International Journal of Management and Commerce Innovations ISSN 2348-7585 (Online)

Vol. 5, Issue 1, pp: (135-144), Month: April - September 2017, Available at: www.researchpublish.com

Effect of Bank’s Lending Behaviour on Loan

Losses of Listed Commercial Banks in Kenya
Michael Ochieng Obuya1, Dr. Tobias Olweny2
1 2
MBA (Finance) Student, Senior lecturer, 1,2Jomo Kenyatta University of Agriculture and Technology, Thika, Kenya

Abstract: This examined the effect of bank’s lending behaviour on loan losses of listed commercial banks in Kenya.
The study employed descriptive research survey design. The target population encompassed 11 listed commercial
banks in Kenya. The study was a census of listed commercial banks in Kenya. The data was extracted from CBK
Annual reports and Audited financial statements of individual commercial bank in Kenya. Descriptive analysis
involved mean, standard deviation, minimum and maximum while for inferential analysis; correlation analysis
was used to test the relationship between banks’ lending behaviour and loan losses of commercial banks in Kenya.
Simple OLS model was used to establish the causal effect relationship between lending behaviour and loan losses
of listed commercial banks in Kenya. The findings showed that total customer loans and quality of loans had a
statistically significant effect on loan losses. However the effect of lending rate, loan growth and loan portfolio
diversification on loan losses was not statistically significant. The study thus concluded that banks’ lending
behaviour has a significant effect on loan losses as shown by p value less than significance level in the ANOVA. The
study recommends to the management of listed commercial banks to take into consideration their total customers
loans and quality of loans when setting loan loss provisions to cover any eventual loan losses.
Keywords: Lending behaviour, Loan losses and commercial banks.

I. INTRODUCTION

A. Background of Study:
Lending institutions play a major role in economic growth and development through provision of credit to execute
economic activities. The major concern of any lender while advancing credit is how they will get their money back
(Fleisig, 1995), and this implies that the engagement between lenders and borrower is accompanied by certain level of
risk. The major types of risks faced by lending institutions globally include market risk, operational risk, and performance
and credit risks (Pyle, 1997). In the Kenyan banking sector for instance, while market risk is a great business concern for
all institutions, credit risk is cited as a major concern by 95 per cent of the banking institutions (CBK, 2011). The overall
observation of risks facing the banking sector is that while market risk can be easily managed through hedging activities,
credit risk has emerged as a new management challenge to financial institutions (Gonzalez-Paramo, 2010). Loan losses is
by far the most significant risk faced by banks and the success of their business depends on accurate measurement and
efficient management of this risk to a greater extent than any other risks (Gieseche, 2004). Coyle (2000) defines loan
losses as losses from the refusal or inability of credit customers to pay what is owed in full and on time. Loan losses is the
exposure faced by banks when a borrower defaults in honouring debt obligations on due date or at maturity. Banks’
lending behaviour describes the lending activities of a bank in terms of loans growth, total credit advanced, the lending
rate, quality of loans advanced and loan portfolio diversification (Foos et al.2010; Altunbas et al., 2011).To measure
banks’ lending behaviour, bank’s loan growth rate has used by most researchers which is defined similarly to Foos et al.
(2010) and Altunbas et al. (2011) as the a bank’s loan growth rate in a year. This takes account of the fact that high rates
of loan growth reflect excessive risk-taking if all other banks have lower growth rates. If banks raise lending by lowering
their lending standards, relaxing collateral requirements or a combination of both, higher rates of loan growth are
Page | 135
Research Publish Journals
 International Journal of Management and Commerce Innovations ISSN 2348-7585 (Online)
Vol. 5, Issue 1, pp: (135-144), Month: April - September 2017, Available at: www.researchpublish.com

associated with greater risk (Foos et al., 2010). Furthermore, banks which exhibit significantly higher loan growth rates
than their competitors may attract customers which have not been given a loan by other banks because they asked for too
low loan rates or provided not sufficient collateral relative to their credit quality (Foos et al., 2010). Loan losses can be
defined as inability of individuals or households to repay all debts fully and on time (Haas, 2006). First of all, a household
or an individual is over-indebted if they cannot cover all payment obligations arising from all debt contracts in a given
period by the excess cash, i.e. periodic cash income not used to cover all periodic expenses of the debtor, during that
period (Wisniwski, 2010). It should be noted that this definition still misses a dynamic perspective. Loan losses only
occur if this situation occurs chronically, i.e. in several periods in a row (Wisniwski, 2010) and against the borrowers’
will (Schicks, 2010). Haas (2006) also states a dynamic relationship between loan losses and characteristics of borrowers
and credit markets, similarly Gonzalez (2008) and Vogelsang (2003). Both the static one-period and the more dynamic
multi-period definition immediately lead to the most direct way of measuring of loan losses.
B. Statement of the Problem:
Financial institutions particularly commercial banks are very important in providing financial assistance to the economic
units in the society. However, just like other financial institutions, commercial banks experience numerous cases of loan
losses. The loan losses negates the profitability of the commercial banks. Additionally, Loan losses are not only argued to
harmfully affect the financial performance of commercial banks, but they also have other far reaching repercussions. This
is due to the fact that, other potential borrowers may be denied to access credit facilities since part of the funds that could
be extended as loans by the commercial banks are lost due to failure by past customers to pay back loans borrowed. The
loan losses also affect the economy of a country which explains the rationale behind the setting of guidelines by the
central bank for enabling financial institutions to alleviate loan losses. Kenyan banks are not an exception on the problem
of loan losses. Statistical information from CBK annual bank supervisory report of 2015, 2014 and 2013 showed that loan
losses category accounted for 1.5 % of the loan book in 2015 compared to 1.3 % in 2014 and 1.2 % in 2013. The
increases were occasioned by deteriorating asset quality, challenges in the business environment and increased interest
rates. Indeed, the Central bank of Kenya put two banks on receivership as of April 2016 that is Chase Bank and Imperial
Bank of Kenya due to poor performance majorly brought by NPLs leading to rising loan losses. This current study goes
one step further by analysing the effect of a bank’s lending behaviour on loan losses of individual banks listed by Nairobi
stock exchange (NSE) .To the best of my knowledge, there is no existing study in Kenya that has looked into effect of
bank’s lending behaviour on loan losses of listed commercial banks in Kenya.
B. Objectives of the Study:
The purpose of this study is to establish the effect banks’ lending behaviour on loan losses of listed commercial banks in
Kenya.
C. Hypotheses of the Study:
The study tested the following hypotheses:
H01: Loan growth has no significant effect on loan losses of listed commercial banks in Kenya.
H02: Lending rate has no significant effect on loan losses of listed commercial banks in Kenya.
H03: Quality of credit advanced has no significant effect on loan losses of listed commercial banks in Kenya.
H04: Total credit advanced has no significant effect on loan losses of listed commercial banks in Kenya.
H05: Loan portfolio diversification has no significant effect on loan losses of listed commercial banks in Kenya.

II. LITERATURE REVIEW

A. Theoretical Review:
This study was underpinned by three theories including Information Asymmetry Theory, Institutional Memory
Hypothesis and Modern Portfolio Theory. Information Asymmetry Theory was proposed by Akerlof (1970) and Stiglitz
& Weiss (1981) is based on the notion that the borrower is likely to have more information than the lender about the risks
of the project for which they receive funds hence they are in a position to negotiate optimal term for the transaction than
the banks. The theory of asymmetric information tells us that it may be difficult to distinguish good borrowers from bad
Page | 136
Research Publish Journals
 International Journal of Management and Commerce Innovations ISSN 2348-7585 (Online)
Vol. 5, Issue 1, pp: (135-144), Month: April - September 2017, Available at: www.researchpublish.com

borrowers (Auronen, 2003 & Richard, 2011), which may result into adverse selection and moral hazards problems
(Matthews & Thompson, 2008). Information asymmetry theory is considered relevant in this study on the effect of banks’
lending behaviour on loan losses of commercial banks in Kenya. The problem of moral hazard associated with the theory
leads to borrowers concealing material information concerning their ability to pay and risks associated with the
investment and the bank lending out credit without having complete information about borrower (Bester, 1994). The
banks end up giving low quality loans that result to increase in loan losses. At the same time due to the problem of
adverse selection associated with information asymmetry, the banks ends up charging high interest rates to covers for
increased risk of default due to opaqueness of customer credit history thereby leading to even more default as borrowers
cannot afford rising interest rates therefore increased non-performing loans that further leads to rising loan losses of the
commercial banks in Kenya.
The second theory Modern Portfolio Theory (MPT) is highly associated with Markowitz (1952. Modern portfolio theory
measures the benefits of diversification. MPT is an investment theory which tries to explain how investors could
maximize their returns and minimize their risks by diversification in different assets by carefully choosing the proportions
of various assets. Tobin (1958) expanded the theory of Markowitz’s (portfolio theory) by adding the analysis of risk-free
assets which made it possible to influence portfolios on the efficient frontier. The centerpiece of this theory is the capital
asset pricing model (CAPM) devised by Markowitz (1952). By combining different assets whose returns are not perfectly
positively correlated, MPT seeks to reduce the total variance of the portfolio return. MPT also assumes that investors are
rational and markets are efficient. Modern portfolio theory approach underpins current study and is relevant for analyzing
the effect of banks’ lending behaviour on loan losses of commercial banks in Kenya. MPT implies portfolio
diversification and the desired portfolio composition of commercial banks are results of decisions taken by the
management of commercial banks. Commercial Banks should consider diversifying loan portfolio to minimize risk of
credit takers defaulting in loans repayments and causing loans losses. Finally, Institutional Memory Hypothesis. The
institutional memory hypothesis was first articulated by Butlin and Boyce (1985).Under the institutional memory
hypothesis, the capacity of loan departments to evaluate risk and identify potential future problems deteriorates as time
passes since their last “learning experience” with problem loans. Early in a bank’s lending cycle immediately after a loan
bust, the lessons of the bank’s last bust are still fresh in the memory of loan officers who have just witnessed the ex post
realization of their prior loan decisions. Thus, as the bank starts its rebound from its most recent experience with problem
loans, the reservoir of lending knowledge is at its peak because these lessons are fresh in the minds of those loan officers
who survived the experience. As time passes since the bank’s last loan bust, the level of loan officer skill tends to
deteriorate. The officers do a worse job of screening, analyzing, and structuring their loans as they are originated,
monitoring them after origination, and designing and implementing work-out strategies when these loans become
distressed. This deterioration in loan officer ability may result in an easing of credit standards as officers become less able
to recognize potential loan problems and lower-quality borrowers as a consequence, banks may pool loan applicants that
might otherwise be rejected with acceptable credits and extend credit to additional borrowers (Office of the Comptroller et
al., 2001) Eventually, a bank’s loan boom turns to a bust and its loan officers turn more of their attention to managing
their distressed credits.
B. Empirical Review:
This section of the study examines relevant empirical literature on the relationship between study variables that is banks’
lending behaviour and loan losses. Igan & Pinheiro (2011) explored the risks posed by rapid credit growth recognizing the
two-way causality between credit growth and bank soundness. The dataset covers banks from 90 countries from 1995 to
2005 in OECD area. The authors show that rapid credit growth during the last decade weakened banks and it became less
dependent on bank soundness. Pakhchanyan & Sahakyan (2014) examined how progressive credit intermediation in the
Armenian banking sector affects financial stability of the banking sector. Particularly, the study analysed the impact of
abnormal loan growth on banks’ solvency, profitability and riskiness. The study used GMM technique to a panel of 22
Armenian banks covering the 2003–2014 period. The study finds that favourable macroeconomic performance and
abnormal loan growth positively affect subsequent loan losses. These findings support the view that, in the case of
economic boom and aggressive loan growth strategy, in order to attract new customers, banks ease their credit standards
and lend to weaker borrowers which further translates into increased loan losses. Further, their findings support their
hypothesis regarding abnormal loan growth, suggesting that banks with aggressive risk taking behaviour suffer more from
loan losses. In a paper by Köhler (2011), the researcher analysed the impact of loan growth and business model on bank

Page | 137
Research Publish Journals
 International Journal of Management and Commerce Innovations ISSN 2348-7585 (Online)
Vol. 5, Issue 1, pp: (135-144), Month: April - September 2017, Available at: www.researchpublish.com

risk in 15 EU countries. Results indicate considerable heterogeneity in risk-taking across banks and countries. Researcher
show that banks with high rates of loan growth are more risky. The effect, however, decreases with bank size possibly
because large banks are more active in volatile trading and off-balance sheet activities such as securitization that allow
them to increase their leverage. Results further indicate that banks become more risky if aggregate credit growth is
excessive. This even affects those banks that do not exhibit high rates of individual loan growth compared to their
competitors. Foos & Norden (2009) provides new comprehensive evidence on the inter-temporal relation between
abnormal loan growth and the riskiness of individual banks. Using Bankscope data on more than 16,000 individual banks
from 16 major countries during 1997-2007. The study finds that past abnormal loan growth has a positive and highly
significant influence on subsequent loan losses with a lag of two to four years. The positive relation between past
abnormal loan growth and contemporaneous losses is robust.
Pinho & Martins (2009) studied the determinants of generic and specific provisions in Portuguese banking. They tested
the model using a comprehensive panel of all financial institutions operating in Portugal between 1990 and 2000. The
study finds that specific provisions are mainly determined by the amount of non-performing loans, loan write-offs and
recoveries. The study also find that generic provisions are mainly explained by the amount of loans outstanding. Isa &
Yaziz, (2011) carried out a study to derive the determinants of loan loss provisions of commercial banks in Malaysia. A
single stage panel data multiple regression model that contains a mixture of quantitative and qualitative elements is
employed. The loan loss provisions is a dependent variable and the independent variables are non-performing loan (NPL),
interest income, net profit, loans & advances; and the Gross Domestic Product (GDP).This paper suggests in loan loss
provisions; non-performing loans, interest income, loans & advances, net profit, and the Gross Domestic Product. Lim
et.al (2013) did a study whose main purpose was to analyse the determinants of bank loan provisions in Malaysia. This
study investigates that what variables that will affect and how to affect the loan loss provision of the banks in order for the
banks to take note on these variables’ movement in managing their loan loss provision more effectively to reduce losses.
This paper is a quantitative research that obtaining data from secondary source. In this 9 years of data was included from
2003 till 2011. The log method is used since the data that collected from annual report of 9 banks in Malaysia is too huge.
In this research, the target population is the licensed local commercial banks in Malaysia. Researchers found that GDP
and market lending rate are insignificant variables while the other five variables (total loan, earnings before tax and
provision (EBTP), bank size, non-performing loan (NPL) and equity ratio) were significant in determining the bank loan
provision in Malaysia. Kanagaretnam et al. (2008) investigated that total loan is positively related to loan loss provision.
Therefore, it is expected that total loans of the banks are positively related to loan loss provision in Malaysia.
Kanagaretnam, Krishnan, and Lobo (2010) found that total loan of the banks is positively correlated to loan loss
provision. When the total loans of the bank increase, loan loss provision will increase too. The increase of total loans may
lead to higher defaults of the loans. Thus, banks normally will set high loan loss provision in order to prevent the banks
from being bankrupt. However, the study of Craigwell & Elliott (2011) is inconsistent with the study of Kanagaretnam et
al. (2010). Based on the result of the study of Craigwell and Elliott (2011), they found that there exists significant
negatively relationship between total loan and loan loss provision. When the total loan of banks increases, loan loss
provision moves in opposite direction. Some banks, for example, larger banks in Barbados that have largest loan
portfolios normally able to forecast the possibility of default of the loans. Thus, banks will expect that loan loss provision
should decrease when total loans increase. Murira (2010) sought to determine the relationship between loan portfolio and
financial performance of commercial banks in Kenya. The researcher used causal research design. The target population
composed of 43 commercial banks in Kenya. The study concludes that there exists a relationship between loan portfolio
and financial performance of commercial banks in Kenya as loan portfolios are the major asset of banks and other lending
institutions. Maina (2003) carried out a research on the risk based capital standards and the riskiness of bank portfolios in
Kenya. The study established that the challenges include taxes, investor preferences, portfolio constraints, lack of
knowledge from consultants and cultural hurdles. Ndung’u (2003), sound asset and liability management have significant
influence on profitability. Among the external factors, high market interest rate was found to have an adverse effect on
financial institution's profitability in Kenya. The study also found that the prerequisites to operational efficiency include
the adaptation of an effective service delivery methodology and significant institutional competence in such areas as
delinquency control, information management, and staff development.
Study by Okoye & Eze (2013) examined the impact of bank lending rate on the performance of 11 Nigerian Deposit
Money Banks between 2000 and 2010. The study utilized secondary data econometrics in a regression, where time-series

Page | 138
Research Publish Journals
 International Journal of Management and Commerce Innovations ISSN 2348-7585 (Online)
Vol. 5, Issue 1, pp: (135-144), Month: April - September 2017, Available at: www.researchpublish.com

and quantitative design were combined and estimated. The result confirmed that the lending rate has significant and
positive effects on the performance of Nigerian deposit money banks. Glen and Mondragon-Velez (2011), loan loss
provision is positively related to lending rate. This is due to there is higher probability that firms and individuals may not
be able to service on their debt during recession. Pain (2003) and Arpa, Giulini, Ittner and Pauer (2001) investigated the
influence of the business cycle on loan loss provision in UK and Austrian banks respectively. They showed that, banks set
a higher provision when GDP reduces. Provision gets a boost during the time when the fall in GDP was concluded. Also,
raise of banks’ profitability in good time can encourage banks to have a lower provision. However, there are some
journals stated that loan loss provision is positively related to GDP. Anandarajan, Hasan and McCarthy (2007) stated that
when GDP increases, firms or companies tend to increase their borrowing to expand activities. This may improve the risk
of borrowing. Banks need to increase loan loss provision to absorb the additional risk. Therefore the relationship between
loan loss provision and GDP is treated as positive.

III. RESEARCH METHODOLOGY

A. Research Design, Population and Collection:
In this study, the researcher used descriptive survey and empirical research design. According to Saunders, Lewis and
Thornhill (2009), descriptive survey research design is meant to give an output of statistical information about an aspect
of a study that is of interest to policy makers in a bid to aid them in making informed decisions. The study described the
effect of banks’ lending behaviour on loan losses of listed commercial banks in Kenya. The target population for this
study was all the 11listed commercial banks in Kenya. According to NSE, there are 11 registered commercial banks that
have shares trading in NSE (NSE, 2016). A census was done in order to provide a true measure of population. As result,
no sampling technique was necessary because the research utilized the entire population. The period of study was five
years 2011, 2012, 2013, 2014 and 2015.Due to the nature of financial studies, the researcher used secondary data sources.
Secondary data was sourced from statistics maintained by the central bank of Kenya and audited financial statement of
individual banks covering five years (2011-2015). The data extracted was recorded on data collection sheets. Reliability
and validity of test instrument was also important concept for this study. For the purpose of this study, data reliability and
validity was ensured by collecting data from official sources such as annual audited accounts and corporate websites.
B. Data Analysis Techniques and Procedures:
The data collected was examined before analysis commenced for completeness and consistency .The data was analyzed
using descriptive and inferencial.The panel methodology was aided by Excel 2013 and STATA software Version 14
software. Descriptive statistics included measures of central tendencies and dispersion. Bivariate Pearson correlation,
multiple regressions and ANOVA were used to test the significance of the effect of banks’ lending behaviour on loan
losses of listed commercial banks in Kenya. The study used empirical model represented by the regression formulae
below.

LNLLPit = β0+ β1LGti t +β2NPLTCLit+ β3LNTCLit +β4LENDRit+β5 PORTDIVit + β6 GDP t +

ɛi…………………………….Equation (1)
Where LNLLP: Natural logarithm of Loan Loss Provision of bank, LG: Loan Growth as a percentage LNTCL: Natural
Logarithm of total customer loans, GDP: Gross Domestic Product, LENDR: Lending rate, PORTDIV: Portfolio
diversification, NPLTCL: Non-performing Loans to Total loans, β0 =intercept term, β1, β2, β3, β4, β5 and β6, : are the
coefficients of the independent variables respectively, i : 1, 2…….11: Number of banks t = 2011,2012,2013,2014 and
2015 and ɛi = Error term

IV. RESEARCH FINDINGS AND DISCUSSION

A. Correlation Analysis:
The researcher carried out par wise correlations analysis to assist explains the association between banks’ Lending and
loan losses of listed commercial banks in Kenya. The researcher used pairwise Pearson Correlation to establish the
relationship as shown in table 1

Page | 139
Research Publish Journals
 International Journal of Management and Commerce Innovations ISSN 2348-7585 (Online)
Vol. 5, Issue 1, pp: (135-144), Month: April - September 2017, Available at: www.researchpublish.com

Table 1: Pairwise Pearson Correlation Coefficients

. pwcorr LENDR PORTDIV LNTCL NPLTCL LG GDP LNLLP, obs sig star(5)

LENDR PORTDIV LNTCL NPLTCL LG GDP LNLLP

LENDR 1.0000

55

PORTDIV 0.2658* 1.0000

0.0498
55 55

LNTCL -0.1715 -0.2086 1.0000

0.2106 0.1264
55 55 55

NPLTCL 0.0302 0.1800 0.0287 1.0000

0.8269 0.1884 0.8351
55 55 55 55

LG -0.3942* -0.5793* 0.0559 -0.2405 1.0000

0.0029 0.0000 0.6853 0.0769
55 55 55 55 55

GDP -0.1269 -0.0510 0.3925* 0.2562 0.1025 1.0000

0.3560 0.7115 0.0030 0.0591 0.4567
55 55 55 55 55 55

LNLLP -0.0430 -0.0329 0.6083* 0.6986* -0.1400 0.4562* 1.0000

0.7552 0.8114 0.0000 0.0000 0.3080 0.0005
55 55 55 55 55 55 55

Table 1presents the following :- Loan growth (LG) and loan losses: The researcher wanted to establish the relationship
.
between loan growth and loan losses (LNLLP) of the 11 listed commercial banks in Kenya. Pairwise Pearson correlation
coefficient was calculated at 0.05 level of significance. Pearson’s correlation (r) indicated that there was a statistically
insignificant negative correlation between loan Losses and loan growth (r = -0.1400, p = 0.3080 and α = 0.05). With loan
growth being negatively correlated with loan losses; it suggests that an increase in Loan growth leads to reduction in loan
losses. This negative association could be explained by the fact that loan growth in the listed commercial banks may does
not necessarily be positively associated with loan losses as the listed banks are usually tire one banks with strict control
from CBK and NSE Lending rate (LENDR) and loan losses: There was a statistically insignificant moderate negative
correlation between lending rate and loan losses (r = - 0.0430, p = 0.7552 and α = 0.05). The negatively correlation
suggests that an increase in lending rate leads to an reduction in loan losses .This negative correlation could be explained
by the fact the positive correlation between lending rate and loan losses is only possible during period of stagnated
economic growth but the period under study was accompanied by increasing economic growth throughout the study
period.
Quality of loans (NPLTCL) and loan losses: There was a strong positive statistically significant correlation between
quality loans and loan losses (r = 0.6986, p = 0.000 and α = 0.05). Quality of loans being positively correlated with loan
losses, suggests that an increase in the stock of poor quality loans leads to an increase in loan losses. This positive
correlation could be explained by the fact that increase in the stock of poor quality loans is associated with banks not
recovering loans which become toxic or impaired assets hence increases in loan losses and such banks must increase loan
loss provision with increase in stock of poor quality loans.
Total loans (LNTCL) and loan losses: Pearson’s correlation coefficient indicated a statistically significant positive
correlation between total customer loans and loan losses (r = 0.6083, p = 0.000 and α = 0.05). Total loans being strongly
and positively correlated with loan losses, suggests that an increase in total loans could be associated with increasing loan
losses. This positive correlation could be explained by the fact that banks with large total customer loans must may also
experience large volume of loan losses hence must set a side big chunk of profits as provision for loan losses.

Page | 140
Research Publish Journals
 International Journal of Management and Commerce Innovations ISSN 2348-7585 (Online)
Vol. 5, Issue 1, pp: (135-144), Month: April - September 2017, Available at: www.researchpublish.com

Loan portfolio diversification (PORTDIV) and loan losses: There was a statistically insignificant negative correlation
between loan portfolio diversification and loan losses (r = -0.0329, p = 0.8114 and α = 0.05). The negatively correlation
suggests that an increase in loan portfolio diversification is associated with reduction in loan losses. This could be
explained by the fact that by diversifying loan portfolio book, the unsystematic risks are spread over the less risky
financial assets like government bonds and bills with almost zero default hence overall loan losses is expected to decline.
Gross Domestic Product (GDP) and loan losses: There was a statistically significant positive correlation between Gross
Domestic Product and loan losses (r = 0.4562, p = 0.0005 and α = 0.05).Gross domestic product being positively
correlated with loan losses, suggests that an increase in Gross Domestic Product is associated with an increase in loan
losses. This positive correlation could be explained by the fact that increasing gross domestic product means increasing
positive business climate, this could make banks to offer more loans which will eventually lead to increasing loan losses
in absolute figures as banks increase loan loss provisions to cover the expected loan losses before they are experienced.
B. Regression Analysis:
Simple OLS Regression analysis was multiple in natures as there were five independent variables and one control
variable. The independent variables were: Loan growth, total customer loans, loan portfolio diversification, lending rate,
quality of loans and gross domestic product. The dependent variable was loan losses proxied by natural logarithm of loan
loss provision. The findings are presented in table 2
Table 2: Summary of Simple OLS Regression Results
. regress LNLLP LENDR PORTDIV LNTCL LG GDP NPLTCL

Source SS df MS Number of obs = 55

F(6, 48) = 42.72
Model 43.4166955 6 7.23611592 Prob > F = 0.0000
Residual 8.13045906 48 .169384564 R-squared = 0.8423
Adj R-squared = 0.8226
Total 51.5471546 54 .954576937 Root MSE = .41156

LNLLP Coef. Std. Err. t P>|t| [95% Conf. Interval]

LENDR 1.220905 1.777048 0.69 0.495 -2.35209 4.793899

PORTDIV -.691394 .7337727 -0.94 0.351 -2.166743 .7839548
LNTCL .9330523 .1090407 8.56 0.000 .7138113 1.152293
LG -.4498123 .8389841 -0.54 0.594 -2.136703 1.237078
GDP 15.81567 14.05253 1.13 0.266 -12.43884 44.07018
NPLTCL 62.36685 5.822996 10.71 0.000 50.65893 74.07477
_cons -4.835193 2.0941 -2.31 0.025 -9.045663 -.6247227

Tables
. 2 indicate that the model explains 84.23 % of the total variations in loan losses as shown by the coefficient of
determination (R2) value of 0.8423. The remaining 15.77% Variations in loan losses is explained by other factors not
included in the model as captured by error terms. It is therefore clear that lending behaviour explains 84.23 % variations
in loan losses. The overall significance of the model was 0.000 with an F value of 42.72. The level of significance was
lower than 0.05 and this means that lending behaviour do show statistically significant effect on loan losses. Table 2
further shows the coefficients of independent variables (Lending rate, loan growth, loan portfolio diversification, total
customer loans and economic growth) the values of p and values of t .The estimated model is shown in equation 2
LNLLPt = -4.8351+ -0.45LGit +62.36NPLTCLit + 0.93LNTCLit +1.22LENDRit- 0.69
PORTDIVit+15.8156GDPt+ɛi…………………………………….Equation (2)
Effect of Loan growth on loan losses: The researcher wanted to test the null hypothesis that loans growth has no
significant effect on loan losses of listed commercial banks in Kenya using simple OLS model. It was established that
loan growth had a negative statistically insignificant effect on loan losses (β1= -0.4498, t = -.54, p = .594 and α = 0.05).
Hence null hypothesis was accepted. The value β1was negative showing that loan growth has a negative effect on loan
losses of listed commercial banks in Kenya hence when loan growth changes by one unit, loan losses changes by 0.4498

Page | 141
Research Publish Journals
 International Journal of Management and Commerce Innovations ISSN 2348-7585 (Online)
Vol. 5, Issue 1, pp: (135-144), Month: April - September 2017, Available at: www.researchpublish.com

in the reverse direction.

Effect of Lending Rate on Loan Losses: The researcher tested the null hypothesis that lending rate has no significant
effect on loan losses in listed commercial banks in Kenya. Results show that lending rate had a statistically insignificant
positive effect on loan losses (β2 = 1.2209, t = 0.690, p = .495 and α = 0.05). The null hypothesis was thus accepted. The
value β2 was positive showing that lending rate has a positive effect on loan losses among the 11 listed commercial banks
in Kenya. Any unit increase in lending rate leads to increase in loan losses by 1.2209 units. However the insignificant
effect could be attributed to that fact that the study used listed commercial banks that contains most of the tire 1 banks that
boast of well-trained credit management teams that ensures that borrowers are clients are financially stable and have high
repayment chances for the loans taken.
Effect of quality of loans on loan losses: The study tested null hypothesis that Quality of loans has no significant effect
on loan losses of listed commercial banks in Kenya .The findings show that Quality of loans had a statistically significant
positive effect on loan losses (β3 = 62, t = 10.71, p = .000 and α = 0.05).The null hypothesis was thus rejected and .The
value β3 was positive showing that quality of loans has a positive effect on loan losses of listed commercial banks in
Kenya. Any increase in the stock of toxic loans by one in the bank leads to increase in loan losses by 62 units. The
possible explanation for this significant positive effect is that when stock of non-performing loans increases as a ratio total
loans, the chances of loan losses increases as some of the non-performing loans will eventually turn into loan losses hence
banks must increase loan loss provisions in anticipation. Empirical review also agrees with this result. Bank increases
their provisions in response to an increase in credit risk. (Eng & Nabar, 2007).
Effect of total customer loans on loan losses: The researcher tested the null hypothesis that total customer loans has no
significant effect on loan losses of listed commercial banks in Kenya. It was established that total customer loans had a
statistically significant positive effect on loan losses (β4 = 0.9330, t = 6.56, p = .000 and α = 0.05).The null hypothesis
was thus rejected. The value β4 was positive showing that total customer loans has a positive effect on loan losses hence
any unit increase in total customer loans leads to 0.933 units increase in loan losses. The positive effect can be attributed
to the fact that when banks offer more absolute loans at any particular time, they are required to increase loan loss
provision to cover for any expected loan losses before they actually incurred. Study by Krishnan and Lobo (2010) agrees
with this study as they also found that total loan of the banks is positively correlated to loan loss provision. However,
Craigwell & Elliott (2011), found that there exists significant negatively relationship between total loan and loan loss
provision.
Effect of loan portfolio diversification on loan losses: The null hypothesis that loan portfolio diversification has no
significant effect on loan losses of listed commercial banks in Kenya was also tested. Study results established that loan
portfolio diversification had a statistically insignificant negative effect on loan losses (β5 = -0.6913, t = -0.94, p = .351
and α = 0.05).The null hypothesis was thus accepted and alternative hypothesis rejected showing that loan portfolio
diversification has no significant effect on loan losses of listed commercial banks in Kenya. The value β5 was negative
showing that a unitary increase in loan portfolio diversification leads to 0.6913 units reduction in loan losses. The
negative effect can be explained by the theory of portfolio that states that a firm can reduce risks of loan default by
spreading the risk in less risky treasury bills and bonds hence the more the diversification the lower the loan losses.

III. CONCLUSION
The study was carried out to establish the effect of banks’ lending behaviour on loan losses of listed commercial banks in
Kenya. The results of the study were as follows: Total customer loans and Quality of loans had statistically significant
effect on loan losses of listed commercial banks in Kenya. However loan portfolio diversification, lending rate and loan
growth had a statistically insignificant effect on loan losses. The study recommends the following: Firstly, basing on this
study, Management of commercial banks will able to make business policy changes regarding their lending behaviour to
reduce loan losses. One of key policy issue could be to increase loan loss provision with increase in total customer loans
as their prepare for eventual loan losses that may result. Other areas of business policy change involve diversifying loan
portfolio book by taking up more government financial assets to spread loan losses. Secondly, the board of CBK will be
able fine tune loan loss provision policies by basing it on lending activities of individual commercial banks. There should
be portion of loan loss provision for individual banks that should be adjusted based on lending behaviour of individual
banks. The risk weighted minimum capital requirements should also be based on the exposure of individual banks to loan
losses that eat into capital of a banks .The minimum capital requirements should be adjusted depending on lending

Page | 142
Research Publish Journals
 International Journal of Management and Commerce Innovations ISSN 2348-7585 (Online)
Vol. 5, Issue 1, pp: (135-144), Month: April - September 2017, Available at: www.researchpublish.com

activities of individual banks. Finally, Future researchers will find the finished report an invaluable document as it will
provide base literature on topics touching on the relationship between lending activities and loan losses of commercial
banks. Concerning areas of further research, the current study was limited to effect of lending behavior on loans losses of
listed commercial banks in Kenya. The study was restricted to listed commercial banks and the data was restricted to five
years. Another study should be carried out that looks at effect of lending behavior on loan losses in all commercial banks
in Kenya to establish if results hold also in other commercial banks. Another study should also be carried out using long
term data covering over twenty years since some variables can be best observed in a long period of time. Future studies
can also relate lagged values of loan growth with contemporaneous loan losses of commercial banks.

REFERENCES
[1] Al-Khouri, R. (2011). Assessing the Risk and Performance of the GCC Banking Sector, International Journal of
Finance and Economics, 65, 72-8.
[2] Allen F., Carletti E. and Marquez R. (2011), Credit Market Competition and Capital Regulation, Review of
Financial Studies, 24,(4).983-1018
[3] Altunbas, Y., Manganelli, S. and Marques-Ibanez, D. (2011), Bank Risk During the Financial Crisis – Do Business
Models Matter?, ECB Working Paper Series, No. 1394, European Central Bank, Frankfurt
[4] Athanasoglou, et al, (2006), Determinants of Bank Profitability in the South Eastern European Region, MPRA Paper
10274, University Library of Munich, Germany
[5] Athanasoglou P. P., Brissmis S. N. & Delis M. D. (2008). Bank-Specific, Industry Specific and Macro-economic
Determinants of Bank Profitability. Journal of International Finance, Markets, Institution and Money, 18, 121-136.
[6] Auronen L. (2003).Asymmetric Information: Theory and Applications. Paper presented in the Seminar of Strategy
and International Business as Helsinki University of Technology, May 21st 2003
[7] Beltratti, A., & Stulz, R. M. (2012). The credit crisis around the globe: Why did some banks perform better?.
Journal of Financial Economics, 105(1), 1-17.
[8] Berger, A. N., & Udell, G. F. (2004). The institutional memory hypothesis and the procyclicality of bank lending
behavior. Journal of financial intermediation, 13(4), 458-495.
[9] Bhattacharya, Kaushik. (2003). How good is the BankScope database? A cross validation exercise with correction
factors for market concentration measures. BIS Working Paper No. 133, Bank for International Settlements,
September 2003.
[10] Bloem, M. A. and Gorter, N. C. (2001), “Treatment of Non-Performing Loans in Macroeconomic Statistics”, IMF
Working Paper, WP/01/209.
[11] Bofondi, M. & Gobbi, G. (2003). Bad Loans and Entry in Local Credit Markets. Rome: Bank of Italy Research
Department.
[12] Bonner C., Van Lelyveld I. and Zymek R. (2014), Banks’ Liquidity Buffers and the Role of Liquidity Regulation,
Journal of Financial Services Research, forthcoming.
[13] Borio, C. and Drehmann M., (2009), Assessing the Risk of Banking Crises Revisited, Bank for International
Settlements Quarterly Review, March.
[14] Boyd J., De Nicolo’ G. and Jalal A. M. (2006), Bank Risk-Taking and Competition Revisited: New Theory and
Evidence, IMF Working Paper 06/29 (Washington: International Monetary Fund).
[15] Central Bank of Kenya (2011) Risk Management Survey for the Banking Sector. Nairobi. Central Bank of Kenya
[16] Cooper, D. R., and Schindler, P. S. (2006). Business Research Methods (9th edition). USA: McGraw-Hill.
[17] Coyle, B. (2000). Framework for Credit Risk Management, Chartered Institute of Bankers, United Kingdom
[18] De Lis, S. F., Pagés, J. M., & Saurina, J. (2001). Credit growth, problem loans and credit risk provisioning in Spain.
bis Papers, 1, 331-353.

Page | 143
Research Publish Journals
 International Journal of Management and Commerce Innovations ISSN 2348-7585 (Online)
Vol. 5, Issue 1, pp: (135-144), Month: April - September 2017, Available at: www.researchpublish.com

[19] Demirgüç-Kunt, A., and Huizinga, H. (1999). Determinants of Commercial Bank Interest Margins and Profitability:
Some International Evidence. World Bank Econ Rev 13, 379408.
[20] Fleisig, H. (1995). The Power of Collateral-How problems in securing transactions limit private credit for Moveable
property. Public Policy for the Private Sector-Note No 43. The World Bank.
[21] Foos, D., Norden, L., and Weber, M. (2010). Loan growth and riskiness of banks. Journal of Banking & Finance,
34, 2929–2940.
[22] Gieseche, K. (2004). Credit Risk Modelling and Valuation: An Introduction, Credit Risk: Models and Management,
2, Cornell University, London.
[23] Gonzalez-Paramo, J.M (2010). The Challenges of Credit Risk Management-lessons learnt from the crisis. Speech
delivered during Risk Europe 2010,in Frankfurt am Main, on 26th May 2010.
[24] Hakenes, H., and Schnabel, I. (2011). Bank size and risk-taking under Basel II. Journal of Banking & Finance, 35,
1436–1449.
[25] Hilbers, Paul, and others, 2005, Assessing and Managing Rapid Credit Growth and the Role of Supervisory and
Prudential Policies, IMF Working Paper No.05/151, Washington: International Monetary Fund.
[26] Igan, D., & Pinheiro, M. (2011). Credit growth and bank soundness: fast and furious?. IMF Working Papers, 1-27.
[27] Isa, M., & Yaziz, M. (2011, March). Determinants of Loan Loss Provisions of Commercial Banks in Malaysia. In
2nd International Conference on Business and Economic Research (ICBER).
[28] Klein, N. (2013). Non-performing loans in CESEE: Determinants and impact on macroeconomic performance.
[29] Laeven, L. and Majnoni, G. (2003), Loan Loss Provisioning and Economic Slowdowns: Too Much, Too Late?,
Journal of Financial Intermediation, 12, 178-197.
[30] Lim, C. K. (2013). The Determinants Of Bank Loan Provision In Malaysia (Doctoral Dissertation, Universiti Tunku
Abdul Rahman).
[31] Okoye, V., & Eze, O. R. (2013). Effect of bank lending rate on the performance of Nigerian deposit money banks.
International Journal of Business and Management Review, 1(1), 34-43.
[32] Pain, D. (2003). The provisioning experience of the major UK banks: a small panel investigation (Bank of England).
[33] Pakhchanyan, S., & Sahakyan, G. (2014). Drivers of Bank Risk, Solvency, and Profitability in the Armenian
Banking System.
[34] Pinho, P. S. D., & Martins, N. C. (2009). Determinants of Portuguese bank’s provisioning policies: Discretionary
behaviour of generic and specific allowances.
[35] Quagliariello, Mario. (2007). “Banks’ riskiness over the business cycle: a panel analysis on Italian intermediaries.”
Applied Financial Economics 17, 119-138.
[36] Saunders, M., Lewis, P and Thornhill, A (2009), Research Methods for Business Students, 5th edition, Harlow,
England, Prentice Hall.

Page | 144
Research Publish Journals