial condition of the
firm. Market liquidity risk is the risk that a bank may not easily offset or eliminate a position at the
market price because of insufficient market depth or market trouble (Vodová, 2013).The purpose of
good liquidity management is to ensure that every bank is able to meet fully its contractual
commitments. The ability of increasing fund to meet their obligations as they come due is critical to the
continuing feasibility of bank. Therefore, effective management of liquidity is the most important
activities conducted by banks.

Having effective risk management is very crucial for banks especially now a day with a volatile and
unstable environment every banks are becoming victim of enormous risks like: liquidity risk ,credit
risk, operational risk, market risk, foreign exchange risk and interest rate risk all of those may have an
impact on the existence and successes of banks (Al-Tamimi & Al-Mazrooei, 2009). For sound and
effective managing of credit risk, National bank of Ethiopia, conducted a survey on December, 2009, to
identify status of risk management practice through distributing questionnaires for a sample of 15
Ethiopian banks.
Based on the study the Ethiopian banks mostly affected by credit risk, operational risk and liquidity risk
as 60%, 53% and 40% weight of response respectively. Based on that study, credit, liquidity and
operational risks were the key bank risks for the last two years and they would continue for the next
five years more with 100%, 80% and 80% weight of response respectively. However, the study does
not see the relationship between those risks such as the relationship between liquidity with credit risk in

2
Ethiopian banks. Hence, the objective of this study will be investigating the effect of liquidity on credit
risk of Ethiopian banks which is open for empirical study.

1.2. Statement of the problem

The existence of bank is to provide financial intermediation services and at the same time maximizing
profit and shareholders' value. Hence, banks to be at health condition for such kind of intermediation
services particularly in developing countries such as Ethiopia where banks become the most dominant
financial institutions in the financial system since there is no capital market. Accordingly Brown &
Moles, (2012), credit risk defined as the possibility that a borrower will fail to pay obligations in accordance
with the agreed terms. Accepting deposit and granting loan and advances is the main function for most
banks and loan is the largest source of income and at the same time it is the source of credit risk mostly
which is directly related to non-performing loans, a loan that borrower might not be paid back
accordingly with the agreement. Managing of credit risk is the most important component of a
comprehensive approach to risk management and essential to the long-term success of any banking
business (BCBS, 2001).

An effective management of credit risk will enable the banks to meet its financial obligations and take
advantage of profitable investments that yield better returns for the future. In other words, financial
institutions which have an appropriate balance of their liquidity risk and credit risk are able to transfer
their funds into profitable investments. The optimal level of liquidity is determined by the credit
management practices implemented by a bank for alleviate the exposure to credit risk (Myers & Majluf,
2004). Poor level of credit risk management and low levels of liquidity position are the two major
causes of Bank failures and considered as the key risk sources in terms of credit and liquidity risk and
attracted great attention from researchers in recent time (Engdawork, 2014).

NPL which is taken as proxy for credit risk measurement in various literatures like (Ahmad & Ariff,
2007), (Zuhair.A and Nasif.F, 2013), have serious impact on the bank’s overall banks profitability and
even economic growth of a given country. For effective controlling of the increasing non-performing
loans in Ethiopian banking business, National Bank of Ethiopia has issued a directive that enforce all
banks to maintain ratio of non -performing below five percent in 2008 which is similar with standard
set by Basel Committee. Despite the NBE requirement, banks in Ethiopia experiencing high percentage

3
of nonperforming loan which indicates they have credit risk problem. According to Tesfa,(2016) un
published research, the ratio of NPL( indicators for credit risk in most literatures) for Dashen Bank it
was 7.4% in 2009, for Wegagen Bank it was 7.7% in 2009; for Cooperative of Bank Oromiya it was
9.5% on 2009, for Nib International Bank it was 9.1% & 7.3% in 2009 & 2010 respectively and for
Zemen Bank it was 8.8% in 2013 and all of them show high level of NPL compared with the standard
set by NBE, and Basel standard limit. According to ( Negera, 2012), the ratio of NPL for Cooperative
Bank of Oromia (CBO) stood at 11.54% on March 31, 2010 which is relativity very high compared
with the standard set by NBE or the industry average. Despite, there are some improvements during
recent quarters, the ratio of NPL remains higher. For instance the ratio of NPL showed 7.62% and
6.75% and 6.1% on June 30, 2010, December 31, 2010 and March 31, 2011 respectively. This indicates
that there is a credit risk problem in commercial banks of Ethiopia.

Different studies have been conducted on the relationship of credit risk and liquidity at international
level such as Harvey & Roper, (2014) , Philip, (2012), Bjorn & Rauch, (2014), Amir & Fatemeh,
(2014), Murage, (2016) have made a study in developed and developing country and reveled
inconsistence result in connection with relationship of credit risk and liquidity risk. Some of them
revealed that there is an inverse relation between those two risks and the other said that there is no
significant relationship among credit and liquidity risk. Despite the very importance of managing credit
and liquidity risk for the effectiveness of banks in performing their activities, there is no research
conducted for analyzing the relationship of credit and liquidity risk in the Ethiopian banking industry.

However, studies like Tehulu & Olana, (2014), Asfaw & Veni, (2015), Tesfa, (2016), conducted a
study for investigating determinants of credit risk in commercial banks of Ethiopia. They consider
micro variables only to see the credit risk factors and in some case and the others try to see
determinants of credit risk in private banks only. Macro variables have an impact on credit risk
management practice of banks and this study will fill this gap. In addition, to the best knowledge of the
researcher there is no study made to see the impact of liquidity credit risk. Does liquidity has an impact
on credit default of banking industry? Hence, the motive for this study is to see the impact of liquidity
on credit risk on commercial banks of Ethiopia by considering liquidity as a main variable and other
bank specific and macro variable as controlled variables.

4
1.3 Objective of the study

1.3.1 General objective

Related with the research problem, the general objective of the research will be determining the impact
of banks liquidity on credit risk in commercial bank of Ethiopia.

1.3.2. Specific objectives

To determine the effect of credit growth on credit risk in commercial banks of Ethiopia.
To determine the effect of Bank size on credit risk in commercial banks of Ethiopia.
To determine the effect of liquidity on credit risk in commercial banks of Ethiopia.
To determine the effect of Capital of bank on credit risk in commercial banks of Ethiopia.
To determine the effect of profitability on credit risk in commercial banks of Ethiopia.
To determine the effects of inflation on credit risk of banks in Ethiopia.

1.4. Research hypotheses

To achieve the objective of this study, the researcher has developed the following alternative hypothesis
to estimate the relationship between liquidity and nonperforming loans and also bank specific and
macroeconomic variables with nonperforming loans of commercial banks in Ethiopia.
H1: There is significant relationship between credit growth and credit risk of commercial banks in
Ethiopia.
H2: There is significant relationship between bank size and credit risk of commercial banks in Ethiopia
H3: There is significant relationship between bank capital and credit risk of commercial banks in
Ethiopia.
H4: There is significant relationship between liquidity and credit risk of commercial banks in Ethiopia
H5: There is significant relationship between inflation and credit risk of commercial banks in Ethiopia.
H6: There is significant relationship between profitability and credit risk of commercial banks in
Ethiopia.

1.5. Significance of the study

This study will have great contribution to the existing knowledge in the area of impacts of liquidity on
credit risk in Ethiopian context. The study will have also importance to identify the main causes of
credit risk; the finding of this study will enable management of the banks to come out with practical

5
policies aimed at improving the quality of their loan portfolios. It will also help the country
policymakers to implement effective monetary policies concerning credits and therefore minimizing the
percentage of non-performing loans in the economy. This in turn contributes to the well-being of the
financial sector of the economy and the society as a whole. Therefore, the major beneficiaries from this
study will be commercial bank, regulatory bodies, the academic staff of the country and the society as a
whole in the country.

1.6. Scope and delimitation of the study

Because of arising different firms in various sectors in Ethiopia, empirical studies to examine credit risk
were highly demanded. Though, scope of this study was only to examine the impact of liquidity on
credit risk of commercial banks of Ethiopia with other controlled variables. Therefore, the main
concern of this study was delimited to examine micro variable and macro variables affecting the credit
risk level of Ethiopian commercial banks. This study incorporated variables like liquidity, profitability,
size of bank, credit growth, bank capital, inflation. In the case of methodology, the scope of the study
was delimited to quantitative approach with the use of secondary data obtained from national bank of
Ethiopia and from their audited annual report for the period of 2007-2016. The analysis of the study
also delimited to quantitative analysis, since the study used secondary data. To make inference about
the credit risk factors of Ethiopian commercial banks in the population descriptive and random effect
model regression analysis were used accordingly.

1.7. Limitation of the Study

For conducting this study there were some limitations. First, despite examining the factors affecting
credit risk of financial and non- financial institution is so important but this study targeted only
commercial banks in Ethiopia ad only secondary data even if conducting the study by using both
primary and secondary data is more valuable. The second limitation of this study was related with its
sample size determination and its study period specifications. This study consider only nine banks are
sampled covering the period of 2007-2016 to investigate the credit risk level. The reason for this
sample size selection was to exclude new entrant banks (i.e. avoids new entrant bias). The researcher
want to take balanced panel data then only nine bank and ten year data were taken.

6
1.8. Organization of the study
This study has five chapters. The first chapter deals with background of the study, statement of the
problem, objectives, hypotheses, significance, scope and limitations of the study. Chapter two contains
a review of the literature including theories on liquidity and credit risk, review of prior empirical
studies. The third chapter discussed the research methodology including the research study description
area, research approach& design, data source and data, definition and measurement of variables, and
method of data analysis. The fourth chapter were present the statistical analysis of data that is
descriptive and regression results. Finally, the last chapter was present conclusions and
recommendations on the study.

7
CHAPTER TWO

2. LITERATURE REVIEW

2.1. Credit Risk

Credit risk has defined in different way by different authors. The most commonly definitions are:
Credit risk it is a financial exposure of bank’s due to relying on another party to perform an obligation
as agreed (NBE, 2010). It also defined by the Basel Committee on Banking Supervision (2001), credit
risk is the possibility of losing the outstanding loan partially or totally because of inability of the
counterparts according to the agreement. It can also be defined as the probability that a contractual
party may fail to meet its obligations in accordance with the agreed terms when it comes due. The
existence of Bank is not only to accept deposits but also to grant credit facilities for individual,
enterprise and inherently they are exposed to credit risk. Credit risk is considerably the most significant
risk faced by banks however, the success of the bank entirely depends on appropriate measurement and
efficient management of this risk to a greater extent than any other risks (Gieseche, 2004).

According to (Chen & Pan, 2012), credit risk is defined as the degree of value fluctuations in debt
instruments and derivatives due to changes in the underlying credit quality of borrowers and
counterparties. In addition (Coyle, 2000) define credit risk as losses from the refusal or inability of
credit customers to pay what is owed in full at time due. Credit risk is the exposure faced by Banks
when a borrower defaults in honoring debt obligations at maturity date. Credit risk sometimes called
‘counterparty risk’ is capable of putting the Bank in sorrow if it is not managed properly. Credit risk
management can maximizes Bank’s risk adjusted rate of return by maintaining credit risk exposure
within acceptable limit in order to provide framework for understanding the impact of credit risk
management on banks’ profitability (Kargi, 2011).

2.2 Credit Risk Management Practices

Risk management process is creating economic value in a firm through using financial instruments to
manage vulnerability to risk, especially credit risk, liquidity and market risk. According to Harker &
Satvros (1998), risk management is the process of making decisions concerning risks and their
subsequent implementation, and flows from risk estimation and risk evaluation. According to NBE,
(2010) effective credit risk management is the process of managing an institution’s activities which

8
create credit risk exposures, in a manner that significantly reduces the likelihood that such activities
will impact negatively on a microfinance institution’s earnings and capital. According to Charles
(2013), risk management is important for the existence of a bank and it enables management of the
bank to allocate resources for the risk units based on a compromise between risk and potential return.
Bank can minimize credit risk by using five main ways like: accurate determination of the price of the
credit, setting credit limits, use of collateral in credit, diversification of credit in various portfolio and
“Securitization” and/or the use of credit derivatives (Sheffernan, 2004).

2.3 Liquidity
According to BIS, (2008) liquidity defined as the ability of bank to fund increases in assets and meet
obligations as they come due, without incurring unacceptable losses. It is defined as the ability of a
business to generate sufficient cash from its operations to cover its obligations. Liquidity risk arises
from the fundamental role of banks in the maturity transformation of short-term deposits into long-term
loans. Therefore, banks have to hold optimal level of liquidity that can maximize their profit and enable
them to meet their obligation. Failure to manage liquidity can quickly result in significant unexpected
losses. It’s obvious the purpose of liquidity management is to ensure that every bank is able to meet
fully its contractual commitments. The ability to fund increases in assets and meet obligations as they
come due is basic to the ongoing viability of any bank. Therefore, proper management of liquidity is
among the most important activities conducted by any banks (NBE, 2010).

Proper liquidity management can minimize the probability of facing serious problems. Which means if
the bank is not liquid their asset becomes deteriorate since the bank loses the confidence of depositor
due to inability of meeting their obligation when it comes due. In fact, the importance of liquidity
transcends the individual bank, since a liquidity shortfall at a single bank can have system-wide
repercussions. For this reason, the analysis of liquidity requires the management of the bank not only to
measure the liquidity position of the bank on an ongoing basis, but also to examine how funding
requirements are likely to evolve under various scenarios. Liquidity risk has two types: funding/cash
liquidity risk and market/asset liquidity risk. Funding liquidity risk is the risk that the bank will not be
able to meet efficiently both expected and unexpected current and future cash flow and collateral needs
without affecting either daily operations or the financial condition of the firm. On the other hand

9
market liquidity risk is the risk that a bank cannot easily offset or eliminate a position at the market
price because of inadequate market depth or market disruption (BIS, 2008).

Liquidity gap (flow approach) and liquidity ratio (stock approach) are the most widely used
approaches to measure liquidity risk of banks. The liquidity gap approach considers the difference
between assets and liabilities both at the current and future periods. A positive liquidity gap means for
deficit, requiring for liabilities to be increased (Bessis, 2009). The liquidity gap consider liquid reserves
as a reservoir: the bank computes the required liquidity by comparing inflows and outflows during a
specified period. On the other hand, liquidity ratio uses various ratios to identify liquidity trend. The
various ratios make for immediate feasible source of funding. This in reality entitles portfolio of assets
that can be sold off without any excitement and also adequate amounts of stable liabilities. Most
importantly, ready credit line with other financial institutions.

2.4. Credit Risk and Liquidity

Liquidity and credit risk are the most risk faced by any financial institution such as banks and they are
closely related. According to the industrial organization models of banking such as the Monti-Klein
framework and the financial intermediation perspective in Harford & Maxwell (2005), Dybvig &
Diamond, (1983), liquidity structure and credit risk of financial institution closely linked, mainly with
related to borrowers failure to pay and fund withdrawals. As industrial organization model and different
literature reveals definitely there is a relationship between liquidity and credit risk in the financial
institution like banks. Liquidity risk considered as a profit lowering cost and default loan will increases
the liquidity risk since the cash flow become lowered. (Hooks & Linda, 2003).

The industrial organization model is based on the argument that banks obtain money from
inexperienced depositors which is used for lending and the problem will exist when too many projects
funded with loans which yield insufficient funds or even defaults. At this time the bank becomes unable
to meet depositors’ demands and due to this asset decline, depositors could claim back their money
which may in turn enforce the banks to recollect their loans and in this manner reduce aggregate
liquidity. Hence, higher credit risk may be accompanied by higher liquidity risk through depositors
demand (Harvey & Roper, 2014). On the other side of the relation of those most commonly risk face by
banks is that the liquidity of banks has an effect on probability of defaults that is credit risk. There is a
common perception that suggests firms that have larger cash holdings in their asset and investment

10
portfolio should be "safer." In particular, cash-rich firms should have a lower probability of default and
lower credit spreads other things remain constant. Other thoughts opined by studies control on liquidity
with default risk and their findings concerning the effect of liquidity on the probability of default are
inconclusive and often confusing.
According (Zmijewski, 1984), (Hillegeist et al.2004) the relationship of liquidity with the possibility of
default of loan positive which means when the liquidity of firm increase (become highly liquid), they
are rush to make loan to dispose of the excess liquidity. To the reverse, when the firm is not as such
liquid they are not ready to make loan and expose to less credit risk. In other case, studies made by in
(Shumway, 2001), (Ohlson, 1980) the finding was negative. This means when the bank liquidity
increase, the bank will safer as they may not want interest through making loan and they may not
interested rush to make loan and this lead cash-rich firms should have a lower probability of default and
lower credit spreads other things remain constant.
However, the finding related with correlation of liquidity with default mainly depends on the time
horizon over which default is being considered. If there is high liquidity (liquid asset reserve is large),
the short term probability of default is lower (lower credit risk) consistent with the common intuition.
Conversely, for long period of time (more than one year), the correlation between liquidity and
probability of default positive sign suggesting higher liquidity leads to higher credit risk.

Another studies made by Cherubini & Lunga (2001), Using a corporate bond as the underlying, they
opined that whenever liquidity risk increases, credit risk also increase. Further studies made by Boss &
Scheicher (2002), revealed a positive relationship between liquidity risk and credit risk in the European
corporate bond market. Moreover, studies made by Ericsson & Renault (2006) found a positive
relationship between liquidity risk and credit risk. As a consequence, any increase in liquidity risk
should be related with an increase in credit risk. Those studies indicate that liquidity has an impact on
credit risk level of firms.

2.5. Theoretical Framework

2.5.1 Credit Risk Theory
Credit risk theory is the first and principal readily available portfolio model for evaluation of credit
risk in the financial institution (Cantor &Frank, 1996). The credit risk approach has an ability to merge
the overall company credit risk across the organization and can provides a statement of value-at-risk

11
(VaR) for credit caused by upgrades, downgrades and defaults. The credit risk model is so crucial for
any firms which were exposed to credit risk in the course of running their business. The theory states
that, a business organization should set constructive method to measure credit risk across a broad range
of instruments, such as traditional loans, commitments and letters of credit; fixed income instruments;
commercial contracts such as trade credits and receivables; and market-driven instruments such as
swaps, forwards, options and other financial derivatives (Padilla & Pagano 2000).

2.5.2 Liquidity Preference Theory

The liquidity preference theory advocates that the premium demanded for parting with cash amount
increase as the term for getting the cash become decreases. When the credit term increase, the rate of
increment for premium slow down (Moore, 2010). According to (Mbole, 2004), Liquidity preference
theory associates the concept with investors who demand a premium for their securities which have
longer maturities and involve greater risk because the investor may prefer to hold cash rather investing
long term securities which result less risk. When an investment is more liquid, it is easy to sell quickly
in the market without losing its value.

According to liquidity preference theory, necessity of liquidity is to achieve the transactions motive, the
precautionary motive and the speculative motive (Pasinetti, 1997). Since financial institution emerged
to provide lending activities to borrowers, they may face liquidity problem especially when the
borrowers become unable or unwillingness to pay the loans at the maturity. Therefore, firms may not
participate in investing on profitable projects even it has higher returns in the future by fearing the
liquidity problem. Based on this concept, liquidity preference theory argues that a firm should hold
more cash for investment activities.

2.5.3 Liquidity Theory of Credit

Initially this theory was suggested by Emery (1984), proposes that credit rationed company use more
trade credit than those with normal access to financial institutions. The main point of this idea is that
when a firm has financial problem the offer of trade credit can make up for the reduction of the credit
offer from financial institutions. According to this view, those firms presenting good liquidity or better
access to capital markets can finance those that are credit rationed. Numerous approaches have tried to
obtain empirical evidence in order to support this assumption. Such as, Nielsen (2002), by using small
firms as a proxy for credit rationed firms, finds that when there is a monetary contraction, small firms
react by increasing the amount of trade credit accepted. As financially unconstrained firms are less
12
likely to demand trade credit and more prone to offer it, a negative relation between a buyer’s access to
other sources of financing and trade credit use is expected. Study made by Petersen & Rajan (1997)
supporting this negative relation.

2.5.3 Credit Risk Modeling

In order to measure and manage credit risk firms can use credit risk model. According to Walsh
(2010), credit risk models have three main functions like estimating the possibility of borrower will
default payments at maturity, measuring of dollar amount that a firm may lost when a borrower defaults
payments and lastly measuring the correlation of overall default risk across the entire credit exposure
portfolio. Jose & Riestra (2002) emphasize that the credit risk models are proposed at supporting the
financial institutions for management of risk, quantifying and aggregating risk across geographical and
the product lines. The result of the models also play important roles in management of risk and
performance measurement processes of financial institutions including performance-based
compensation, customer profitability analysis, risk-based pricing, portfolio management and capital
structure decisions. According to Robinson (2001), for the better internal risk management practice and
potentially to use in the supervisory oversight of microfinance institutions credit risk modeling is so
very importance.

2.6 Determinants of Credit Risk

Different academic literatures reveal that, credit risk of banks is influenced by several bank specific and
macro factors. Commonly Profitability, Liquidity, Bank size, deposit rate, operating inefficiency, credit
growth and capital adequacy are some of bank specific factors that were mostly used in a study related
to credit risk determinant whereas GDP, interest rate, unemployment rate, inflation, exchange rate and
money supply are some of widely employed macroeconomic determinants of credit risk. However,
Therefore for this study liquidity, Credit growth, profitability, bank size, inflation rate and capital of
bank were take as determinants of credit risk in commercial banks of Ethiopia.
Bank specific factors
a. Return on Asset (ROA)
It is expressed as the ratio of net income of the year to total asset of the banks and it shows what
amount the return of the shareholder earn from their investment. It represents the rate of return
generated by the investors from their asset and considered as one of the proxy to measure profitability
/performance of bank. In different literature ROA is considered as proxy for profitability of banks and

13
 the relation is either positive or negative. The negative relationship supports the fact that a bank
which has strong profitability has less incentive to produce income and as a result less forced to take
on in risky activities such as giving way risky loans. In reverse, inefficient banks are forced to grant
credits considered risky and subsequently achieve high levels of impaired loan. In the other case,
studies such as Garciya & Robles (2008) revealed that high levels of profitability are achieved
through absorbing a high risk in the future. And their argument is that the profit maximization policy
is accomplished by high levels of risk. Creation of huge-risk and lower quality loans to improve
reported short term financial performance may lead the bank in to losing long term profitability.
Therefore, the return on asset will be positively correlated to credit risk.

b. Liquidity of bank
Various literature elsewhere in the world argue that at least theoretically there is a relationship between
credit risk and liquidity risk in banking business and a bank become succeeds if the bank properly
manage those most significant types of risk. The relationship between credit risk and liquidity obvious
however the debate comes in does credit risk affect liquidity or liquidity affect credit risk. Various
literatures opined that credit risk has an effect on liquidity of financial institution especially deposit
taking financial institution like banks and the premises is that when the credit risk of bank become
higher ( there is higher default on loan) there will be shortage of fund to pay for depositor which is
called liquidity shortage. Hence credit risk should aggravate the problem of the liquidity problem.

On the other hand liquidity of bank create problem of credit risk in such a way the when the bank
highly liquid then intends to rationed higher credit for borrowers and the higher the loan granted the
less likely to repay on time or totally not collected which is termed as borrower default that is credit
risk and the reverse is true if the bank is not as such liquid, less amount of credit granted or may not
credit granted. Therefore liquidity have an impact on credit risk of bank and the relation is seems
positive.
c. Size of Bank
It is clear that there is relationship between bank size and credit risk of banks. However, there are
different arguments regarding relation between bank size and credit risk, in one case bank size and
credit risk has inverse relationship and stated that if the bank have huge asset it has an ability to deal
with sound risk management through portfolio of asset diversification and advantage of economies of

14
scale. On the other hand banks credit risk and size of asset have positive relationship and the reason is
that when the bank becomes huge it will have more branches through expansion and enforce the bank
to pay attractive deposit rate that expose the bank for credit risk and it is positively correlated.
D. Credit growth of banks
Credit growth which is sometimes called loan growth that shows the expansion of credit by banks. It is
clear that there is a possibility of non-repayment of the loan as the level of credit growth becomes
increase. Conceptually, increment of unit of credit growth is not without bearing a risk that is creation
of one more unit of credit is only possible by taking a risk. Hence, the banks become exposed for risk
when the bank needs to extend credit level. Several literatures revels that credit growth has an effect on
credit risk of banks and the relation is either positive or negative.
E, Bank capital
Capital for business organization especially for bank is so crucial for starting up the business by
financing their investments and reduces the prospect of bankruptcy. The capital of a firm can compute
by subtracting the total liability from total asset of the firm. It is a personal fund which is available to
maintain the bank's business and serve as a cushion in case of unfavorable situation. Capital of firm
indicates the ability of the firm that liability could be privileged. Capital of bank is the level of capital
required by the banks to enable them bear up the risks such as credit, operational and market risks they
are exposed to in order to absorb the potential loses and protect the bank's debtors (Tesfa, 2016).

It is essential for maintaining soundness of the banking system since it acts as a cushion against bank
run or uncertainties (Keovongvichith, 2012). The capital amount of a firm is a measure of the overall
financial strength of a bank. Although the relationship between capital of the bank and bank risk level
is vague. In one case, some scholars opined that capital of the firm reduces the risk level which mean
banks that have high levels of risk will try to increase their capital in order to avoid being penalized and
the others think that capital increase the risk level through that banks with high capital level will engage
in more risky activities.
Macroeconomic factors
Banks can play a key role in the economic activity of a given country through providing of various
financial services. Macro and micro economic variable affect the activities of bank in a given country
and some of the macro variables are GDP, Interest rate, Inflation and unemployment rate. In this study
inflation only as macroeconomic reviewed in connection with credit risk of banking business.

15
 a. Inflation
Different literature indicates that inflation has an impact on credit risk and inflation may have an
important for banks in their capacity of financial intermediation having adjusted for anticipated
inflation, and can bear massive default risk depending on the fluctuation of inflation between the
anticipated and actual inflation rates on their fixed instruments (Glogowski., 2004). The rising in
inflation lead to an increase in credit risk. Several studies such as Mileris (2012), (Kochetkov, 2012),
(Derbali, 2011), Renou, (2011), (Hess et al. 2008), (Beck et al., 2013),found inflation rate as a
significant variable explaining credit risk and found that an increase in inflation rate had positive
relationship to credit risk. In contrast to the above studies, the study conducted by Skarica (2013),
(Fofack, 2005), found as inflation had negative and significant impact on credit risk.

2.7. Review of related empirical studies

2.7.1 Determinants of banks credit risk-empirical studies
A number of studies have been conducted on the relationship between credit risk and liquidity at
international level such as; (Abdullah et.al 2012) Conducted research using Johansen’s co-integration
test to assess the long-term relationship between Credit risk and bank specific factors. Researchers
found that Bank size had a positive and significant relationship with credit risk in domestic banks.
Liquid assets and credit risk had negative and significant in foreign banks. A study made by (Ganic,
2012), on bank specific determinants of credit risk in the banking sector of Bosnia and Herzegovina
using the panel regression model and found that inefficiency and credit growth had a significant
negative influence on credit risk while profitability and deposit rate had significant positive impact on
credit risk. However, capital adequacy, liquidity, market power and reserve ratio had an insignificant
impact on credit risk.

Amir & Fatemeh, (2014), studied on investigate the relationship between liquidity risk and credit risk
of banks. In the study, liquidity risk taken as dependent variable and credit risk as the independent
variable and storage of losses from loans and past due loans, total loans and credits to total assets ratio
of banks, capital adequacy ratio and bank size as control variables are considered. The statistical
population of this study is listed banks on Tehran Stock Exchange during the period 2008 to 2013 and
the final sample size is 20 banks. In the study, Eviews software and panel data are used with fixed

16
effects and obtained results of data analysis showed there is a significant and inverse impact between
liquidity risk and credit risk.

A study made by (Ahmad & Ariff, 2007) in multi-country study of bank credit risk determinants in
Malaysia, Mexico and Japan, capital adequacy ratio has significant and positively related to the credit
risk of banks in the study period and firms. The researchers reveal the reason for their finding that
banks to increase their capital as a cushion to take up potential losses that might arise from an increase
in credit risk.

Murage, (2016), conducted a study on the effect of credit risk on corporate liquidity of deposit taking
microfinance institutions in Kenya, the population of the study was 9 DTMs in Kenya and use
secondary source of data for the period 2011 and 2013.the result of the regression revealed that credit
risk has strong and significant effect on corporate liquidity of DTMs. Harvey & Roper, (2014)
conducted a study on effects of credit risk on corporate liquidity among commercial banks in
Netherlands. The data for the study was collected from a sample of 65 banks for a period of five years
ranging from 2008-2012. According to the results of the study there is an inverse relationship between
the credit risk and corporate liquidity among commercial banks

Bjorn & Rauch, (2014) Conducted on the relationship between liquidity risk and credit risk in banks by
took sample of all US commercial Banks during the period of 1998- 2010 and the finding revealed that
both risk categories do not have economically meaning full reciprocal relationship. A study made by
Younes & Nabila, (2011), studied to identify the factors influencing bank credit risk in Tunisia banks
and the sample period’s was from 1995 - 2008 by taking ten commercial banks. The study considered
both macroeconomic factors and microeconomic variables that have an influence on credit risk. The
result of the study indicates that ownership structure, prudential regulation of capital, profitability and
macroeconomic indicators are main determinants of bank credit risk in Tunisia.

A study conducted by (Castro, 2013), to examine the macroeconomic determinants of the credit risk in
the banking system: The case of the GIPSI to analyze the link between the macroeconomic
developments and the banking credit risk in a particular group of countries – Greece, Ireland, Portugal,
Spain and Italy abbreviated in (GIPSI) by Employing dynamic panel data approaches to these five

17
countries over the period 1997 quarter one -2011 quarter three. The researcher concludes that credit risk
is significantly influenced by the macroeconomic variables and a negative relationship that is the credit
risk of bank increases when GDP growth becomes decrease and has positive relationship with
unemployment rate, interest rate, and credit growth.

Washington, (2014) made a study to investigate the Effects of macroeconomic variables on credit risk
in commercial banks of Kenya. The researcher used was nonperforming loans as dependent variable
under investigation and macroeconomic variables like: Lending interest rates GDP per capita growth
rate, Exchange rate between the US dollar and the Kenyan Shilling, Inflation rate and Domestic credit
to the private sector by the Commercial Banks as independent variables. The study covered from 1990
up to 2013 and secondary data was used. The result of the study shows that only GDP have a negative
and significant impact in the short run and other macro variable have significant effect in the long run
on credit risk of bank. In further the study revealed that exchange rate, domestic credit to private sector
granted by commercial banks and inflation has negative relationship with credit risk. On the other hand
lending interest rate has positive effect on credit risk of banks.

Another studies made by (Zmijewski, 1984) on methodological issues related to the estimation of
financial distress prediction and the probability of default is positively correlated with the liquidity.
Similarly study made by (Hillegeist et al.2004) revealed a positive relation between liquidity and credit
risk. In other case probability of default and liquidity negatively correlated in (Shumway, 2001) and
(Ohlson, 1980)
Hertrich, (2015) made a study on the title “Does credit risk impact liquidity risk”? Evidence from
credit default swap markets and the purpose of the study was building light on the dynamic interactions
between credit and liquidity risk in the credit default swap market by talking samples of German and
Swiss companies. The result of the study which is also tested by Granger causality test and revealed
that opposite to the common belief that illiquidity leads to credit risk decline in financial markets,
negative credit shock typically leads to a subsequent liquidity shortage in the credit default swap
market
In the Ethiopian context, some studies have been conducted on credit risks determinants and their
relationships such as (Tehulu & Olana, 2014) conducted a study to see the relationship between credit
risk and bank specific determinants in Ethiopia.

18
The quantitative research approach was adopted for the study. A panel data of 10 commercial banks
both state-owned and private owned for the period 2007- 2011 has been analyzed using random effects
GLS regression. The study revealed that credit growth and banks size have negative impact on credit
risk. While operating inefficiency have positive impact on credit risk. However, capital adequacy and
bank liquidity has no strong impact on the credit risk.

Asfaw & Veni, (2015), made as study on Determinants of Credit Risk in Ethiopian Private Commercial
Banks. The study analyzed the link between the bank specific factors and credit risk using a panel data
covered the period of 2006-2012. Three Panel data estimation method, pooled OLS regression, fixed
effect and random effect model, were used for extracting good result and F-test ascertained the
appropriateness of Pooled OLS regression model. The result revealed that the credit growth and return
on equity had statistically significant and negative impact on Credit risk indicator of the large Ethiopian
private commercial banks. On the other hand inefficiency, and deposit rate had statistically insignificant
positive influence on the Credit risk.

Jabir & Terye, (2016,) conducted a study to investigate the factors that affect credit risk of Ethiopian
commercial banks and the study covers a time period from 2003 - 2009.The researchers used both
macroeconomic and bank specific credit risk determinants to investigate by using fixed effect panel
data model. The result of the study shows that leverage, operating inefficiency, loan growth, ownership
and loan to deposit ratio are significant determinants of credit risk of Ethiopian commercial banks in
the test period.

2.7.2 Empirical Studies Related to the Study Variables

Credit Risk
As empirical studies show credit risk sometimes called asset quality of banking industry affected by
different factors which are macro and micro factors. In similar to this, studies conducted on the area of
credit risk used different measurement for credit risk. It is commonly known that credit risk measured
by the NPL which is non-performing loan divided by the total loan however as NPL is confidential and
not disclosed for third party different proxy measurement is used by scholars. In this study loan loss for
provision is used to measure the credit risk of the sampling banking industry in Ethiopia.

19
 I. Liquidity and Credit risk
As Empirical studies revealed the relationship between credit risk and liquidity risk is not consistent.
Studies like (Harvey & Roper, 2014), (Amir & Fatemeh, 2014), (Philip, 2012), (Abdullah et.al 2012)
and (Ganic, 2012), (Jabir & Terye, 2016), found that there is significant and an inverse relationship
between credit risk and liquidity of banks. Rach et al (2013) conducted a study on the relationship
between liquidity risk and credit risk of bank and the result shows that there is a significant and direct
relationship between credit risk and liquidity risk of banks. Similarly a study conducted by (Murage,
2016) on the title effects of credit risk on corporate liquidity found a direct and strong relationship
between credit risk and corporate liquidity in deposit taking micro finance institution in Kenya. In other
case (Bjorn & Rauch, 2014), found that liquidity risk has insignificance relationship with credit risk in
banking business. (Tehulu & Olana, 2014) Studied bank specific determinants of credit risk and found
that liquidity has a negative but not strong impact on the credit risk.

II. Bank size and Credit risk

Studies such as: (Saunders, 1999), (Chen et al. 1998), (Cebenoyan et al. 1999), (Megginson 2005),
(Salas & Saurina, (2002); (Thiagarajan, 2011), (Das & Ghosh, 2007), (Tehulu & Olana, 2014) found
that a negative relation between of bank size and credit risk. On the other hand Studies like: (Abdullah
et.al 2012), (Jabir & Terye, 2016), (Ziribi & Younes, 2011) found a positive relationship between size
of bank and credit risks.

III. Profitability and credit risk

Various studies showed that profitability have a relationship with credit risk of the firm. Although, the
result of the studies show inconsistent result regarding to the direction and magnitude of the
relationship. Studies like: (Ziribi & Younes, 2011), (Glogowski., 2004), (Messai et.al. (2013) study on
determinants of credit risk found negative relationship between credit risk and profitability measured
by ROE. (Belayneh, 2011) studied on determinants of CBE profitability measured by ROA and found
that profitability has negative relationship with credit risk of the banks. In addition (Tefera, 2011),
(Mekash 2011), (Habtamu, 2012), (Asfaw & Veni, 2015), (Misker, 2015), (Tesfa, 2016) found that
credit risk is negatively correlated with profitability.

20
IV. Loan growth and Credit risk
Studies such as, (Das & Ghosh, 2007), (Jimenez & Saurina ,2006), (Thiagarajan, 2011), (Ahmad,
2013), (Hess et.al.2009), (Jabir & Terye, 2016,) found a positive and significant relationship between
credit risk and credit growth. However (Mehmed, 2014,), (Ganic, 2012), (Tehulu & Olana, 2014),
(Asfaw & Veni, 2015) found a negative relationship between credit risk and credit growth.
V. Capital and credit risk
Regarding to the relationship between credit risk and capital of banks Berger & DeYoung, 1997
studied the USA banks and resulted that thinly capitalized banks take increased portfolio risk, which
results in higher levels of problem loans in the future. Hence the lower the capital of the firm, the
higher loan default of the firm. Further studies like, Hussain & Hassan (2004), Nor & Mohamed
(2007), (Glogowski., 2004), (Tesfa, 2016) and (Tehulu & Olana, 2014) found an inverse and significant
relationship between credit risk and bank capital. In other case Ali & Metin, (2015) and (Ahmad &
Ariff, 2007) found a positive relationship between capital adequacy and bank credit risk.
VI. Inflation and Credit risk
Studies such as Fofack, (2005), Mileris (2012), Kochetkov, (2012), Derbali, (2011), Renou, (2011),
Hess et al. (2008), Beck et al., (2013), Tesfa, (2016), found inflation rate as a significant variable
explaining credit risk and has positive relationship to credit risk. In other case study conducted by Jabir
& Terye, (2016,) found a positive and insignificant relationship between credit risk and inflation rate in
Ethiopian commercial banks. In contrast to the above studies, the study conducted by Skarica (2013),
Washington, (2014), found as inflation had negative and significant impact on credit risk.

2.8. Knowledge Gap of the Study

In line with the above theoretical as well as empirical reviews, liquidity is important to all business
organization specially for banking industry since the function of bank is creation of liquidity both on
the asset and liability side of their balance sheet. At the same time credit risk management also have an
important for any financial institution such as banks. Different study revealed that banks credit risk can
be affected by both bank specific and macroeconomic variables like profitability, inflation rate,
liquidity, bank size, and deposit rate. However this study will be focus only on liquidity as main
variable and some of the bank specific and macroeconomic factors as control variable to see their effect
on credit risk. As the researcher try to mention in the empirical study section various study conducted
on the study area but none of them indicate the relationship between credit risk and liquidity of bank.

21
 There are literatures that support the relation of those two types of risk however the question is does
credit risk affect liquidity or liquidity of bank has an effect on credit risk of banks? A study made by
Murage, (2016) tries to see the effects of credit risk on corporate liquidity. Therefore this study have an
aim of examining the effects of liquidity on credit risk of commercial banks and to the best knowledge
of the researcher this study is the first to see this research area which is open to researchers. Since the
banking industry is in the growth stage with opening of new banks and the absence of active secondary
stock exchange in the country, it is important to notify the relationship between liquidity and credit risk
performance by making empirical investigation to already established banks.

2.9. Conceptual Framework of the Study

The conceptual frame work which describes the relationship between bank liquidity with bank specific
and macroeconomic determinants of credit risk based on the theoretical and empirical perspectives was
formulated as follows:
Fig. 1 Relation between credit risk liquidity and other control variable

Independent Variable

Liquidity=
Commercial Measured by Total loan
Banks credit PLLR/ Total
Risk Loan Total deposit

Controlled Variables
 Profitability
 Credit Growth
 Bank Size
 Inflation
Dependent Variable
 Bank Capital

22
 Chapter Three
3.1 Research Methodologies
This Chapter discusses the methodology that provides a detailed direction about the methods that the
researcher uses to conduct the research. According to Jankowicz (1991), methodology in respect to
research is ‘the analysis of, and rational for, the particular method or methods used in general”. Based
on this definition, methodology of the study is all about the procedures employed by the researcher in
carrying out the study. This chapter explains the research design, source and methods of data collection,
methods of data analysis, model specification and definition of variable and measurement.

3.2 Research Approach and Design

Research design can be defined as the plan that guides the researchers in the course of collecting,
analyzing and interpreting the raw data. According to Creswell, (2009) there are three commonly
approach especially in social science namely qualitative, quantitative and mixed approach. Hence, the
nature of the problem and objective of the study always determine the type of research approach
employed by the researcher. The choice of research approach shows the priority of a researcher about
the dimensions of the research process and methods. The objective of this research was to see the
impact of liquidity on credit risk in commercial banks of Ethiopia.

Given the objective of the study, the researcher was adopted causal or explanatory research
method/design/ analysis as the study involves cause and effect relationship. The purpose of this study is
to determine the relationship between liquidity the independent variables and credit risk (PLLR) as
dependent variable. Hence, the quantitative research approach was used for this study. The study aims
to develop hypothesis and theoretical framework, which can only be examined by quantitative
measures. The other reason for selecting this method is the support of numerous literatures on the
relevant studies, where they employ quantitative methods to investigate their research problems and
verify their hypothesis.

3.3. Data types and Collection Methods

The data for this study was gathered from the audited annual financial report published by the listed
eighteen (18) Banks. The annual data for the all listed banks during 2007 to 2016 was used in order to
see the impact liquidity on credit risk of Ethiopian banking industry. In addition to this, other sources

23
like magazines, brochures, journals, newspapers, websites, etc. were also considered whenever found
necessary. The paper was based on secondary data and the sources of data for the study was balance
sheets and income Statements of banks over 10 years period from 2007 till 2016, which are mainly
extracted from National Bank of Ethiopia, which can provide comprehensive database for all banking
industry. The data for macroeconomic variable: Inflation gathered from World Bank data outlook and
MOFED.

3.3 .1 Study Population and Sampling Technique

According to NBE (2015/16), the number of bank decline from nineteen to eighteen due to the merger
of CBB with CBE. From 18 banks 16 are private and 2 public. The population of this study consists of
all Ethiopian banks, public and private banks such as Commercial Bank of Ethiopia (CBE),
Development bank of Ethiopia( DBE), 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),
Berhan International Bank S.C (BIB), Buna International Bank S.C (BUIB), Oromia International Bank
S.C (OIB), Zemen Bank S.C (ZB), Debub Global Bank S.C(DGB), Abay Bank S.C, Addis
International Bank S.C(AIB), Enat Bank S.C.

Even though, no need to sample from the eighteen banks as they are already few in numbers to collect
information, for meaningful analysis, the researcher used purposive sampling for selecting the sample
frame from the total population. The length of time in this study will be 10 years from 2007-2016 and
the reason for selecting the period is due to the researcher intention to provide the reliable and most
recent result. However, the listed banks may not have the required period information. Due to this
reason, the year service below 10 years is not included in sample frame to make panel data model a
balanced structured.
Therefore, banks that were established after 2007 and started to provide financial statement in the
following fiscal year was not included in this study because this study incorporated only banks that
have financial statements for the year, 2007, and onwards. Therefore, only nine (9) banks namely
(CBE), (CBO), (DB), (AIB), (WB), (UB), (NIB ,( LIB) and (BOA) information were used in the study
to examine the impacts of liquidity on credit risk in commercial banks of Ethiopia. The researcher
believed that the sample size is sufficient to make sound conclusion about the population as far as it
covers above 50% of the total population of the study.

24
The nine commercial banks chosen to examine liquidity factors in Ethiopia are presented as below:

S.No Name of Banks Year of Establishment

1 Commercial bank of Ethiopia Aug 1942
2 Cooperative Bank of Oromiya Feb 2005
3 Awash international Bank S.C Nov 1994
4 Dashen Bank S.C Sept 1995
5 Bank of Abyssinia S.C Feb 1996
6 Wegagen Bank S.C June 1997
7 United Bank S.C Sept 1998
8 Nib International Bank S.C Sept 1999
9 Lion International Bank S.C Oct 2006

Table 3.1: Source: Bank Supervision Directorate, NBE June 30, 2016

3.4. Definition and Measurement of Variables

3.4 .1. Credit Risk (dependent variable)
Credit risk is mostly defined as the potential on which a borrower or counterparty will fail to meet its
obligations in accordance with agreed terms. There are different proxy used to measure the level of
credit risk in several studies, such as the ratio of nonperforming to total loan, loan to total asset, loan
loss reserve to total loans, loan losses to total loan, loan loss provision to total loans and provision for
loan losses to total assets as well as total loan to total deposit were mostly used as a proxy for credit
risk in several credit risk related literatures.

However, the ratio of non-performing loan to total loan was considered better indicator for credit risk in
most literature such as Ahmad & Ariff, (2007), Zuhair & Nasif, (2013), Eftychai & Sofoklis, (2014),
Mehmed, (2014,) and Chaibi & Z.Ftiti, (2015) has been used NPL to measure credit risk level. To
measure credit risk of banks the study used the loan-loss provisions to total loans ratio as proxy
measure since non-performing loan are not available due to confidentiality and this proxy used by

25
different scholars like Belyneh, (2011), Tehulu & Olana, (2014), Tibebu, (2011), Nigist &
Laximikantham, (2015), Jabir & Terye, (2016,).

3.4.2. Liquidity (independent variable)

Liquidity is the ability of any bank to increases fund and meets obligations as they come due, without
incurring unacceptable losses. Commonly, there are two methods to measure the liquidity of banks
which are liquidity ratios and funding gap. The liquidity ratio uses different ratio of balance sheet and it
is very simple to compute. Whereas, funding gap is the difference between inflows and outflows and it
is difficult to measure because it is more data intensive and even there is no standard technique to
forecast inflows and outflows (Crosse & Hempel 1980; Yeager and Seitz 1989; Hempel et al. 1994
;Vodová, 2013). For this study liquidity ratios will be use due to the availability of data to measure
banks liquidity.

As empirical literature shows there are different ways to measure liquidity ratio of banks such as liquid
assets to total assets, liquid assets to deposits and short term financing, loans to total assets and loans to
deposits and short term borrowings. Various researchers like Vodová, (2013), Moore, (2010), Praet,
(2008), Rychtarik, (2009), (Tseganesh, 2012) used total loan to total deposit ratio for measuring
liquidity of bank. Similarly for this study the researcher used total loan to total deposit ratio for
measurement of liquidity positions of commercial banks in the Ethiopia which expressed in the
following form.
LQ= Total loan
Total Deposit

This loan to deposit ratio relates illiquid assets with liabilities of the bank (i.e., relating the loan of the
bank with the deposit). It shows what percentage of the volatile deposit of the bank is attached to in
illiquid loans. Hence, the higher of this ratio the less liquid the bank and the less the ratio the bank is
highly liquid. So the interpretation of this ratio is in reverse, the negative result interpreted as positive
and the positive result interpreted as negative.

26
3.4.3. Controlled Variables
Profitability

As academic literature reveals in most case, return on equity, return on assets and net interest margin
are the proxy to measure profitability of banks. Similarly literatures show that profitability of bank has
a relationship with the asset quality of the bank and it has both positive and negative relationship. In the
positive relationship scholars opined that when the firm becomes profitability the grant for loan
becomes high means the will rationed more loan for creditor and the higher the profitability the higher
the credit risk sine ration of higher loan leads to credit defaults. In other case when the firm becomes
profitable the firm can handle the risk by investing different portfolio of investment. Hence higher
profitability leads to lower risk level for the firm. Return on asset is ratio of net income to total asset
and it indicates the ability of firm to efficiently allocate and manage its resources. How much of net
income generate by investing on the asset. In most of the studies ROA were used as bank’s profitability
indicator such as: Nigist & Laximikantham, (2015), Engdawork, (2014), Belete , (2015) used ROA for
measuring of profitability of banks and similarly in this study ROA will be used for measuring
profitability of Banks.

Bank size
It sometimes called asset size and it measures the size of the bank through its total asset position. Bank
size has become common determinant of credit risk recently and measured by logarithm of total asset.
There are two arguments related to asset size and credit risk, banks which have large size are most
likely have low credit risk that arise from its capacity to establish better credit risk management system
through portfolio of asset diversification. In other case when the bank become huge since it become
expanded through expansion and paying attractive deposit rate which expose the bank for credit risk.
Salas and Saurina, (2002) ,Thiagarajan, (2011) , Younes, (2011), Das & Ghosh, (2007) , used logarithm
of total asset for measuring asset size or bank size and this study also used logarithm of total asset for
measuring bank size.
Credit growth
Credit growth which is sometimes called loan growth that shows the expansion of credit by banks. It is
clear that there is a possibility of non-repayment of the loan as the level of credit growth becomes
increase. Conceptually, increment of unit of credit growth is not without bearing a risk that is creation
of one more unit of credit is only possible by taking a risk. Hence, the banks become exposed for risk

27
when the bank needs to extend credit level. Credit growth can be measured by annual change in total
credit as it was indicated in different studies like Tehulu & Olana, (2014) and in this study credit
growth is measured through change in annual credit growth and in this study the researcher used
change in credit annual credit growth as measurement of loan growth.
Bank Capital
Accordingly previous literatures Bank capital and bank credit risk showed a negative relationship. In
the area of this topic Hussain and Hassan (2004), conducted a study in developing countries and find a
negative relationship between capital of the bank and risk. Similarly Nor and Mohamed (2007) have
tried to presented a comparative study of all factors causal to the credit risks of commercial banks in a
multi-country setting: Japan, France, Australia, and the U.S. represent developed economy banking
systems whereas emerging ones are represented by Mexico, Korea, Malaysia, India, and Thailand.
They have found that the regulatory capital is an important factor influencing the credit risk of any
banking system that offers a range of services. This study also highlights that the credit risk in
emerging economy banks is higher than that in developed economies and that risk is formed by a larger
number of bank specific factors in emerging economies compared to their counterparts in developed
economies. In the context of emerging countries, Glogowski., (2004) have found that the regulation of
capital and risk are negatively related.
Inflation

Inflation can be define as an increase in the general price level of goods and services and is usually
expressed as an annual percentage rate of change. Different literature indicates that inflation has an
impact on credit risk and it is measured by consumer price index. The inflation rate data for this study
will get from the National bank of Ethiopia for the ten year period of 2007 to 2016 and positive
coefficient expected from this study. The following table (Table 3.2) summarizes the list of variables
that will be used for this research study with their operational definitions (measure to represent
variables).

28
Table 3.2: Variable-measurement List
No Dependent variables SYMBOL Measurement Method of Computation
1 Credit risk CRRI Provision for loan loss/total loan
Independent variables
2 Liquidity of firm LQ LQ=total loan/ total deposit

3 Controlled variables
Bank size BSIZE Natural Logarithm of total asset
Credit( loan) growth CRGR Current year Loans minus Previous year
Loans/previous year loan
Bank Capital CB Total capital/ total asset
Profitability/ROA/ ROA Net income / total asset
Inflation INF Annual inflation rate of Ethiopia./
consumer price index ( CPI)

3.5 Model Specification

In the first relationship, credit risk represents as dependent variable and liquidity and other controlled
variables are the independent variables. Namely, liquidity, bank size, profitability, credit growth, bank
capital and inflation. The model is specified on an empirical framework using the variable mention for
the study to investigate the impacts of liquidity on credit risks of banking industry.
The econometric model is specified as follows;
Yit = β 0 + β1LQit + β2CVit+μit………………………………………… ( 1)
Where:
Yit - is dependent variable.
β 0 =is the intercept (constant variable)
B1it=is coefficient of independent variable (liquidity)
LQ= liquidity of firm i - The number of firms and
CVit= controlled Variables t - The number of time period
μit- are the error terms.
The above Model will be written in detail form as follows;
PLLR = β0 + β1LQit + β2BSIZE it+ β3ROAit + β4CBit+ β5CGit+ β6INFit + it (2)

29
 Where:
PLLR–Provision Loan Loss Ratio (credit risk indicator of the firm)
β0 - Constant coefficient
β1 – β6= Regression coefficients for measuring independent variables
LQ = Liquidity of the firm CB= Bank Capital
Size = Bank size CG = Credit growth.
ROE=Return on equity. INF=Inflation
it = Error component showing unobserved factor

3.6 Method of data analysis and Presentation

This study was used descriptive and inferential statistics. Mean, standard deviation, minimum, and
maximum were calculated and presented in tables for the purpose of descriptive analysis. For inferential
statistics, the researcher used STATA software version 12‟s output. Econometric model specification tests
including, Hausman-test were used to select the best suite model among, fixed effect model, and random
effect model. In the same fashion, diagnostic tests for the classical linear regression model assumptions
were carried out. Shapiro Wilk test for normality test, Variance inflation factor test for multicollinarity,
Ramsey RESET test (regression specification error test) for functional form, Modified Wald test for group
wise heteroskedasticity and Wooldridge test for serial correlation (autocorrelation) were used.

30
Chapter Four

4. Data presentation and analysis

This part of the research presents the major findings of the effects of liquidity on credit risk and the other
controlled variables in the context of Ethiopian banking industry and also the analysis and discussion of
the results in association to the theories and previous empirical results discussed and presented in previous
chapter. The stated hypotheses were addressed carefully in this section as to add insight into the different
aspects of determinants of credit risk. The researcher starts first through looking at the firm specific
factors in the study period and investigates by the credit risk of sample banking firm as independent
variables and the liquidity and other controlled variables as a dependent variable. It also presents the
results of panel data regression analysis results, data taken from balance sheets and income statements in
sample Ethiopian banking industry.
The study used credit risk (PLLR) as dependent variable for measuring firms risk level and the
independent variables liquidity, profitability, bank size, credit growth, bank capital and inflation. In order
to achieve the objectives of the study, the following hypotheses were developed.
H1: There is significant relationship between credit growth and credit risk of commercial banks in
Ethiopia.
H2: There is significant relationship between bank size and credit risk of commercial banks in Ethiopia
H3: There is significant relationship between liquidity and credit risk of commercial banks in Ethiopia.
H4: There is significant relationship between inflation and credit risk of commercial banks in Ethiopia.
H5: There is significant relationship between profitability and credit risk of commercial banks in Ethiopia
H6: There is significant relationship between bank capital and credit risk of commercial banks in Ethiopia

In the previous chapters important literatures concerning to the title were reviewed which gives adequate
understanding about the title and also used to identify the knowledge gap on the area of the study. To
achieve the research objectives and to test research hypotheses the research design used by the researcher
for this study also discussed in chapter three. In this chapter data collected were presented and important
correlation and regression analysis findings were discussed. This chapter has five sections. Under the first
section (section 4.1.) descriptive statistics of dependent and independent variables were clearly presented.
Next in section 4.2 Person correlation analyses presented. Under Section 4.3 test Unit root test stationary)
presented. In section 4.4 Econometric Model Estimation procedures and specification tests stated. In

31
section 4.5 Diagnostic test for the classical liner regression model presented. Finally, the results of the
regression analysis were presented under section 4.6 of this chapter.

4.1. Descriptive Statistics of the Data

The descriptive statistics for the dependent and independent variables which show the mean, standard
deviation, minimum and maximum are values of Ethiopian banking business were indicated. The
dependent variables are credit risk measured by provision for loan loss to total loan ratio. The remaining
are the independent variables such as: liquidity, profitability, bank capital, bank size, loan growth and
inflation. The following table (Table 4.1) presents the descriptive statistics of the dependent and
independent variables.

Variable Mean Std. Dev Min Max Observation

CB 13.7553 4.398276 7.752 31.03 90
ROA 2.989778 .9776169 -1.88 5.01 90
LIQ 59.49022 11.39745 25.46 93.40 90
BSIZE 22.91156 1.346259 19.41 26.63 90
CRGR 26.87592 21.67229 -15.94 85.33 90
CRRI 2.559111 2.063005 .06 17.05 90
INF 412.2611 159.109 168.56 647.214 90
Table 4.1 Descriptive statistics
Source: computed from financial statement of Ethiopian Banking Business.
As indicated in the above table (table 4.1), the mean value of the credit risk ratios measured by PLLR for
the sample Ethiopian Banking industry is 2.56 percent (2.559111), which implies that in the sample
Ethiopian Banking industry on average the credit risk is were 2.56 percent. The maximum and minimum
value is 0.1705 and 0.0006 with a standard deviation of 2.06% which has some dispersion from the mean
value. The maximum value of 17.05% indicates the existence of high credit risk in some of the banks.
Liquidity which defined as the ability of firm to pay its obligation when it comes due without incurrence
of unacceptable loss which is funding liquidity has a mean value of liquidity 59.49 percent
(mean=0.5949022) in the sample banking industry which is lower than the international standard for loans
to deposit ratio (i.e. 75% (CBRC, 2012)) that measured by total loan over total deposit and it reflects the

32
sample banking business perform its operation with less significance level of liquidity since it is below the
standard requirement. This show on average for the commercial banks in Ethiopia lower amount of
deposits/liabilities (i.e. when 1 birr is deposited 59.49 cent lend out for creditor which is illiquid loans.
There was high dispersion/deviation/ of the ratio from the mean value among the sample banks which is
shown by the standard deviation of 11.4%. The maximum and minimum value for the sample banking
business is 93.40 and 25.46 percent respectively.

The bank capital of the sample Ethiopian banking industry in average 13.7553 percent (mean=0.137553)
as measured by total capital over total asset of the banks which was above the international standard for
capital ratio (i.e.8% Reporter (13 March 2010). The Maximum value of the capital ratio among the sample
Ethiopian banking industry is 31.03 maximum and the minimum value is 7.752 and it shows a standard
deviation for the sample banks mean value by 4.4 percent and revealing some dispersion towards the mean
among sample commercial banks in Ethiopia. The mean value of the profitability of the sample banking
business measured by ROA is 2.989778 percent (0.02989778), this implies sample Ethiopian Banks on
average earned a net income of 2.99 percent from the asset invested with a maximum and minimum value
of 0.0501 and -0.0188. The standard deviation is 97.7 percent from the average value, which reflects the
presence of higher variation among across the sampled banking industry.

Among the independent variables size of banks was highly dispersed from its mean value (i.e. 22.911) with the
standard deviation of 1.3462. The maximum and minimum values are 26.63 and 19.41 respectively. The
maximum value indicating the CBE and the minimum value was some of privately owned commercial
banks in Ethiopia such as LIB and CBO the next lower value and in case of size CBE overshadow some
banks more than 100%.The average value of credit growth which is measured by current year loan minus
previous year loan divided by previous year loan was 26.87% with a maximum and minimum values of
85.33% and -15.94% respectively. In terms of credit (loan) growth the sample commercial banks in
Ethiopia were highly different with the standard deviation of 21.67%.
The last independent variable is the macroeconomic indicators that can affect banks credit risk over time. The
mean value for the inflation of the country is 412.26 with a maximum inflation was recorded in the year 2016
(647.214) and the minimum was in the year 2007 (i.e. 168.56) which is measured by consumer price index.
The rate of inflation was highly dispersed over the periods under study towards its mean with standard
deviation of 159.109.

33
4.2 Pearson Correlation Matrix:
It shows the likely dependence of one variable upon another variable or the likely relationship among the
variable in the study. Test of correlation a common task to carryout in research that relate with regression
in order to determine whether colliniarity exist among the independent variable used in the study or not.
The reason for test of this problem is to avoid a distorting result on the relationship of dependent variable
and independent variable. The most commonly type of correlation coefficient used is Pearson, sometimes
also called linear or product moment correlation.

According to (Brook, 2008), when we say that y and x variables are correlated, it means that y and x
variables are being treated in completely symmetrical way. Thus, it doesn’t mean that changes in x cause
changes in y or indeed that changes in y cause change in x rather, it is simply stated that there is evidence
for a linear relationship between the two variables, and that movements in the two are on average related
to an extent given by the correlation coefficient. Correlation coefficient always ranges between +1, (which
is perfect positive relationship) to -1 (which perfect negative relationship) and if the correlation coefficient
is 0 (zero) it indicates that there is no linear relationship between variables. The Pearson correlation for
the independent variables; liquidity, size of bank, profitability, credit growth, capital and inflation and the
dependent variable credit risk measured by the provision for loan loss ratio presented in the following
table below, which data taken from balance sheet and income statement of sample banking industry during
the period 2007-2016.
| CRRI CB ROA LIQ BSIZE CRGR INF
-------------+---------------------------------------------------------------
CRRI | 1.0000
CAR | -0.1017 1.0000
ROA | 0.2371 -0.3120 1.0000
LIQ | -0.1947 0.2885 -0.0568 1.0000
BSIZE |0.1657 -0.5859 0.3444 -0.3772 1.0000
CRGR | -0.2122 0.3157 -0.2052 0.2404 -0.2913 1.0000
INF | -0.2653 -0.0694 0.2300 -0.1094 0.4840 -0.0025 1.0000
Table 4.2: Pearson Correlation matrix for Banking Industry
*Source: Financial statement of sampled Ethiopian Banking industry and personal computation.

34
As indicated in the Pearson correlation matrix in the above table, credit risk was negatively correlated with
bank capital, liquidity, credit growth and inflation, with a coefficient estimates of correlation -0.1017, -
0.11947, 0.2122 and -0.2653 respectively whereas profitability of bank and size of bank has a positive
correlation with the banks credit risk with a coefficient of 0.2371 and 0.1657 respectively. As we observe
in the table 4.2, when capital of bank, credit (loan) growth and inflation increases, the credit risk of
Ethiopian banking industry decreases whereas increment in profitability, liquidity and size of bank for the
sampled Ethiopian banking industry result an increase of banks credit risk. From the correlation matrix the
highest correlation is existed between size of bank and inflation rate which is 0.48 according to above
table 4.2.

4.3 Unit Root Test

This study were employed a panel data research which combines the attributes of time series and cross-
sectional. Thus, the researcher initially tested the data and variables to a unit root test since this is
necessary in order to ascertain from the beginning, the researcher is dealing the nature of data and to know
whether or not the result of regression and invariably of the findings can deal to the long run. Particularly
for this test of stationary Augmented Dickey Fuller (ADF) unit root testing was conducted through Stata
version 12, software. Based on the test results, it indicates that all the variables are stationary at 1% and
5% level of significance except last two variables. (See the appendix table 4. 6). As a result, the results
point out that, whatever outcome the researcher gets from the hypotheses testing, the findings can hold in
a long-run perspective.

4.4 Econometric Model Estimation Procedures and Specification Tests

The objective of this study was to see the effects of liquidity on credit risk of Ethiopian banks using panel
data which was collected from annual financial report. Before the panel data studies begin it is important
to make choice among, random effect mode and fixed effect model when estimating econometric models.
Hence, to choose random and fixed effect model is very important as it highly affect conclusions on the
individual coefficients Gujarati (2003). For the model choice of suitable panel data model Hausman test
were carried to make choice among random and fixed effect model. The null hypothesis in this test is
random effect model is appropriate and the alternative is fixed effect model is appropriate.

35
Based on Hausman test we are fail to rejected the null hypothesis, suggesting that random effect model is
better suited compared to the fixed effect (Insert Table 4.1 here).. Based on this test random effect
regression model is appropriate. Thus, the regression results of the random effect regression model were
used for statistical inference and further analysis of the individual coefficients.

4.5. Diagnostic Tests for Classical Linear Regression Model (CLRM) Assumptions
4.5.1. Test of the Mean Value of Error term is Zero (E (ut) = 0)
The first assumption from the classical linear regression model is that the mean values of the error become
zero. According to Brooks, (2014), since a constant term is included in the regression equation, the
assumption zero means vale will never be violated. In this case, a constant term is included in the
regression model of this study to satisfy the assumption of the classical linear regression model that
requires the value of the errors to be zero.

4.5.2. Test of Normality Assumption

Normality assumption is the most fundamental assumption in data analysis that can be a benchmark for
statistical methods. Normality refers to the shape of data distribution for an individual variable. According
to Gujarati , (2003), Normality of the residual shows that, the estimator is unbiased, minimum variance
and consistent commonly named BLUE. Normality can be tested using graphical and statistical tests.
However for this study Shapiro Wilk test was used to test normality distribution of error term with null
hypothesis that residuals are normally distributed. The Result of this test shows that Prob>z = 0.85165, it
is statistically insignificant and fail to reject the null hypothesis which indicates that the residuals are
normally distributed supporting the null hypothesis (Insert Table 4.2 here).

4.5.3. Test of Multicollinarity

The assumption for multicollinarity is that the explanatory variables should not correlate with one to
another. The problem of multicollinarity occurs when the explanatory variables are very highly correlated
with each other. If there is problems of multicollinarity is present and ignored, R square will become high
however the individual coefficients will have high standard errors, as a result that the regression looks
good as a whole, but the individual variables are insignificant (Brook, 2008). The severity of the problem
of multicollinarity across the independent variables is examined in terms of the variance inflation factors
(VIF). According to Gujarati, (2003), variables are considered as highly collinear if the VIF higher than10.

36
In this study, the results of VIF for each explanatory variable included in the regression model are very
low (less than 3), which suggest that there is no serious multicollinarity problem among the explanatory
variable in the study. (Insert Table 4.5 here).
4.5.4 Autocorrelation: (cov(ui , uj ) = 0 for i _= j)
Another assumption of classical linear regression model is serial correlation which is the error term is
uncorrelated each other. Which means the shock that happen in the previous year have no effect on the
current year. Even though autocorrelation is a time serious data problem this study contains both times
serious and cross section data it so important to test this assumption. Actually, ignoring autocorrelation
when it is present has similar consequence to those of ignoring Heteroscedasticity problem when is
actually exist and wrong inferences could be made. For the of test autocorrelation Wooldridge test was
used. The null hypothesis is that there is no autocorrelation and the alternative is there is autocorrelation in
the error term. Based on Wooldridge test shows that Prob>z = 0.6425 and it indicates statistically
insignificant result supporting absence of autocorrelation problem in this study. (Insert Table 4.3 here).
4.5.5 Test of Heteroscedasticity: Var (ui/Xi ) = δ 2i
It states that the variance of the error term is constant in regression results. According to Gujarati , (2003)
when the error term have a constant variance in the regression results which is termed as
homoscedasticity. If the error term has no a constant variance it is called Heteroscedasticity problem. In
this study the researcher used Modified Wald test to identify Heteroscedasticity problem. The null
hypothesis in this test is that the error variances are all equal variance (Homoskedasticty) and the
alternative that the error variance is not constant (Heteroscedasticity).

Based on the test of Heteroscedasticity, the result is revealed P-value is 0.0000 (Insert table 4: 4 here)
which is significance and the researcher fail to reject the null hypothesis that supports there is
heroskedasticty problem in this study. Therefore with this problem the estimation of panel model could
lead to biased statistical result. For handling this Heteroscedasticity problem in the model, studies with
regression of panel data can adjust the standard errors of coefficient of estimates and as a result, it ensures
the validity of statistical result (Hoechle 2007). Therefore, if there is Heteroscedasticity or panel serial
correlation problem, STATA provides different commands and options to estimate standard errors and for
this study Heteroscedasticity problem is controlled by clustered robust test.

37
4.6. Regression Analysis and Discussion

Variable Coefficient Robust Std. Err Z P>|z|

BC .0934934 0.668728 1.40 0.162
ROA .5458829 0.1621384 3.37 0.001**
LIQ -.0265839 .0428023 -0.62 0.535
BSIZE .5322452 .0971553 5.48 0.000 **
CRGR -.0090314 .0062107 -1.45 0.146
INF -.0064241 .0022414 -2.87 0.004 **
Constant -8.081311 4.704937 -1.72 0.086
Number of obs = 90
F( 8, 75) = 2.33
Prob > F = 0.0000
R-sq:within=0.1719
Between=0.6201
Avg = 10.0
Overall = 0.2838
sigma_u | .30410429
sigma_e | 1.7010256
rho | .03097134 (fraction of
variance due to u_i)
Table 4.3, Regression result
Note: ** indicates that significance at 1% level of significance
Source: Financial statements of Ethiopian Banking industry.

CRRIit =-β0 -β1 LIQ it + β2ROAit + β3CBit + β4 BSIZEit - β5 CRGRit -β6INF+ εit

CRRI= -8.081311-.0265839LIQ+.5458829ROA+0934934CB + .5322452BSIZE -.0090314CRGR -.0064241INF

(4.704937) (0.0428023) (0.1621384) (0.0668728) (0.971553) (0.0052107) (0.0022414)

The above table presents the random effect model regression results of credit risk and its determinants.
According to Gujarati, (2004), the coefficient of determination (R2) measures the proportion of the
variation in the dependent variable that jointly explained by the independent variables. In this study the
coefficient of variation (R2 ) in the sample banking industry disclose about the joint significance of the
independent variables (liquidity, profitability, capital , size of bank, loan growth and inflation to explain

38
the dependent variable that is credit risk in the regression model. From the regression model the
coefficient of variation (R2) value is 62 % which means that about 62% of the variation in credit risk of
sampling banks is explained by liquidity, profitability, capital, loan growth, size of banks and inflation
jointly.

However, the remaining 38% variations in credit risk of Ethiopian commercial banks are explained by
other explanatory variables which are not included in the model. In addition, the F-statistics of the random
regression model indicates the joint significance of the independent variables: liquidity, profitability, Bank
capital, loan growth, size of bank and, inflation to explain the dependent variable (credit risk) in the
population, with the null hypothesis of there is no statistical relationships between dependent and
independent variables, against the there is a statistical relationship between dependent and independent
variables. 0.0000 probabilities of the F- statistics that indicates that all explanatory variables included in
the model jointly, significantly explain the variation of credit risk of Ethiopian commercial banks.

The panel random effect estimation result above also indicates the intercept value of about -8.081, means
that if the value of all explanatory variables were fixed at zero, the mean credit risk of Ethiopian
commercial banks would be about -8.08. Hence, the following discussions present the relationship
between dependent variable of credit risk and all the explanatory variables in the model.

Liquidity and credit risk

According to the table, the regression result for liquidity reveals negative result and suggesting that as
liquidity increase, credit risk will increase. Since the study employed inverse proxy for liquidity.
However, the regression result was statistically insignificant even at 10% of level significance. The
negative sign of the regression result implies that as liquidity increase, banks will rush to make loan to
dispose of the excess liquidity and this lead to more credit default. This result was consistent with the
prior empirical studies such as Bjorn & Rauch, (2014), Tehulu & Olana, (2014) all found that negative
and insignificant relationship between liquidity and credit risk. The reason for insignificance relationship
might be in the study period the sample banks have no as such excess cash ( has no excess liquid asset)
and this resulted no impact on credit sik. This is suggesting that even if increase in liquidity resulted
increase in credit rsik, there is no as such high credit rsik in the study period for sample banking firms.

The study also the share the finding of Harvey & Roper, (2014), Amir & Fatemeh, (2014) , Philip, (2012)
,Abdullah et.al (2012), Ganic, (2012) , Jabir & Terye, (2016,) in all of them found a negative relation ship

39
between credit risk and bank liquidty. However this result contradicts the finding of Inyang et.al (2014),
Diamond & Rajan, (2005), Murage, (2016) and Rach et al, (2013) who founds a positive relationship
beteween credit risk and bank liquidity. .Consequently, the researcher reject the alternative hypothesis of
there is a significant relationship between liquidity and credit risk of banks and concludes that the
relationship between liquidity and credit risk is insignificance.

Profitability and Credit Risk

The regression result indicates that profitability of firm measured by ROA has a positive and significant
relationship with credit risk of banks. Hence, the result suggested that when profitability of banks
increases, credit risk will also increase. The positive sign of this regression intends toward to the
perception of highly profitable firms exposed to credit risk as they grant more and more credit. In this case
the higher the profitability of the firm the higher the credit risk of the firm. Studies such as Garciya &
Robles (2008) revealed that higher levels of profitability are achieved through absorbing a high risk in the
future and their argument is that the profit maximization policy is accomplished by high levels of risk.

This result of regression is supported the finding of Rajan, (2005), Garciya & Robles (2008) , Younes &
Nabila, (2011) who found a positive relationship between credit risk and profitability of banks. However
this result contradicts the finding of Ganic, (2012) , Asfaw & Veni, (2015) and Tesfa, (2016) who founds
a positive and strong relationship beteween credit risk and bank liquidity. .Consequently, the researcher
accepts the alternative hypothesis of there is significant relationship between profitability and credit risk
of banks.

Bank Capital and Credit risk

According to the table the regression result, the coefficient of Capital of bank indicates positive suggesting
that when bank capital increase, credit risk of firms will increase. However the result was statistically
insignificant even at 90% confidence interval. The positive sign of the regression implies that as bank
capital increase the firm credit risk also increase. The perception here is that when firms capital increase
the risk level through that bank with high capital level will take on more risky activities. Hence the higher
the capital of the bank the higher for the credit risk due to involvement in risky investment by the bank.
The reason for insignificance relationship might be due to implementing sound credit risk management by
the firms in the study period. As bank become strong enough in capital they can implement better credit
risk management like investing in portfolio of asset, creation of awareness and consultancy for borrows
this resulted little impact on credit risk of the firm even if the firm may involve in more risky investments.

40
This result was consistent with the finding of Ahmad & Ariff, (2007) and Ali & Metin (2015) both found
that positive and insignificance relationship between bank capital and credit risk of banks.

However this result contradicts the finding of Berger & DeYoung (1997) , Hussain & Hassan (2004) ,
Nor & Mohamed (2007) , Glogowski, (2004) , Tehulu & Olana, (2014) and Tesfa, (2016) who found an
inverse relationship between capital and credit risk of bank and opined that the credit risk of bank
becomes low when the banks capital comes high . Which means capital of the firm reduces the risk level
that is banks which have high levels of risk will try to increase their capital in order to avoid risk.
Consequently, the researcher reject the alternative hypothesis of there is significant relationship between
bank capital and credit risk of banks. This means the relationship between bank capital and credit risk of
the company in the study period and firms insignificance.
Bank Size and Credit risk
The regression result indicated in the above table the relationship between credit risk and size of bank
measured by logarithm of total asset is positive and strong. This means when the size of the bank increase
the credit risk of firm will also increase other variable in the study assumed constant. This indicates that
size of bank is one of the important determinants of credit risk. The implied perception is that when the
firm size is becomes huge it will have more branches through expansion that expose the bank for credit
risk and it is positively correlated. The result of this study is consistent with the study of Rajan & Dhal,
(2003). Abdullah et.al (2012), Jabir & Terye, (2016), Ziribi & Younes, (2011) who found a positive and
significance relationship between size of bank and credit risk of banks.

Although this result is inconsistent with the finding of Saunders, (1999), Chen et al. (1998), Cebenoyan et
al. (1999), Megginson, (2005), Salas & Saurina, (2002), Thiagarajan, (2011), Das & Ghosh, (2007) ,
Tehulu & Olana, (2014) who found a negative relationship between credit risk and size of bank. Based on
these studies, an inverse relationship means that huge banks have better risk management strategies
through diversification in portfolio of asset that typically translate into more superior loan portfolios than
their smaller counterparts. Hence, the researcher fail to reject the alternative hypothesis of there is a
significant relationship between size of bank and credit risk of the sample banking industry for the study
period. This means there is significant relationship between size of bank and credit risk.

41
Credit (Loan) Growth and Credit Risk
According to the table the regression result, the coefficient of credit growth is negative suggesting that as
credit growth increase, credit risk of the bank will decrease. However, the result was statistically
insignificant at 10% level of significance. This is due to the reason that during study period, there is no as
such constant growth in loan (variation in loan growth even in some period there were decreasing in loan)
and this resulted insignificance result on credit risk of sample banks for the study period. The perception
here is that as there is stable growth in loan there is impact in credit risk either in positive or negative. The
negative relationship between loan growth of the sampling banking industry and their credit risk is due to
the reason that the banks may be develop best experience with borrowers how they repay the loan. The
bank may develop strong and unified credit risk culture, developing sound credit risk management system
as well as gaining best experience of dealing with borrowers and building the capacity of solving the
repayment problem. Hence the higher the credit growth of the bank the lower for the credit risk due to
lower of credit risk by controlling using experience and good customer relationship and follow up by the
bank.
This result was consistent with the finding of, Al-Smadi & Ahmad, (2009), Altunbas et al (2007) ,Mehmed
(2014,) Ganic, (2012), Tehulu & Olana, (2014), Asfaw & Veni, (2015) found a negative and insignificance
relationship between credit risk and credit growth. However this result contradicts the finding of Das &
Ghosh, (2007), Jimenez & Saurina ,(2006), Thiagarajan, (2011), Ahmad , (2013), Hess et.al.(2009), Jabir
& Terye, (2016,) found a positive and significant relationship between credit risk and credit growth of
bank and opined that the credit risk of bank becomes high when the banks loan groqth comes high since
additional loan is never exist without incurrance of addional risk Which means loan growth of the firm
increase the risk level of the banks. Consequently, the researcher accepts the null hypothesis of there is
insignificant relationship between credit growth and credit risk of banks.
Inflation Rate and Credit Risk
According to the table the regression result, the coefficient for inflation is negative suggesting that an
increase in inflation results a decrease in credit risk. The result also has statistically significant effect at
1% level of significance. The negative sign of the regression result implies that as inflation increase, it
leads too much money chasing goods and that business person will pass on the burden of inflation to the
consumer hence will always able to service their loan on time. The probable interpretation of this result is that
inflation leads the firm to more profitability as more money chases few goods. Most borrowers are business people
who seem to pass over the cost of inflation to consumers. For example, when fuel prices go up, road transport
players raise fare to consumers of their services. Thus business people retain their ability to repay their loans.

42
Hence higher inflation leads low default probability. Thus business people retain their ability to repay their
loans.
This result was consistent with prior periods studies like Skarica (2013), Warue, (2013), Washington,
(2014), all found that inflation had negative and significant impact on credit risk. However this result
contradicts the finding of Mileris (2012), Kochetkov, (2012), Derbali, (2011), Renou, (2011) who founds
a positive and significan relationship beteween credit risk and inflation. Consequently, the researcher
rejects the null hypothesis of there is insignificant relationship between inflation and credit risk of banks.

Table 4.4 Summary of actual and expected signs of explanatory variables on the dependent variables

Independent Expected relation Actual relation Decisions for the null hypothesis
variables with Credit risk with Credit risk

Liquidity Significance Insignificant Accept the null hypothesis

Profitability Significance Significant Failed to reject the alternative hypothesis

Size of Bank Significance Significant Failed to reject the alternative hypothesis

Loan growth Significance Insignificant Accept the null hypothesis

Capital of bank Positive & Sig Insignificant Failed to reject hypothesis

Inflation Positive & Sig Significant Failed to reject the null hypothesis

43
Chapter Five

5. Conclusions, Recommendation and implications of the study

This chapter of the thesis includes presentation of the major findings, discussion and empirical
conclusions derived from the research study. Finally this section finishes by providing recommendation
for future research in this area.

5.1. Conclusions of the study

The purpose of this study was to see the impacts of liquidity on credit risk of commercial banks of
Ethiopian. As empirical evidence of previous study indicates, credit risk of firms can affected by both
firm specific and macro variables such as liquidity ,profitability, size of banks, credit growth, capital of
banks and inflation respectively. To achieve the objectives of the study the researcher used panel data
from nine commercial banks for the periods of 2007-2016.
The data for the study obtained from national bank of Ethiopia and from their audited annual reports for
bank specific factors and from World Bank data outlook for macro variables. All banks included in the
study if they had the ten year audited financial statements. This paper examined empirically the
implication of impact of liquidity on credit risk in Ethiopian banking companies. The results of regression
analysis reveal that firm liquidity, credit growth size of bank, capital of banks, profitability and inflation as
independent variable whereas the PLLR of the firm as dependent variable. The study shows that the
expected sign for is confirmed by actual relation for the model under the study by credit risk of firm
(PLLR) measures in regression model result.

The researcher used Hausman test to select the appropriate panel data estimation technique for this study
and as a result random effect model is appropriate for this .Random effect panel regression model was
used in order to determine the effect of the recognized explanatory variables on PLLR of Ethiopian
commercial banks. Result of the regression analysis indicates that liquidity is not a significant factor that
determines Ethiopian commercial banks credit risk for sample banks for the study periods due to the
reason that no excess liquidity for banks in the study period. The negative regression result implies that
when the liquidity situation of banks increases, they will rush to make loan to dispose the excess liquidity.
This result is consistent with different scholars finding and thus, liquidity measured by total loan to total
deposit is insignificant factor that affects the credit risk of Ethiopian commercial banks in the study
periods. The random estimation model regression result showed that profitability is significant factor of
Ethiopian commercial banks credit risk and it has positive effect on credit risk. This result of the study

44
shows that commercial banks of Ethiopian were more vulnerable for credit risk when they become more
profitable as their profit lead to expansion and more credit. Therefore, profitability is a significant factor
that affects the credit risk of Ethiopian commercial banks in the study periods.

The regression result indicates that credit growth of commercial banks is negatively affects the credit risk
off Ethiopian commercial banks for the study periods. The negative and insignificant result indicates that a
bank which has high credit growth measured by loan in current period minus loan in last period divided by
loan in last period are going to absorb low credit risk than banks with low credit growth, this is due to the
reason that experience of bank and good consulting habit with customer reduce problem of loan
repayment and result low problematic loan. Therefore, credit growth is one factor that affects the credit
risk of Ethiopian commercial banks in the study periods even if it is insignificant one.

The random effect regression results of the study reveal that size of banks measured by natural logarithms
of total asset was a significant factor that affects the credit risk of Ethiopian commercial banks for the
study period and it has a positive effect on to PLLR of sample banks. The result shows that when banks
are large in size they are going to more expose to credit risk than banks with low in size, because the
larger the firm have an advantage to expand the branch and mobilize more deposit than low size banks
which have low deposit mobilization and low credit risk. Hence, bank’s size is a significant factor that
affects the credit risk of Ethiopian commercial banks in the study periods.

The regression result reveal also that bank capital which measured by total capital to total asset is
positively affect the credit risk of Ethiopian commercial banks for the study period. The perception here is
that when firms capital increase the risk level through that bank with high capital level will take on more
risky activities. Hence the higher the capital of the bank the higher for the credit risk due to involvement
in risky investment by the bank. However, the result is insignificance due to better credit risk management
practice as their capital increase. Therefore, bank’s capital is one factor that affects Ethiopian commercial
banks credit risk for the study periods.

The random effect estimation model regression results indicated that inflation is one of the significant
factors that affect credit risk of commercial banks of Ethiopia for the study period. The result of this study
was consistent with the Skarica (2013), Warue, (2013) ,Washington (2014), all found that inflation had
negative and significant impact on credit risk. This result of the study shows that if there is high inflation
which is measured as consumer price index Ethiopian commercial banks are going to absorb low credit
risk than in low inflation situation. This is due to the fact that inflation will lead to too much money

45
chasing goods and that business people will always pass on the burden of inflation to the consumer hence
will always be able to service their loans. Therefore, inflation of the country is one of the significant
factors that affect the credit risk of Ethiopian commercial banks for the study periods.

5.2 Recommendation
Based on the results of this study which was discussed in the previous chapters, the following
recommendations were stated. Besides of this study, various empirical studies on the relationship between
liquidity and credit risk revealed a negative relationship between liquidity and credit risk of the banking
business. The result proves that with the increase in liquidity positively affects the credit risk Ethiopian
banking industry. So, the researcher recommends that bank should have adequate level of liquid asset for
minimizing their credit risk. Even if the magnitude was insignificance, the negative relationship suggests
that more efforts should be taken regarding adequate level of liquid asset on hand to reduce credit risk.

The random effect estimation model result, size of banks has a positive and significant impact on credit
risk of sampled Ethiopian banking industry. The positive relationship between size of bank and credit risk
of the bank suggests that bank size is positively correlated to the credit risk because big size bank can
grant more credit and this lead to more credit risk than small size banks.

The positive relationship between profitability and credit risk of the firm suggests that Ethiopian banking
industries may reduce their risk by implementing effective controlling risk management mechanisms like
diversified its operation and follow-up of customer after granting credit for customer. Generally, the
banks (manager of the bank) should devote their time and effort on the significant variables to minimize
their credit risk exposure as the same time to maximize the shareholders wealth.

5.3 Implications of the study

There are three reason for conducting research, Firstly, research going to undertaken to contribute to
existing knowledge through providing extra results to prove or disprove results of earlier studies and add
value to existing knowledge since research is the process of re-investigation of the problems . Secondly,
research is undertaken to propose ways of improvements for perform. Thirdly, it is important to provide
information to policy makers (Creswell, 2009). Hence, the findings of this study have practical,
theoretical, and policy contributions and suggestions.

46
Implications for decision makers
The results of this study provides that an important information to board of directors and managers to
revised their credit risk control mechanism by considering the significant factors affecting the credit risk
of the bank. Mainly, if the board of directors in the bank were interested to minimize the credit risk, in this
case the effect of profitability, size of banks and inflation desires careful attention according to the study
results. The result of this empirical study also suggests that the management team in the bank should make
every effort as much as possible in the significant factors to retain the existing investors and to attract the
new investors as well as to satisfy their shareholders goal of wealth maximization.
Implications for investors
The result of this study also provides important information to existing and potential investors to make
investment decisions in the banks. If investors expect higher return from their investment, they should
focuses on the significant credit risk factors in the banks before making investment decisions. According
to the result of the study shareholders can make prediction or decision what will be the credit risk of
banks, since, the result of this study accordingly indicated the significant credit risk determinants of firms.
Implications for further research
This study was not conducted without its limitations. Firstly, for examining the determinants of credit risk,
commercial banks of Ethiopia are the only target populations and only secondary data was used. Thus,
this study recommends that further studies be conducted in the area including both financial and non
financial firms in the country. The use of primary data such as interviews, questioners may provide more
informed data on factors that affect the credit risk of firms. For future researchers in the area this study
recommends that further research be conducted to examine factors affecting credit risk by employing
other model. This study also recommends that further studies be conducted in the area to investigate the
effects of ownership structure (private and public firms) on the credit risk level of financial ad non
financial firms in Ethiopia.

47
Reference.
Abdullah, A. K. ,Khan,A,Q.,&Nazir,N; A comparative studyof credit risk managmnt:A case study of domestic and
foreign banks in pakistan. Academic Research international, 371-377

Ahmad, B. (2013). Explanatory Power of Bank Specific Variables as Determinants of Non-Performing Loans:
Evidence from Pakistan Banking Sector. . World Applied Sciences . , Journal , 9 (22).

Ahmad, N., & Ariff, M. (2007). Multi Country Study of Bank Credit Risk Determinants. International Journal of
Banking and Finance, , Vol. 5.

Alemayehu.F. (2016). Determinants of liquidity of CBE. Empirical study on commercial banks in Ethiopia’, MSc
Thesis, Addis Ababa University .

Ali, M. a. (2015). “Determining Impacts on Non-Performing Loan Ratio in Turkey,”. in Journal of Applied Finance &
Banking, , ol. 5, no. 1, pp. 1-11.

Al-Tamimi, H. H., & Al-Mazrooei, F. M. (2009). Banks' risk management: a comparison study of UAE national and
foreign banks. ,. The Journal of Risk Finance , 8 (4), 394 - 409.

Amir, M., & Fatemeh, M. (2014). Investigating the relationship between liquidity risk and credit risk in listed banks
on Stock Exchange ,Tehran. Today Science Journal of Humanity , Volume: 3 Issue: 2 Pages:58-68.

Asfaw, A. H., & Veni, P. (2015). Credit Rsk Managment Tools Practiced In Ethiopian Comercial Banks. International
Journal of Social Science & Interdisciplinary Research , 4 (5).

ASPACHS, O. –N.–T. (2005). Liquidity, Banking Regulation and the Macroeconomy. Evidence on bank liquidity
holdings from a panel of UK-residents banks. Available at:
http://www.bis.org/bcbs/events/rtf05AspachsNierTiesset.pdf.

Basel, ( 2009). on banking supervision announces enhancements to the Basel II capital framework.www.bis.org.

BCBS. (2001). Risk Management Practices and Regulatory Capital: Cross-Sectional Comparison (available at
www.bis.org) . Basel Committee on Banking Supervision .

Belete.F. (2015). Factors Affecting Liquidity of Selected Commercial Banks in Ethiopia. Empirical study on
commercial banks in Ethiopia’, MSc Thesis, Addis Ababa University .

Bessis, J. (2009). Risk Management in Banking. Chichester: John Wiley & Sons.

BIS. (2008). Instituets of Bankining Supervision.

Bjorn, I., & Rauch, C. (2014). The relationship between liquidity risk and credit risk in banks. Journal of Banking &
Finance , 40 (2014) 242–256.

48
Brook, C. (2008). Introductory Econometrics for Finance. The ICMA Centre, Henley Business School, University of
Reading .

Castro, V. (2013). Macroeconomic determinants of the credit risk in the banking system:The case of the GIPSI. N.º
12.

CBRC. (2012). China Banking Regulatory Commission/CBRC , „Stricter bank rules will take effect in early available
at http://www.chinadaily.com.cn/business/2011-05/04/content_12440802.htm.

Chen, K., & Pan, C. (2012). An Empirical Study of Credit Risk Efficiency of Banking Industry in Taiwan,, . Web Journal
of Chinese Management Review , 15(1), 1-16.

Coyle, B. (2000). Framework for Credit Risk Management, Chartered Institute of Bankers, United Kingdom.

Creswell, J. W. (2009). Research design: Qualitative, quantitative, and mixed methods approaches, , Sage
Publications, New York. 3rd ed.

Crowe, K. (2009). ''Liquidity risk management-more important than ever",. Harland Financial Solutions, , p. 3.

Das, M., & Ghosh, D. (2007). Determinants of Credit Risk in Indian Stateowned Banks: An EmpiricalInvestigation.
MRP , paper , no. 17301.

Derbali. (2011). Determinants of Banking Profitability Before and During the Financial Crisis of 2007 The Case of
Tunisian Banks. Tunisian Banks.,. 3, 1256-1269.

Diamond .Dw and Rajan, R. (2001). ‘Liquidity risk, liquidity creation, and financial fragility: atheory of banking’,.
Journal of Political Economy , Vol. 109, No. 2, pp.287-327.

Dybvig, D., & Diamond. (1983). , ‘Bank runs, deposit insurance, and liquidity’,. Journal of Political Economy, , Vol.
105, No. 91, pp. 401-419.

Engdawork, T. (2014). Impact of Credit Risk on the performance of Commercial Bank of Ethiopia. Thesis Paper .

Fofack, H. (2005). Non-performing loans in sub-Saharan Africa: Causal Analysis and MacroeconomicImplications. , .
World Bank Policy Research Working Paper , No.3769.21.

Ganic, M. (2012). Bank Specific Determinants of Credit Risk An Empirical Study on the Banking Sector of Bosnia and
Herzegovina. International Journal of Economic Practices and Theories , , 4 (4).

Gieseche, K. (2004). “Credit risk modeling and valuation: an introduction”, Cornell University, Credit Risk: Models
and Management, . Vol. 2.

Glogowski. (2004). Macroeconomic determinants of Polish banks’ loan losses,. National Bank of Poland. IMF
Working paper , 53.

Gujarati. (2004). Basic Econometrics. fourth edition.

Gujarati, D. (2003). Basic Econometrics. United States Military Academy, West Point , 4th edition.

49
Harford, M. a. (2005). . Corporate Governance and Firm Cash Holdings.,. Unpublished working paperUniversity of
Washington.

Harvey, R. L., & Roper, H. (2014). The effect of Credit Risk on Corporate liquidity. Journal of Financial Economics, ,
74, 3-30.

Hertrich, M. (2015). Does Credit Risk Impact Liquidity Risk? International Journal of Applied Economics , 1-46.

Hillegeist, S. E. (2004). Assessing the probability of bankruptcy. . Review of studies , 9:5-34.

Hussain, H. A.-A. (2012). Risk management practices of conventional and Islamic banks in Bahrain. The Journal of
Risk Finance , , 13, 215-239.

Ikhide, S. (2000). . “Efficiency of Commercial Banks in Namibia.” . BON Occasional , Paper No. 4.

Inyang, N. E. (et.al 2014). The Relationship and effect of Credit and Liquidity Risk on Bank DefaultRisk among
Deposit Money Banks in Nigeria. Research Journal of Finance and Accounting www.iiste.org , Vol.5, No.16,.

Jabir, T. T., & Terye, N. D. (2016,). Credit Risk Analysis of Ethiopian Banks: A Fixed Effect Panel Data Model. British
Journal of Applied Science & Technology, Article no.BJAST.23563 .

Jappelli, T. &. (2002). Information Sharing, Lending and Defaults: Cross- Country Evidence,. Journal of Banking &
Finance, . , 26, 17-45.

Kargi, H. (2011). Credit Risk and the Performance of Nigerian Banks, AhmaduBello University, Zaria.

Kochetkov. (2012). Modern Model of Interconnection of Inflation and Unemployment in Latvia. Inzinerine
Ekonomika-Engineering Economics , 2(23), 117-124.

Mbole, W. (2004). The relationship between credit risk analysis and the level of non-performing loans,.
Unpublished MBA Project, University of Nairobi .

Mehmed, G. (2014,). “Bank Specific Determinants of Credit Risk - An Empirical Study on the Banking Sector of
Bosnia and Herzegovina,. International Journal of Economic Practices and Theories , ol. 4, No. 4 pp. 428 – 436.

Mekibib.S. (2016). Determinants of Liquidity in Commercial Banks of Ethiopia: The Case of Selected Private Banks.
un published research, .

Misker, B. (2015). The Impacts of Credit Risk on performance of Banks in Ethiopia. Thesis Paper .

Moore, W. (2010). . How do financial crises affect commercial bank liquidity? Evidence from Latin America and the
Caribbean. . Munich: Munich Personal RePEc Archive. MPRA Paper , 2010-21473.

Murage, C. W. (2016). The Effect of Credit Risk on Corporate Liquidity of Deposit Taking Microfinance of kenya.
International Journal of Business and Social Science , Vol. 7, No. 4;.

Myers, & Majluf. (2004). Myers, C. & Ma Corporate financing and investment decisions when firms have
information thoseinvestors do not have. Journal of Financial Economics, , 13, 187- 21.

50
NBE. (2010). RISK MANAGEMENT GUIDLEIES for MICROFINANCE INSTITITTIONS..

Ngugi, W. (2012). The impact of credit information sharing on credit risk for commercial banks in Kenya,.
Unpublished MBA Project, University of Nairobi .

Nigist, & Laximikantham. (2015). Determinants of Banks Liquidity: Empirical Evidence on. Journal of Economics and
Sustainable Development , Vol.6, No.15,.

Ohlson, J. (1980). Financial ratios and the probabilistic prediction of bankruptcy. . Journal of Accounting Research ,
18:109-31.

Philip, S. (2012). Liquidity Risk and Credit in the Financial Crisis, . Journal of Financial Economics , 101, 297–312.

Praet, H. (2008). Market liquidity and banking liquidity: linkages, vulnerabilities and the role of disclosure. In
Banque de France Financial stability Review. France: Banque de France,. pp. 95-109.

Prakash, R. &. (2013). Macroeconomic Determinants of Credit Risk in Nepalese Banking Industry. Proceedings of
21st International Business Research Conference. Toronto: Ryerson University.

Rajan, R. (2005). The Effect of Credit Market Competition on Lending Relationships,. The Quarterly Journal of
Economics, , 1,1-5.

Rychtarik, Š. (2009). Liquidity Scenario Analysis in the Luxembourg Banking Sector. BCDL. Luxembourg: Banque
Centrale du Luxembourg. Working Paper 2009-41.

Saunders, A. &. (1999). New Approaches to Value at Risk and Other Paradigms (2nd ed.). New York.:. John Wiley &
Sons, Inc.

Shumway, T. (2001). Forecasting bankruptcy more accurately: A simple hazard model. . Journal of Business ,
74:101-24.

Tehulu, T. A., & Olana, D. R. (2014). “Bank- Specific Determinants of Credit Risk: Empirical Evidence from Ethiopian
Banks”,. in Research Journal of Finance and Accounting , Vol.5, No.7, pp. 80 – 85.

Tesfa, G. (2016). Determinants of Ethiopian Commrcial Private Bank's in Ethiopa Asset Quality. Thesis Paper .

Thiagarajan, S. e. (2011). Credit Risk Determinants of Public and Private Sector Banks in India. . European Journal
of Economics, Finance and Administrative Sciences , 34.

Tibebu, T. (2011). CREDIT RISK MANAGEMENT AND PROFITABLITY OF COMMERCIAL BANKS IN ETHIOPIA. Thesis
paper Addis Ababa Univeristy .

Tseganesh.T. (2012). ‘Determinant of banks liquidity and their impact on financial performance:. Empirical study
on commercial banks in Ethiopia’, MSc Thesis, Addis Ababa University.

Vodová. (2013). ‘Determinants of commercial banks liquidity in Hungary’,.

Washington, G. K. (2014). EFFECTS OF MACROECONOMIC VARIABLES ON CREDIT RISK IN THE KENYAN BANKING
SYSTEM. International Journal of Business and Commerce , Vol. 3, No.9: [01-26].

51
Younes. (2011). . The factors influencing bank credit risk: The case of Tunisia. Journal of accounting and taxation ,
Vol. 3(4) PP. 70-78.

Younes, B., & Nabila, Z. (2011). The factors influencing bank credit risk: The case of Tunisia. Journal of Accounting
and Taxation , Vol. 3(4), pp. 70-78,.

Ziribi, N., & Younes, B. (2011). The factors influencing bank credit risk: The case of Tunisia. Journal of accounting
and taxation Vol. 3(4) PP. 70-78.

Zmijewski, M. (1984). Methodological issues related to the estimation of financial distress prediction models.
Journal of Accounting Research , 22-59.

52
Appendix
Table: 4. 1
Hausman Specification Test of Random-Effects against Fixed-Effects
Null: Random effect model is appropriate
Alt: Fixed effect model is appropriate

Source: STATA 12 output

Table 4.2: Shapiro Wilk W Test Normality Test

Shapiro-Wilk W test for normal data

Variable Obs W V z Prob>z

u 90 0.99176 0.623 -1.044 0.85165

Source: STATA 12 output

53
Table 4.3: Autocorrelation Test

Wooldridge test for autocorrelation in panel data

H0: no first order autocorrelation
F( 1, 8) = 0.233
Prob > F = 0.6425
Source: STATA 12 output
Table 4.4: Test for Heteroscedasticity

Modified Wald test for groupwise heteroskedasticity

in fixed effect regression model

H0: sigma(i)^2 = sigma^2 for all i

chi2 (9) = 4809.05

Prob>chi2 = 0.0000

Source: STATA 12 output

Table 4.5, Test for Multicolliniarity

| CRRI CB ROA LIQ BSIZE CRGR INF
-------------+---------------------------------------------------------------
CRRI | 1.0000
CAR | -0.1017 1.0000
ROA | 0.2371 -0.3120 1.0000
LIQ | -0.1947 0.2885 -0.0568 1.0000
BSIZE |0.1657 -0.5859 0.3444 -0.3772 1.0000
CRGR | -0.2122 0.3157 -0.2052 0.2404 -0.2913 1.0000
INF | -0.2653 -0.0694 0.2300 -0.1094 0.4840 -0.0025 1.0000

54
 Variable VIF 1/VIF

BSIZE 2.41 0.414110

CAR 1.76 0.566736
INF 1.49 0.672049
LIQ 1.21 0.823961
ROA 1.21 0.829195
CRGR 1.18 0.844436

Mean VIF 1.55

Table 4.6: Unit Root Test

Output result of unit root test

. xtunitrootllcroa

Levin-Lin-Chu unit-root test for roa

------------------------------------
Ho: Panels contain unit roots Number of panels = 9
Ha: Panels are stationary Number of periods = 10

AR parameter: Common Asymptotics: N/T -> 0

Panel means: Included
Time trend: Not included

ADF regressions: 1 lag

LR variance: Bartlett kernel, 6.00 lags average (chosen by LLC)
Statistic p-value
Unadjusted t -5.6573
Adjusted t* -2.5018 0.0062
. xtunitrootllcliq
Levin-Lin-Chu unit-root test for liq
------------------------------------
Ho: Panels contain unit roots Number of panels = 9
Ha: Panels are stationary Number of periods = 10

AR parameter: Common Asymptotics: N/T -> 0

Panel means: Included
Time trend: Not included
55
ADF regressions: 1 lag
LR variance: Bartlett kernel, 6.00 lags average (chosen by LLC)
Statistic p-value
Unadjusted t -16.3385
Adjusted t* -12.9683 0.0000
. xtunitrootllcbsize
Levin-Lin-Chu unit-root test for bsize
--------------------------------------
Ho: Panels contain unit roots Number of panels = 9
Ha: Panels are stationary Number of periods = 10

AR parameter: Common Asymptotics: N/T -> 0

Panel means: Included
Time trend: Not included

ADF regressions: 1 lag

LR variance: Bartlett kernel, 6.00 lags average (chosen by LLC)
Statistic p-value
Unadjusted t -2.3925
Adjusted t* -1.7178 0.0429

. xtunitrootllccrgr
Levin-Lin-Chu unit-root test for crgr
-------------------------------------
Ho: Panels contain unit roots Number of panels = 9
Ha: Panels are stationary Number of periods = 10

AR parameter: Common Asymptotics: N/T -> 0

Panel means: Included
Time trend: Not included

ADF regressions: 1 lag

LR variance: Bartlett kernel, 6.00 lags average (chosen by LLC)
Statistic p-value
Unadjusted t -7.9097
Adjusted t* -3.9185 0.0000

56
xtunitrootllccrri

Levin-Lin-Chu unit-root test for crri

-------------------------------------
Ho: Panels contain unit roots Number of panels = 9
Ha: Panels are stationary Number of periods = 10

AR parameter: Common Asymptotics: N/T -> 0

Panel means: Included
Time trend: Not included

ADF regressions: 1 lag

LR variance: Bartlett kernel, 6.00 lags average (chosen by LLC)
Statistic p-value
Unadjusted t -2.9542
Adjusted t* 3.1386 0.9992
xtunitrootllcinf

Levin-Lin-Chu unit-root test for inf

------------------------------------
Ho: Panels contain unit roots Number of panels = 9
Ha: Panels are stationary Number of periods = 10

AR parameter: Common Asymptotics: N/T -> 0

Panel means: Included
Time trend: Not included

ADF regressions: 1 lag

LR variance: Bartlett kernel, 6.00 lags average (chosen by LLC)

Statistic p-value
Unadjusted t -0.0944
Adjusted t* 0.3568 0.6394

57
58