The International Journal of Business and Finance Research

Vol. 10, No. 1, 2016, pp. 95-108

ISSN: 1931-0269 (print)
[Link]
ISSN: 2157-0698 (online)

A COMPARISON OF CREDIT RISK MANAGEMENT IN

PRIVATE AND PUBLIC BANKS IN INDIA
Isaiah Oino, University of East London

ABSTRACT

Like other corporations, banks want to create value and seek ways to control risk while aiming to enhance
productivity and performance. This is achieved by granting credits to customers from the money deposited
by the depositor, thus placing them at risk in the case of defaulting. Despite this risk, banks must continually
issue credit since it is the key source of its profitability. This research study assesses the impact of credit
risk management on Indian public and private banks during the 2009-2012 period. Using pooled OLS, fixed
effects and random effects, the study examines credit risk management in seven private banks and seven
public banks. The results show that private banks are more capitalized and more profitable than public
banks. In addition, in both cases asset quality measured using non-performing assets with negative
coefficients significantly influenced bank profitability. The study extrapolates the importance of regulatory
capital and the importance of risk management in ensuring stability in the financial industry.

JEL: G02, G18

KEYWORDS: Capital Adequacy, Non-Performing Assets, Performance, Net Interest Margin

INTRODUCTION

D uring the last thirty years, the world has experienced a number of financial and banking crises.
Most of these crises occurred in developing countries. The dominant crises corresponded to
deregulatory processes that forced extension of credit in short timeframes. Continuous rises in asset
prices in the long run precipitate bubbles. Ultimately, bankruptcies resulted from non-performing loans,
leading to acute banking crises and credit losses. Over the last five years, the banking sector has undergone
great metamorphosis as a result of the financial crisis of 2008. There is more emphasis on not only capital
adequacy, but also on moral hazard. The global bailout of banks was a major paradigm shift, with taxpayers
stepping in to rescue banks as a result of their short-term profits or bonuses. The argument for the bailout
is that banks play a significant role of intermediation in the economy.

Significant reforms have been suggested in response to banking crises. For instance, Basel I, and Basel II
and Basel III represent the banking supervision accords proposed by the Basel Committee (Felix and
Claudine, 2008). The Basel Accord implemented a framework in 1988 by G-10 central banks focusing on
credit risk and safe banking. The Reserve Bank of India (RBI) issued a guideline on credit risk management
in 2002 per international law. Basel I played a critical role in strengthening the financial system, through
several measures like weak incentives and deficiency of risk management. As a result, Basel II sought to
reveal banks’ fundamental risk exposure and response to financial innovation like securitization.
Paradoxically, the incidence of crisis did not decline despite the introduction of succeeding development.

The recent international financial crisis signifies that risk management in banking sectors is significantly
inadequate. The drive for globalization, innovation and financial deregulation has not eliminated credit risk
even if the off-balance and market risk hold more interest in the wake of the disruption to the global
financial markets (Paradi et al. 2012). Thus, credit risk is still the greatest concern to banking authorities

95

Electroniccopy
Electronic copy available
available at:
at:[Link]
[Link]
I. Oino | IJBFR ♦ Vol. 10 ♦ No. 1 ♦ 2016

and regulators. There is huge economic impact that is linked with bank failure because of the ripple effects
that spread from banking to other sectors of the economy. Therefore, credit risk management is an issue of
great value given that the core function of every bank is credit granting. The character of the banking sector
has been so perceptive since more than 85% of their liability is deposits (Saunders & Cornett, 2005), and
banks mobilize these deposits to credit for borrowers, which in fact is an income-generating function of
banks. Besides all other services, bank must generate credit for customers to make money, enhance growth
and remain competitive in the market.

Multiple studies have already been carried out on the effects of bank credit risk management, such as
Kithinji (2010) and Poudel (2012). However, all research focuses on the component of credit risk
management in banks regarding credit risk measurement, provisioning, credit derivative and its influences
on bank profitability. This work goes further by incorporating the capital adequacy, taking account of the
recent revisions and guidelines of Basel III.

Research objective: The primary objective of this research is to elucidate how Indian banks practice and
manage credit risk, and thus attention is tailored towards assessing the influence of credit risk management
on profitability over four years (2009-2010). The ultimate objective of this study is to consider different
parameters applicable to credit risk management and how they influence financial performance.

The rest of this paper is structured as follows: the next section examines the literature review. I then describe
the data and the methodology and discuss the results. The final section concludes.

LITERATURE REVIEW

Commercial banks play leading roles in lending and intermediation between lenders and borrowers. A bank
has several functions: mainly accepting deposits and granting credit facilities like loans and advances,
which comprise its primary function. Regardless of the significant role that banks play in financial markets
by linking lenders to borrowers, instability in the global economic environment, currency values and
financial markets has impinged deleteriously on bank functionality and profitability.

Although the key causes of severe bank disruptions and failures continue to be inadequate credit risk
management, credit granting remains the principal business of every bank in the world. Well, in reality,
operating banks are considered a channel for economic prosperity and growth, whereas weakly functioning
ones do not merely obstruct economic progress but also intensify poverty. However, banks are vulnerable
to various risks such those from the market, interest, credit, and operational risk, which impact financial
performance in various ways. The size and level of the loss caused by credit risk can be seen to be severe
compared to other risk, as it directly threatens bank solvency (Frederick, 2012).

Credit Risk

The survival, performance and sustainability of banks are hugely reliant on correct measurement, effective
and sound management of credit risk. As stated by the Basel Committee on Banking Supervision, or the
BCBS (2006), “Credit risk is the potential risk of loss due to the failure payment by the obligators in the
terms of loans or other types of credit”. It implies that the risk emerges from the perspective that bank
counterparties or borrowers are unwilling to perform or fulfill their obligations. Moreover, in other words,
the value of the bank’s assets, particularly its loans, will reduce worth and probably become valueless,
thereby damaging the solvency state of banks. This is in line with Chen & Pan (2012), who termed credit
risk as ‘’the extent of value fluctuation in the debt derivatives and instruments due to transform in the core
credit quality of counterparties and borrowers”. BCBS (2006) claimed that historical understanding and
occurrence reveals that concentration of credit risk mostly in the asset portfolio is the foremost cause of
bank dysfunction.

96

Electroniccopy
Electronic copy available
available at:
at:[Link]
[Link]
The International Journal of Business and Finance Research ♦ VOLUME 10 ♦ NUMBER 1 ♦ 2016

When credit risk increases, there is inadequate capital because the bank will search for different sources to
meet and mitigate losses. In addition, this leads to a decline in its liquidity status. This will consequently
lead to a decline in profitability. It is worth noting that credit risk and returns are interlinked such that the
higher the credit risk, the less return and vice versa. The trade-off between the two illustrates that high-risk
securities (higher yield loans) reward a risk premium (higher average return) because of greater insecurity
of payment. Thus, the return / value and average revenue can be increased only by increasing risk. Greuning
and Bratanovic (2003) stress that it is critical to understand that credit risk has always been the major hazard
to any bank’s performance and the major cause of bank collapse.

Credit Risk Management in Banks

Credit granting is the foremost source of revenue in banks. Credit risk management needs to be integrated
into the decision making process before granting credit. Simply, this involves identifying, analyzing and
assessing, monitoring and controlling credit. This has a direct impact on the level of non-performing loans
as well as on the sum of loans and advances extended to customers (Kithinji, 2010). It should be the top
priority of bank operations in order to enhance sustainability. Despite these facts, significant bank problems
have increased considerably in both established and emerging economies (Fredrick, et al. 2012). Several
studies and researchers have identified the causes of these disruptions as localized to the banking sector in
addition to numerous other factors. The problem regarding the credit, especially weakness in credit risk
management, has been recognized as the main reason for bank problems (Richard et al. 2008).

Over the years, despite the innovations in the financial services sector, Hennie (2003) stated that credit risk
still remains the most prominent reason for bank failure. For this reason, “more than 80% of the bank’s
balance sheet commonly relates to this aspect of risk management.” The major reason for serious banking
problems directly relate to poor portfolio risk management, loose credit standards for counterparties and
borrowers and lack of awareness of changes in economics. Collectively, these observations indicate the
enormous critical role played by credit risk management in the entire bank risk management approach. The
ultimate objective of credit risk management is to intensify the risk-adjusted rate of return by controlling
and standardizing credit risk exposure.

Credit Risk and Bank Profitability

Empirical evidence suggests that credit risk management is a predictor of bank performance. For
illustration, non-performing asset as a parameter of credit risk can reduce the worth and undermine credit
structure. As posited by (Afriyie and Akotey, 2010) loan default shrink the resource support for further
lending, affect the borrower’s confident and deteriorate the staff morale.

Banks incur significant costs in controlling overdue loans and this can naturally affect profitability levels.
The major source of credit risk emanates from inappropriate credit policies, volatile interest rate, low capital
and liquidity, direct lending, poor loan underwriting, poor loan lending, government intervention and
improper supervision from the central bank. When credit risk increases, it leads to bank solvency and
liquidity problems. If the bank lends and the borrower for some reason defaults, i.e. repayment and interest
are not forthcoming, the problems will be twofold. First, the bank has to cease interest accrual on the
doubtful loan, so there is an immediate earnings loss. Secondly, the bank has to maintain provision for non-
performing assets from the net interest income that implies the profit will be decreased. Therefore, increase
in credit risk will cause loss and elevate the marginal expenditure of bank equity and debt to get funds from
alternative sources to cover the losses (Sobhy, 2013).

97

Electronic copy available at: [Link]

I. Oino | IJBFR ♦ Vol. 10 ♦ No. 1 ♦ 2016

Review of Related Empirical Literature

Several scholars like Kolapo, Ayeni and Ojo (2012), Kinthinji (2010) and Li Yugi (2007) carried out broad
research studies on this topic and delivered mixed results. For example, Kolapo, Ayeni and Ojo (2012)
noted that 100 percent increases in non-performing loan reduce profitability (ROA) by about 6.2 percent,
although the study was characterized by serial correlation depicted by high Durbin-Watson. On the other
hand, Kinthinji (2012) observed that there is no relationship between profits, amount of credit and the level
of nonperforming loans. However, the study produced a moderate R squared of 39%, which computes to a
negative adjusted R squared (-0.226).

Boahene, Dasah and Agyei (2012) adopted the regression analysis to evaluate the significant relationship
between credit risk and Ghanaian bank profitability. Their research followed Manzura and Juanjuan (2009)
by using the ratio of non-performing loans to total assets as an indicator for credit risk management and
return on equity as a measure of bank profitability. They highlighted that credit risk management impinges
dramatically on bank profitability. The study indicated that higher capital adequacy positively contributes
to bank profitability.

Poudel (2012) assessed the effect of credit risk management in bank performance of Nepal during the 2001-
2011 period using 31 banks. The capital adequacy ratio, cost per loan and default rates were used as credit
parameters, whereas ROA was a performance indicator. The results showed that credit risk management
has a strong impact on bank financial performance.

Li yuqi (2007) studied the determinants of bank profitability and its impact on risk management practices
in the United Kingdom. The study utilized regression analysis between 1999 and 2006. Six measures of
determinants of bank profitability were employed. He used capital, liquidity and credit as internal factors
in bank performance. Inflation rate, interest rate and GDP growth rate were used as external determinants
of bank profitability. Return on Asset (ROA) was used as a measure of a bank’s performance. It was found
that liquidity and credit risk have a negative impact on bank profitability.

Kolapo, Ayeni and Ojo (2012), while analyzing credit risk management efficiency from 2004-2009 in
commercial banks of Nigeria, suggested some additional views into credit risk as profit-enhancing
apparatus. Regression analysis was used for data analysis and revealed there is nominal causation between
bank performance and deposit exposure. Kithinji (2010) determined the impact of credit risk management
in Kenyan banks for the 2004-2008 period. He employed credit indicators as the ratio of non-performing
loans and advances the ratio of loans and advances to total assets. The study revealed that the volume of
profit of commercial banks is not determined by the level of non-performing loans and credit, as the
implication recommends that other factors apart from non-performing loans and credit influence bank
profitability.

Besides, Naceur and Kandil (2008) evaluated the influence of capital obligation on bank performance and
cost of intermediation employing Generalized Method of Moment (GMM) on time series data during the
1989-2004 period. They used the ratio of net loans to deposit and ratio of capital to total asset and deposit
as independent variables while return on equity and return on asset as the dependent variables to measure
bank profitability. The results showed that the capital adequacy is a forecaster of a bank's performance.
Gurdmundssoa, Ngok-Kisingula and Odongo (2013) assessed the task of regulatory capital obligation on
bank control and competition in Kenya from 2001-2011 using panel data estimation of time series data.
The results showed that, regulatory competence enhances the competition in banking sectors. Ravindra,
Vyasi and Manmeet (2008) studied the impact of capital adequacy of the performance of selected
commercial Banks in India using panel data models. The authors concluded that there is a positive
association between capital adequacy ratio and profitability.

98

Electronic copy available at: [Link]

The International Journal of Business and Finance Research ♦ VOLUME 10 ♦ NUMBER 1 ♦ 2016

Common in the aforementioned studies is that they excluded the public banks, hence the need for the current
research. To a certain extent, one may argue that commercial banks are heterogeneous from public banks
because of political interference.

DATA AND METHODOLOGY

For the empirical investigation, both static and dynamic panel data analyses are utilized and effectively
applied to a dataset consisting of the seven largest private and public Indian banks spanning the 2009-2012
period. The term ‘panel data’ refers to the pooling of time series and cross-sectional observations of banks,
on the same individual variables over several time periods (Baltagi, 2003). Panel data allow one to account
for heterogeneity of the entities being observed. In addition, because of the size of the data set,’ there is
more variability and hence less collinearity among the variables. Private banks were separated from public
banks because of the heterogeneity between the two.

The Static Model

The use of pooled time series and cross sections allows us to take into account the unobserved and time-
invariant heterogeneity across different banks. For the estimation of the models, the dataset that consists
of N, which denotes partial units, was used; i = 1,…,N observed at T time periods, t = 1,…,T. Therefore,
the total number of observations is T × N. Then, y is a (TN × 1) vector of endogenous variables, X is a (TN
× k) matrix of exogenous variables, which does not include a column of units for the constant term. In the
context of the research, N = 7 and T = 4. Given this, one can write a generic pooled linear regression model
by ordinary least square procedure as shown in equation 1. I checked the regression specification using
Ramsey’s reset shown in Appendices 1 and 2.

N
y it = β 0 + ∑ β1 X it +ε it (1)
N =1

where yit is the dependent variable, β0 is the intercept term, βi is a k×1 vector of parameters to be estimated
on the explanatory variables, and xit is a 1×k vector of observations on the explanatory variables, t = 1, . .
. , T, i = 1, . . . , N and εit is random error term. Pooled OLS enables the researcher to capture the variation
of what emerges through time or space simultaneously.
The specification in equation (1) suggests a linear panel data model. The associated assumptions about the
model that can be taken into account are as follows:

a-Error terms are normally distributed and have zero mean and standard deviation si2, eit ~ i.i.d. (0 ,si2 )
b-Similar variances among banks, si2 = se2 "i
c-Zero covariances among banks, Cov(eit ,ejs) = 0 for i ≠ j

If the homogeneity hypothesis is rejected, the estimates based on the pooled model are as follows:

Y1 X1 ε1
Y2 X2 ε2
. = . β + . = Xβ + ε (2)
. . .
YN X N εN

99

Electronic copy available at: [Link]

I. Oino | IJBFR ♦ Vol. 10 ♦ No. 1 ♦ 2016

However, if the difference between βs however significant is thought to be small, then one could consider
a trade-off of accepting some bias in order to reduce variances. If the departure of homogeneity is so great,
then this could result in serious distortion in the conclusion, hence the choice of the best alternative static
specification that links the pros and cons of each specification. I then tested for heteroskedasticity and serial
correlation as shown in Tables 1 and 2. The presence of heteroskedasticity, the statistical inference based
on σ2(X ′X)-1, would be biased, and t-statistics and F-statistics are inappropriate.

Table 1: Heteroskedasticity Test: Breusch-Pagan-Godfrey

Panel A: Private Banks

F-statistic 1.579 Prob. F(4,23) 0.213

Obs*R-squared 6.032 Prob. Chi-Square(4) 0.196
Scaled explained SS 2.694 Prob. Chi-Square(4) 0.610
Panel B: Public Banks
F-statistic 0.645 Prob. F(4,23) 0.635
Obs*R-squared 2.826 Prob. Chi-Square(4) 0.587
Scaled explained SS 1.424 Prob. Chi-Square(4) 0.839
This table tests for heteroskedasticity using Breusch-Pagan-Godfrey under the assumption that the error variance is a linear function of Xt. This
can be written as: δ 2 = α1 + α 2 X t for t =1,2,…..n
t

Table 2: Breusch-Godfrey Serial Correlation LM Test

Panel A: Private Banks

F-statistic 0.073 Prob. F(2,21) 0.9295
Obs*R-squared 0.194 Prob. Chi-Square(2) 0.9074
Panel B: Public Banks
F-statistic 1.885 Prob. F(2,21) 0.176
Obs*R-squared 4.261 Prob. Chi-Square(2) 0.118

The LM tests for serial correlation often in time series data. That is, the violation of: Cov(ut ut − s ) = E (ut ut − s ) = 0 for all t ≠ s

The fixed effect model assumes that despite the variation in intercept across the banks, each individual
intercept does not vary from time to time. Therefore, the intercept β1it means it is time-invariant. Therefore,
the fixed effect model can be expressed as follows:

N
yit = β1it + ∑ β1 X it +ε it (3)
N =1

Where y is the dependent variable for profitability and X denotes the variables of interest (capital adequacy,
non-performing loans, credit to deposit and net interest margin ratios). The fixed effect across the firm was
not significant individually or as a group as shown in Table 3.

The common slope coefficients and constants may not be fixed but random. In this case, the random effects
model would be appropriate. In a nutshell, the random effect is a compromise between pooling under
complete homogeneity and pooling with common slope coefficient, but with the intercept varying cross-
sectionally. That is, all of the elements in the coefficient vector, slopes as well as intercepts, are random
variables rather than fixed parameters.

100

Electronic copy available at: [Link]

The International Journal of Business and Finance Research ♦ VOLUME 10 ♦ NUMBER 1 ♦ 2016

Table 3: Fixed Effects on Banks

Private Banks Public Banks

Bank Effect Bank Effect
ICICI Bank 0.012 STATE -4.439
HDFC -0.030 Punjab -1.373
Axis Bank 0.308 Canara 3.955
Kotak Mahihndra 0.022 baroda -0.272
ING Vysya Bank -0.593 INDIA -0.937
Induslnd bank 0.134 UNION 1.598
Yes bank 0.156 IDBI 1.469
This table shows the fixed effects across each bank, both public and private.

Under the assumption of intercepts for the cross-section, which are random variables, and slope coefficients,
which are fixed parameters, the vector would represent slopes while only the random error term would have
two components. Thus:

µ i η i1
µi ηi 2
εi = . + . (4)
. .
µi η iT

The µ i represents randomness, which is due to the choice of the cross-section (the random intercept), while
η it represents the randomness stemming from the cross-section and time period.

The argument in favor of the random effects model is that the fixed effects model often results in a loss of
many degrees of freedom and also eliminates a large portion of the total variation in the panel. Another
argument is that βi combines several factors specific to the cross-sectional units and as such they represent
‘specific ignorance’ (Maddala, 2001). Hence, βi can be treated as random variables by much the same
argument that ε it represents ‘general ignorance’ can be treated as random variables. On the other hand,
there are two arguments in favor of the use of the fixed effects model. The first, common in the analysis of
variance literature, is that if the analysis wants to make inferences about only this set of cross-sectional
units, then βi can be treated as fixed. On the other hand, if one wants to make inferences about the population
from which these cross-sectional data come, then βi should be treated as random. The Hausman test, which
is derived from the t test, can identify the best model using the restricted F test, which can be expressed as
follows:

2 −R2 ) / m
( RUR
F= R (5)
2 ) /( n − k )
(1− RUR

Where RUR 2
and RR2 are the values obtained from the unrestricted and unrestricted regressions, respectively.
In addition, model specification was tested using the Ramsey reset test as shown in Appendix 2. Further, in
order to test for structural break, recursive least squares (RELS) was used. This assesses how β changes
over time. The basic idea is that if β changes significantly, then there is a structural break. The results of
the two panels indicated no strong evidence to suggest poolability of the model.

101

Electronic copy available at: [Link]

I. Oino | IJBFR ♦ Vol. 10 ♦ No. 1 ♦ 2016

RESULTS

As shown in Table 4, the mean return on assets that measures profitability for private banks is 14.01, which
is slightly higher than that of public banks (13.30). This could be attributed to the efficiency of private
banks in managing the non-performing assets. As shown, the mean for non-performing assets for private
banks is just 0.38 compared with 13.30 for public banks. One of the biggest assets (excluding properties)
is advances. If such advances are not readily accessible, a bank’s financial performance will be adversely
affected. The Reserve Bank of India has classified advances into basically four categories: standard asset,
which refers to a loan that is easily recoverable. The second category is substandard assets, which are non-
performing assets for periods of less than 12 months. The third category is doubtful debts, which are assets
classified as non-performing loans exceeding 12 months. The fourth category is a loss where the loss has
been identified and will be written off as bad debt.

Table 4: Descriptive Statistics

Private Banks Public Banks

CAR NPA NIM CD ROE CAR NPA NIM CD ROE
Mean 16.85 0.38 4.28 70.83 14.01 13.30 13.30 13.30 13.30 13.30
Maximum 21.22 1.11 6.86 100.34 20.89 15.38 15.38 15.38 15.38 15.38
Minimum 11.65 0.03 2.34 31.57 7.06 11.31 11.31 11.31 11.31 11.31
Std. Dev 2.67 0.29 1.15 17.01 3.79 1.06 1.06 1.06 1.06 1.06
N 28 28 28 28 28 28 28 28 28 28
This table shows the descriptive statistics results for both private and public banks. CAR is capital adequacy ratio; NPA is non-performing assets;
NIM is the net interest margin; CD is credit to deposit ratio; ROE is the return of equity.

In terms of capital adequacy, the results indicate that private banks are more capitalized, with a mean of
16.85 compared with 13.30 for public banks. Better capitalized banks may reflect higher management,
quality, thereby generating a positive and a negative coefficient sign in the income and cost regression,
respectively, resulting in an expected positive impact on profitability. In addition, Berger (1995) noted that
well capitalized firms face lower expected bankruptcy costs, which in turn reduce their cost of funding and
in turn increase their income. The results also show that the net interest margin for private banks is 4.28
compared with 13.30 for public banks.

Table 5 shows the correlation matrix for both private and public banks. Clearly, the capital adequacy ratio
is positively associated with profitability in both cases. This implies that the more capitalized the bank, the
more profitable it is. The capital adequacy ratio is the total of Tier 1 and Tier 2 capital and is measured as
the ratio of capital to risk-adjusted assets and off-balance sheet exposure determined on a risk-weighted
basis. A higher ratio reflects a bank’s ability to absorb unanticipated capital losses.

As regards to asset quality, I analyzed non-performing loans. The degree to which provisions are made in
anticipation of, or concurrent with, actual impairment in the loan portfolio reflects credit quality. A
privatized bank may aggressively build its loan portfolio and could be forced to make large provisions for
unanticipated bad debts. As shown in the correlation matrixes, there is a negative association between NPA
and ROE. Although in both cases, the coefficient is negative, the magnitude for private banks is much
higher than for the public banks. It is also possible that a privatized bank may be more efficient in managing
its loan portfolio and therefore carry only a small loan loss provision. However, it is paramount to note that
banks can smooth incomes by making higher provisions than necessary when credit quality and net income
are high. Consequently, they may not increase provisions as much as they should if credit quality is
deteriorating. Gunther and Moore (2000) argue that income smoothing will ensure that banks with asset
quality problems can raise net income and retained earnings, thereby boosting Tier 1 capital.

102

Electronic copy available at: [Link]

The International Journal of Business and Finance Research ♦ VOLUME 10 ♦ NUMBER 1 ♦ 2016

Table 5: Pearson Correlation Matrix for Private Banks

CAR NPA NIM CD ROE

Panel A: Private Banks
CAR 1
NPA 0.273 1
NIM -0.015 0.058 1
CD -0.079 0.044 0.319 1
ROE 0.031 -0.706 -0.106 0.226 1
Panel B: Public Banks
CAR 1
NPA -0.321 1
NIM 0.277 0.019 1
CD -0.093 0.313 -.243 1
ROE 0.340 -.352 .619 117 1
The table shows the Pearson correlation between the independent and dependent variables. Panel A is on the private banks and panel B is on
public banks. CAR is capital adequacy ratio; NPA is non-performing assets; NIM is the net interest margin; CD is credit to deposit ratio; ROE is
the return of equity.

Further, as shown in Table 6, capital adequacy has positive coefficient and significant at 1% using pooled
OLS for private banks. However, as regards to the public banks, although positive coefficient, it's not
significant. This is in line with Pasiouras and Kosmidou (2007) and García-Herrero et al. (2009) implying
that, the best-performing banks are those that maintain a high level of equity relative to their assets. One
could attribute this to the fact that, the banks with higher capital ratios tend to face lower costs of funding
due to lower prospective bankruptcy costs. For a bank with capital below its equilibrium ratio, the expected
bankruptcy costs are relatively high, and an increase in the capital ratios raises the expected profits by
lowering the interest expenses on uninsured debt. However the negative association between capital
adequacy and profitability was noted by Goddard et al. (2010) in eight European Union member countries
between 1992 and 2007. The negative coefficient could be because banks are required to retain a certain
amount that is not lend out and hence is tied capital. However, banks with higher capital–asset ratios are
considered relatively safer in the event of loss or liquidation. On the other hand, low-capitalized banks may
be considered risky. The results show a positive effect of inefficiency on risk-taking, which supports the
moral hazard hypothesis that poor performers are more vulnerable to risk-taking than high-performance
banks.

The need for risk management in the banking sector is inherent in the nature of the banking business. Poor
asset quality is one major cause of bank failures. During periods of increased uncertainty, financial
institutions may decide to diversify their portfolios and/or raise their liquid holdings in order to reduce their
risk. Abreu and Mendes (2002), who examined the banks in Portugal, Spain, France and Germany, noted
that the loans-to-assets ratio, as a proxy for risk, has a positive impact on the bank’s profitability. Table 6
shows that in both private and public banks, non-performing assets have a significant negative coefficient
using pooled OLS in line with Bourke (1989) and Molyneux and Thornton (1992). This implies that, the
more the non-performing loans, the less profitable the bank is hence decreasing liquidity position. An
interesting result is the effect of interest margin. The results indicate a negative coefficient for both public
and private banks. Although it is significant at 1% for public banks, it is 10% for private banks. The negative
and significant coefficient implies that the lower the interest the bank charges, the more profitable they are.
The lower the interest, the more the bank is likely to attract borrowers and the ease it likely the borrower to
honor the obligation. Indeed, the correlation for both private and private banks indicates that there is a
positive association between the NIM and NPA indicating the higher the interest rates, the higher the non-
performing.

103

Electronic copy available at: [Link]

I. Oino | IJBFR ♦ Vol. 10 ♦ No. 1 ♦ 2016

Table 6: Regression Analysis Using Pooled OLS, Fixed Effect and Random Effect

Private Banks Public Banks

Variable Pooled OLS Fixed Effect Random Effect Pooled OLS Fixed Effects Random Effect
C 15.849** 20.701** 16.332*** 3.650 5.959 4.176
(1.392) (6.087) (3.742) (3.770) (10.059) (8.751)
CAR 0.348*** -0.110 0.299 0.306 0.005 0.291
(0.065) (0.255) (0.178) (0.258) (0.888) (0.611)
CD -0.030** -0.021 -0.032 0.006 -0.052 0.002
(0.010) (0.041) (0.027) (0.017) (0.065) (0.042)
NIM -0.352* -0.360 -0.232 4.465*** 5.793* 4.446***
(0.154) (0.686) (0.436) (0.446) (2.563) (1.074)
NPA -10.730*** -4.737 -10.664*** -3.496*** -2.148** -3.472**
(0.597) (7.839) (1.716) (0.666) (3.656) (1.619)
Adj. R sqrd 0.67 0.68 0.57 0.49 0.41
F –Statistic 87.396*** 6.593** 9.677*** 41.395*** 0.57 5.601**
AIC 4.383 4.69 5.169 5.286
Hannan-Quinn 4.421 4.844 5.208 5.446
DW 1.761 2.030 2.029 1.600 1.968 0.953
Hausman test Chi sqr stat. 5.838 Chi sqr stat. 10.091
p. value. 0.212 p. value. 0.038
This table shows the result of running multiple regression analyzing applying the pooled ordinary least square, fixed effects and random effects.
Independent variables: CAR is capital adequacy ratio; NPA is non-performing assets; NIM is the net interest margin; CD is credit to deposit ratio
and dependent variable, ROE is the return of equity. AIC is Akaike Information Criteria. ***, **, and * indicate significance at the 1,5 and 10
percent levels respectively.

CONCLUSION

The objective of this study was to assess how private banks differ from public banks in terms of risk
management. Using 14 banks for the 2009-2012 period, the data indicates that private banks are more
capitalized compared with public banks. The CAR is 17% for private banks and 13% for public banks. The
RBI has set CAR for Indian banks to be 9%, which is higher than in most developed countries. The capital
adequacy ratio on public sectors continues to decline due to an elevated credit demand and obligation of
higher provisions to buffer against asset quality deterioration. NPA negative impacts on the capital
adequacy ratio, profitability and bank credibility (Kumar and Singh, 2012). Using fixed effects, the results
are consistent with previous studies (Kaaya and Pastroy, 2013) Frederick, 2012), Kithinji, 2010) Felix and
Claudine, 2008). That is, the effect of NPA is significant in influencing negatively the profitability of both
private and public banks. The increase of NPA on both public and private banks could be attributed to the
diversion of funds away from the actual purpose for which they were granted as well as misappropriation
of funds by borrowers. Apart from that severe economic conditions and market factors stemming from
regulatory changes, recessionary conditions and feeble resources for inefficient management and stressed
labor relations have affected the conditions of business and forced them to default on their loan repayments.
The research denotes that the public sector banks had to face a reduction in NIM significantly high
compared to private banks. This could be because private banks could hedge themselves by diversification.
Therefore, the results indicate that credit risk management significantly affects both banks, though
compared to the public sector, private banks are far better capitalized and managed more effectively in
terms of asset quality. However, for future research, there is a need to include other macroeconomic
variables and the size of the bank in assessing capital adequacy and profitability for both private and public
banks.

104

Electronic copy available at: [Link]

The International Journal of Business and Finance Research ♦ VOLUME 10 ♦ NUMBER 1 ♦ 2016

Appendix 1: Measurement of the Variables

Variable Proxy Measurement

Profitability ROE Profit available to shareholders/ shareholders’ funds.
Capital adequacy ratio CAR Tier 1 + Tier 2 capital
Asset quality NPA Non-performing assets/ total assets.
Management quality Credit Deposit Credit (loans) / Total deposits
Earning ability NIM Interest earning- Interest on deposit.
This table shows the measurements of variables. Dependent variable, profitability measured by Return on Equity (ROE); dependent variables,
capital adequacy ratio (CAR), asset quality measured by non-performing assets (NPA), management quality and earning ability measured by net
interest margin (NIM)

Appendix 2: Test of Specification. Ramsey RESET

Panel A: Public Banks

Value df Probability
t-statistic 0.151 22 0.880
F-statistic 0.023 (1, 22) 0.880
Likelihood ratio 0.029 1 0.864
F-test summary:
Sum of Sq. df Mean Squares
Test SSR 0.284 1 0.284
Restricted SSR 271.72 23 11.81
Unrestricted SSR 271.43 22 12.338
LR test summary:
Value df
Restricted LogL -71.546 23
Unrestricted LogL -71.531 22
Panel B: Private Banks
Value df Probability
t-statistic 0.297 22 0.768
F-statistic 0.088 (1, 22) 0.768
Likelihood ratio 0.112 1 0.737
F-test summary:
Sum of Sq. df Mean Squares
Test SSR 0.495 1 0.495
Restricted SSR 123.80 23 5.382
Unrestricted SSR 123.30 22 5.604
LR test summary:
Value df
Restricted LogL -60.541 23
Unrestricted LogL -60.484 22

This table shows the result of Ramsey Reset test employed to test a linear specification against a non-linear specification using F test statistic: The
F test statistics is expressed as:

( SSRYˆ − SSRYˆ 2 ) / M
F( M ; N − k −1) =
SSRYˆ 2 /( N − K )
( SSRR − SSRUR ) / M
=
SSRUR /( N − K )
Where SSRs are the sum of squared residuals for the respective regressions;
M is the number of restrictions; N is the number of observations;
K is the number of parameters estimated in the unrestricted equation.
Specification: ROE C CAR CREDIT_DEPOSIT NIM NPA. Omitted Variables: Squares of fitted values

105

Electronic copy available at: [Link]

I. Oino | IJBFR ♦ Vol. 10 ♦ No. 1 ♦ 2016

Appendix 3: Redundant Fixed Effects Tests

Panel A: Public Banks

Effects Test Statistic d.f. Prob.

Cross-section F 2.383 (6,17) 0.074

Cross-section Chi-square 17.092 6 0.008
Panel B: Private Banks
Cross-section F 1.551 (6,16) 0.024
Cross-section Chi-square 12.379 6 0.054
This table shows that in both panels, the p-values associated to the F-statistics and the Chi-square statistics are both significant at 10%, which
provides evidence against the null hypothesis that the fixed effects are all equal to each other. This suggests that there is unobserved heterogeneity.

REFERENCES

Abreu, M. and Mendes, V. (2002). Commercial Bank Interest Margins and Profitability: Evidence from
E.U. Countries. Working Paper Series, Porto.

Afriyie, H. O., and Akotey, J. O (2010): credit risk management and profitability of selected rural and
community banks in Ghana. Catholic University College of Ghana.

Baltagi, B. H. 2003. Econometric analysis of panel data. London: Wiley.

Banking Supervision. Basel Committee on Banking Supervision. Available at:

[Link] Accessed on 10 January 2014.

Basel Committee on Banking Supervision (2009): Principles for the Management of Credit Risk -
consultative document; Basel, Switzerland, Available on at http/www/Bank for International Settlements.

Basel Committee on Banking Supervision (September 2006). Principles for the Management of Credit
Risk, Risk Management Group of the Basel Committee on Banking Supervision. Accessed at:
[Link] Accessed 02 Feb 2014.

Berger, A. (1995). The profit–structure relationship in banking: Tests of market-power and efficient-
structure hypotheses. Journal of Money, Credit and Banking, Vol. 27, pp. 404–431.

Boahene S. H. Dasah J. and Agyei S. K. (2012). Credit Risk and Profitability of Selected Banks in Ghana.
Research Journal of Finance and Accounting. Vol. 3(07). pp. 6-14.

Bourke, P. (1989). Concentration and other determinants of bank profitability in Europe, North America
and Australia. Journal of Banking and Finance, Vol. 13 (1), pp. 65–79.

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

Felix, A.T and Claudine, T. N (2008). Bank Performance and Credit Risk Management, Unpublished
Masters Dissertation in Finance, University of Skovde

Frederick, T., Masood, O., Thapa, P. D. P., Bellalah, M., Levyne, O. (2012). Does Co-integration and
Causal Relationship Exist between the Non-stationary Variables for Chinese Bank's Profitability? An
Empirical Evidence. // International Journal of Business; Vol. 17 Issue 2, p194.

106

Electronic copy available at: [Link]

The International Journal of Business and Finance Research ♦ VOLUME 10 ♦ NUMBER 1 ♦ 2016

García-Herrero, A., Gavilá, S. and Santabárbara, D. (2009). What explains the low profitability of
Chinese banks? Journal of Banking and Finance, Vol. 33, pp. 2080–2092.

Goddard, J., Molyneux, P., Liu, H. and Wilson, J. O. S. (2010). The Persistence of Bank Profit. Bangor
Business School Working Paper.

Greuning H. V. and Bratanovic, S. B. (2003) “Analyzing and Managing Banking

Risk: A framework for assessing corporate governance e and Financial Risk”, The World Bank,
Washington D.C.

Gudmundssoa, R. Ngoka-Kisingul, K. and Odongo M. T. (2013), “The role of capital requirement on

bank competition and stability: The Case of Kenyan Banking Industry” Centre for Research on Financial
Markets and Policy.

Gunther, J. and Moore, R. (2000), “Early warning models in real time,” Federal Reserve Bank of Dallas,
Financial Industry Studies, working paper, No. 1-00, October.

Hennie V. G (2003), Analyzing and Managing Banking Risk: A Framework for Assessing Corporate
Governance and Financial Risk, 2nd Edition, Washington DC: World Bank Publications.

Kaaya, I. and Pastory, D. (2013). Credit Risk and Commercial Banks performance in Tanzania: a Panel
Data Analysis. Research Journal of Finance and Accounting, 4(16), pp.55 -62.

Kithinji, A. M. (2010). Credit Risk Management and Profitability of Commercial Banks in Kenya, School
of Business, University of Nairobi, Nairobi.

Kolapo, T. F, Ayeni R. K. and Oke, O. (2012). “Credit Risk Management and Banks Performance”
Australian Journal of Business and Management Research. Vol. 2 (2), pp 31 – 38.

Kumar, M. and Singh, G. (2012), Mounting NPAs In Indian Commercial Banks, International Journal Of
Transformations In Business Management, (Ijtbm, Vol 1, Issue 6, Apr - Jun 2012 ISSN 2231 - 6868)

Li Yuqi (2007). “Determinants of Banks profitability and its implication on Risk management practices:
Panel Evidence from the U.K”, University of Nottingham

Maddala, G. S. (2001), Introduction to econometrics. 3rd Ed. John Willey & Son Ltd.

Manzura, B and Juanjuan S. (2009). “Credit risk management and profitability in commercial banks in
Sweden” School of Business Economics and Law in Gothenburg. Available at:
[Link] Accessed on 01 Feb 2014.

Molyneux, P. and Thornton, J. (1992). Determinants of European bank profitability: A note. Journal of
Banking and Finance, Vol. 16, pp. 1173–1178.

Naceur, S. B. and Kandil, M., (2008). “The impact of capital requirement on banks costs of
intermediation and performance” IMF working paper. WP/13/160. Available at:
[Link] Accessed on 10 March 2014.

Paradi, J. C., H. Zhu and B. Edelstein (2012). 'Identifying managerial groups in a large Canadian bank
branch network with a DEA approach.' European Journal of Operational Research. Vol. 219: pp. 178-
187.

107

Electronic copy available at: [Link]

I. Oino | IJBFR ♦ Vol. 10 ♦ No. 1 ♦ 2016

Pasiouras, F. and Kosmidou, K. (2007). Factors influencing the profitability of domestic and foreign
commercial banks in the European Union. Research in International Business and Finance, Vol. 21, pp.
222–237.

Poudel, R. P. S. (2012). The impact of credit risk management on financial performance of commercial
banks in Nepal. International Journal of Arts and Commerce, Vol. 1(5), pp. 9-15.

Ravindra Y, Vyasi R. K. and Manmeet S. (2008). “The impact of capital adequacy requirements on
performance of scheduled commercial banks” Asian-pacific Business review Vol. 1. pp.1-5.

Richard E, Chijoriga M, Kaijage E, Peterson C, Bohman H. 2008. Credit risk management system of a
commercial bank in Tanzania. International Journal of Emerging Markets. 3(3):323-332.

Saunders, A. and Corneet, M. (2005), Financial Institution Management, McGraw Publication, London.

Sobhy. N. A. M. (2013), The Impact of Service Quality on Financial Performance and Corporate Social
Responsibility: Conventional Versus Islamic Banks in Egypt, International Journal of Finance and
Accounting, Vol. 2 (3), pp. 150-163.

ACKNOWLEDGEMENT

I would like to thank the journal editors, Mercedes and Terrance Jalbert and two anonymous referees, for
their insightful comments. Any errors are my own.

BIOGRAPHY

Dr. Isaiah Oino is a Senior Lecturer at University of East London, Business and Law School. He can be
contacted at the following address: Stratford Campus, Water Lane London E15 4LZ, United Kingdom. Tel:
02082234896. Email: [Link]@[Link]

108

Electronic copy available at: [Link]