Pacic-Basin Finance Journal 28 (2014) 5875

Contents lists available at ScienceDirect

Pacic-Basin Finance Journal

journal homepage: www.elsevier.com/locate/pacfin

Capital adequacy and lending and deposit

behaviors of conventional and Islamic banks
Mastura Abdul Karim a, M. Kabir Hassan b,
Tauq Hassan c, Shamsher Mohamad d
a

Ministry of Science and Technology, Malaysia

University of New Orleans, New Orleans, USA
c
Universiti Putra Malaysia, Malaysia
d
INCEIF, Malaysia
b

a r t i c l e

i n f o

Article history:
Received 21 August 2013
Accepted 15 November 2013
Available online 19 December 2013
JEL classication:
G21
G28
G32
Keywords:
Capital adequacy ratio
Islamic and conventional banks
Loans and deposits

a b s t r a c t
Capital adequacy plays an important role in determining banking
activities. A bank must hold a minimum level of capital to ensure
sufcient funds to buffer against unexpected losses or adverse shocks.
This study analyzes and compares Islamic and conventional banks in
14 Organization of Islamic Conference (OIC) countries from 1999 to
2009. The empirical evidence suggests that capital requirements have
a signicant impact on the deposit and lending behaviors of the 52
Islamic banks (IBs) and 186 conventional banks (CBs) in the sample.
There is a strong positive relationship between capital requirements
and deposit and loan growth for both IBs and CBs.
2013 Published by Elsevier B.V.

1. Introduction
The issue of capital requirement for nancial institutions has become more important since the nancial
crisis in the USA since 2007, and is now found in most of the current banking literature1 ever since the
introduction of the Basel Accord. Insufcient equity capital has been partly blamed for the Global Financial Crisis,
and a similar situation also occurred during the Asian nancial crisis of the late 1990s. The U.S. economic crisis
has put the spotlight back on the issues of subprime lending and the importance of maintaining sufcient
E-mail addresses: amam68@yahoo.com (M. Abdul Karim), mhassan@uno.edu, KabirHassan63@gmail.com (M.K. Hassan),
tauq_h@upm.edu.my, tauuk@yahoo.com (T. Hassan), shamshermohd57@gmail.com (S. Mohamad)
1
Some of the studies on the impact of capital requirement include: Chiuri et al. (2002)a high minimum capital requirement
causes bank credit to shrink, Barrell and Gottschalk (2006)an increase in capital adequacy ratios has an adverse impact on Brazil
and Mexico's GDP, but for Mexico, increases in capital adequacy ratios do not affect commercial loan supply, as is the case in Brazil,
and Peek and Rosengren (1995)binding capital requirements cause bank loans to shrink.
0927-538X/$ see front matter 2013 Published by Elsevier B.V.
http://dx.doi.org/10.1016/j.pacn.2013.11.002

M. Abdul Karim et al. / Pacic-Basin Finance Journal 28 (2014) 5875

59

enough capital to support adverse market conditions. During the crisis, many institutions did not have a
sufcient enough nancial cushion to absorb the losses from the large loan defaults, and the capital growth of
the banks fell far below the growth of total credit and overall riskiness of the assets (Cannata and Quagliariello,
2009). The losses have adversely impacted banks and their ability to lend, causing economic activity to slow
down.
Ensuring that banks develop and maintain a minimum capital adequacy requirement (CAR) is crucial in
preventing them from failing. A bank with a sound capital position is able to pursue business opportunities more
effectively and has more time and exibility to deal with problems arising from unexpected losses, thus
achieving increased protability (Athanasoglou et al., 2008). When the Bank of International Settlements (BIS)
rst introduced the minimum capital requirement in 1996, it was to ensure that banks are prudent in
maintaining adequate reserves as a shield to protect themselves and their depositors. The idea behind the CAR is
to ensure that banks set aside capital from their own money for each set of investments they make. The riskier
the banks or their businesses, the more capital is required to be put aside. This ensures a safer investment return
for both shareholders and depositors.
The nature of banking activities is another issue that needs to be addressed in determining the adequacy of
bank capital. Since banks are funded by deposits, which are short-term in nature, banks should only be allowed
to nance short term loans from depositors' funds. However, if banks use this short-term fund to nance long
term securities such as mortgage securities, it will cause a crucial maturity mismatch. Therefore, if there is an
unexpected rise in interest rates, particularly once the economy declines, banks may not be able to maintain the
investments on their books. As banking theories recognize, the function of banks is to transform short-term
securities into the long-term securities that borrowers desire (Freixas and Rochet, 1997). The maturity
mismatch exposes the bank to interest rate risk and is a pertinent issue that needs to be addressed.
During the 2008 U.S. nancial turmoil, the collapse of major nancial institutions was due to increasing
amounts of capital, mainly by the top nancial institutions, being placed into long term and risky investments,
such as those of the real estate market. These investments were made during the booming economy without
diligent analyses of the risks involved. The collapse resulted in a credit or capital crunch, and other banks began
to tighten their lending due to a heightened fear of defaults. A similar scenario also happened in Indonesia in
2001. The dependency on bank loans by most of the Indonesian businesses resulted in a mismatch crisis,
whereby banks were using too much of their short term deposits to nance long term investments, which later
contributed to the economic crisis in Indonesia (Vandenbrink, 2005). The nancial crisis of the late 1990s
that occurred in many East Asian countries and the 2007 credit crisis in the U.S. were due, in part, to the
vulnerabilities of banks, and conventional banks (CBs) were often the most signicantly affected. Additionally, in
light of the fact that Islamic banking has become more prominent in the global nancial sector, this concern has
also led to an increased interest from many researchers. There is a general belief that Islamic banks (IBs) are less
vulnerable than CBs to economic distress, due to the fact that the assets of IBs are backed by prot-loss sharing
deposits.
From one point of view, the minimum CAR is irrelevant in IBs, because the prot-loss sharing contract
helps to reduce the overall risk of investment faced by the bank (Pellegrina, 2007). Islamic banking is an
ethical banking system based on the principles of no interest and prot loss sharing contracts (see
Adebayo and Hassan (2013), Farook et al. (2012), Rashid et al. (2012)). Theoretically, the risk sharing
nature of the prot and loss principle in IBs should conceptually make them less risky, but the conceptual
theory is difcult to translate into the real world, due to market imperfections and information
asymmetries (Muljawan et al., 2004). These information problems can cause bank managers to react
heterogeneously, which could affect the stability, soundness, and efciency of the bank. Furthermore,
Hassan et al. (2011), Smolo and Hassan (2010), Hassan and Chowdhury (2010), Grais and Kulathunga
(2007), and Hassan and Dicle (2005, 2007) argue that CAR is an essential safety net for IBs, due to the
specic risks of their products and the nature of IBs as intermediaries. The importance of capital
requirements is still as poignant and signicant to IB performance as it is to CBs, and it deserves further
empirical study, particularly in countries with dual banking systems. Furthermore, due to the unique
principles upheld by IBs, it is important to contrast the impact of CAR between CBs and IBs with regard to
the capital crunch hypothesis, because the applicability of CAR is potentially different between the two
banking systems.
The rst objective of this paper is to investigate how deposit and loan growth react differently in CBs
and IBs. It is found that both IB and CB deposit and loan growth react positively with increased level of

60

M. Abdul Karim et al. / Pacic-Basin Finance Journal 28 (2014) 5875

capital. It is also found that credit creation by both types of banks is more inuenced by supply-side theory
implying more capital results in more deposit and loan growth. The second objective of this paper is to
examine whether capital crunch phenomenon is equally applicable to both CBs and IBs. It is found that
impact of bank capital changes is larger on lower capitalized banks than on higher capitalized banks.
This paper contains 5 sections. The rst section introduces the topic and related issues. This is followed by a
summary of the theoretical and empirical backgrounds in Section 2. Next, we discuss the data and methodology
in section 3. Section 4 presents the estimation results. The nal section concludes.
2. Literature review
2.1. Banking theory
Santomero (1984) uses three different approaches in explaining the existence of banks or nancial
institutions as an intermediary. The rst approach relates to the function of the bank itself, which is that of
asset transformation. The two distinct fundamental functions of banks are those of asset diversication
and the evaluation of the riskiness of nancial assets. The second approach dening the existence of banks
is explained using the demand deposit liability as the medium of exchange. The liability side of the bank
balance sheet explains the deposit-taking activities of the bank as well as its capital structure. The nature
of a bank's activities is to borrow money from depositors and use it to generate returns for the bank's
owners. The third explanation for a bank's existence by Santomero is from the perspective of integrating
the asset and liability side of the bank.
Information asymmetries have proven to be useful in describing banking theory in that banks exist to
minimize transaction costs. Depositors face high transaction costs and are unable to hold risky assets on their
own. In this case, banks are able to benet depositors by investing their wealth in assets in which the bank has
special knowledge (Leland and Pyle, 1977). Diamond and Dybvig (1983) and Gangopadhyay and Singh (2000)
focus on the role of banks through the process of liquidity transformation. Without banks as intermediaries,
individual investors would be trapped in illiquid investments. Campbell and Kracaw (1980) and Diamond and
Dybvig (1983) show nancial intermediaries as information specialists and cost delegators, where they
minimize the cost of information monitoring and facilitate investors in lending and borrowing. Mester (1992)
cites that, in order for a bank to be efcient, it must also undertake nontraditional activities such as loan selling
and buying, as opposed to the more traditional banking practices of originating and monitoring loans.
Freixas and Rochet (1997) classify bank functions into four categories: offering access to a payment
system; transforming assets; managing interest rates and liquidity risk; and processing information and
monitoring borrowers. Freixas and Rochet summarize that the functions of banks are considered
important, because of the demand for different monies, the demand for divisible, low risk and short term
liabilities, the demand for indivisible, risky and long term capital, and the demand for project monitoring.
Allen and Santomero (1997) analyze the role and function of banks as nancial intermediations in the
modern context, which has been neglected in the traditional banking theory. They argue that emphasis
should be given to the important roles of the bank relating to risk transfer and reducing participation costs
and not solely on the reduction of information asymmetry and transaction costs, which are stressed by
traditional theories. Banks issue demand deposits to borrowers, and this allows them to share the risks,
thus improving market competitiveness.
In Islamic banking, there is a lack of theory explaining the existence of Islamic banking within the industry.
Most theorists use the contemporary conventional banking system to explain the Islamic banking model.
Aggarwal and Yousef (2000) model Islamic banking from a prot sharing perspective and conclude that this
model is not being used widely, due to agency problems, and they conclude that the model will only work well
when moral hazard is low.
2.1.1. Bank capital management
Frost (2004) categorizes bank capital into four categories:
i Funding capitalcapital from the shareholders who require return on their capital
ii Risk capitalcapital available to absorb possible losses before the bank faces insolvency

M. Abdul Karim et al. / Pacic-Basin Finance Journal 28 (2014) 5875

61

iii Economic capitalcapital from the investment made by the shareholders in the business (share capital,
premium accounts and retained earnings)
iv Regulatory capitalcapital that needs to be held in order to protect depositors from losses resulting
from loan failure.
Management of capital includes all the choices that have to be made in order to ensure the best optimal
combination of capital instruments. Banks are among the most important institutions, as they provide liquidity
to the markets (Diamond and Rajan, 2000). The optimum level of capital allocated by the banks will take into
account the mandatory controls imposed by regulators, because banks are the most heavily regulated industry,
particularly with regard to the capital adequacy requirement. Even though the capital adequacy requirement is
set as the minimum ratio to be followed by the banks, a bank may optimally increase its capital ratios to give
assurance that it is in stable condition (Berger, 1995).
Lower capital ratios have been cited as the main reason for the decline in bank credit. However, this does not
mean that highly capitalized banks cannot be insolvent. Bank Islam Malaysia Berhad (BIMB), for example,
became insolvent in 2006 in spite of having a high CAR of 31% in 2004 (Chong and Liu, 2009). A high level of CAR
means that the banks have additional capital for additional investment. However, if the banks are not selective in
their investments and are not assessing the risk objectively, they may be exposed to more risk than they can
manage.
On the other hand, the minimum capital requirement has been blamed for the reduction in lending
activities. This results in the contraction of bank lending and a shrinkage in loan supply in the market, causing
loan growth to be low, which directly affects the performance of banks. In meeting capital requirements,
banks have to choose whether to raise capital, reduce the asset portfolio, or invest in less risky or zero risk
assets, such as government securities. In Malaysia during the 19971998 nancial crisis, for example, despite
the fact that the average risk-weighted capital ratio remained high at 11% for the whole banking sector, loan
growth was only 34% annually.2 This shows that banks were restricting their lending to ensure a high level of
capital or to maintain the minimum capital requirement of 8% in order to avoid the penalties imposed if the
requirement is not met.

2.2. Capital crunch hypothesis

The capital crunch always comes under the investigation of researchers when a crisis affects nancial or
economic stability on a national or international scale. The capital crunch hypothesis is also frequently cited,
along with the increasing burden of nonperforming loans (NPLs), as a possible explanation for retractions in
lending (Ferri and Kang, 1999).
Bank loan supply has shown to have a dramatic impact on real banking activities. van den Heuvel (2004) and
Gambacorta and Mistrulli (2004) suggest that a shortfall in the desired level of capital ratios may cause loan
supply to shift downward. Adrian and Shin (2008) suggest that a negative shock to capital leads to a downward
shift in loan supply, which forces banks to adjust their balance sheet items. Peek and Rosengren (1995) argue
that banks will face lending problems if their capital is declining, which also causes difculties in meeting capital
requirements. They describe this scenario as capital crunch, whereby the contraction in liabilities causes banks to
reduce their assets. In a study on behavior of New England banks, Peek and Rosengren conclude that these banks
experienced a capital crunch during the early 1990s recession. Banks that were poorly capitalized were
relatively more affected by the capital crunch than the better capitalized banks.
Banks in Korea lowered their loan investments and increased their investment in government securities in
order to fulll capital requirements during the Asian nancial crisis. Berger and Udell (1994) refer to this
scenario as a supply side credit crunch, where banks are not willing to lend, because of reductions in loan supply
due to the depletion of banking capital.
In order to meet the capital requirements determined by regulators, banks have the option to either increase
equity capital or shrink assets and liabilities. However, raising new equity capital is costly and later pushes the
banks to reduce assets and liabilities anyway, because banks often believe that the current capital adequacy ratio
will determine future bank lending (Bernanke and Lown, 1991).
2

Data taken from Bank Negara Malaysia.

62

M. Abdul Karim et al. / Pacic-Basin Finance Journal 28 (2014) 5875

Not all previous research agrees that the contraction in lending during adverse economic situations is caused
by a contraction in loan supply. Bernanke and Gentler (1987, 1989) conclude that the magnitude of the credit
slowdown during the 19901991 U.S. recession was not mainly caused by the demand factor alone, but that it
could also be linked to supply factors like the loss of bank capital. Additionally, Sharpe (1995) singles out the
tightening of bank regulatory standards and heightening market scrutiny of bank capital as the major factors
behind credit slowdowns.

3. Data and methodology

3.1. Data
The study of bank behavior involves two dependent variables, changes in loans and changes in
deposits, which are analyzed using the same independent variable, the capital adequacy ratio, and set of
control variables. Multivariate analysis is used in the estimation. Bank-specic and economic variables are
used in the estimation as control variables. Multivariate analysis is selected in order to avoid multiple
univariate tests on the variables. The two dependent variables are analyzed separately. The estimation
model used follows Peek and Rosengren (1995) and Chiuri et al. (2002), with some modication of the
variables.
The data involved in this study come from a cross-country sample from the Organization of Islamic
Conference (OIC) country members. There are 57 countries currently in the organization. However, of the 57
member countries, less than 20 provide Islamic banking services in their nancial systems. Therefore, this study
only considers countries that provide both Islamic and conventional banking services. The sample also excludes
countries that have adopted banking systems that consist only of Islamic banks such as Sudan, Iran, and Pakistan.
The reason for limiting the data in this manner is to provide a better comparative analysis of dual banking
systems, i.e. a comparison of the Islamic banking system with its conventional banking counterpart in Muslim
countries over a certain period. The main sources of the data are from the audited annual nancial reports of
each bank. Data on the banking annual reports are taken from the Bureau van Dijk Electronic banking database
(Bankscope), the Islamic Bank and Financial Institution Information (IBIS) online database, and the web sites of
each bank. In addition, data on macroeconomic and country specic variables are obtained from the World Bank
Development Indicators and the IFS.
Following the literature, this study utilizes a panel data methodology. Panel data is most appropriate in this
study, because it contains the necessary properties when dealing with inter-temporal dynamic and individual
bank behaviors. Unfortunately, some of the data for Islamic banks are missing and incomplete. Since missing
data is common in panel data, it is widely used in most of the Islamic banking literature. Panel data provides us
with several benets, such as the ability to control for individual heterogeneity, particularly when it involves
countries or time-invariant variables. Panel data is also more exible, as it provides more degrees of freedom, is
more informative, and there is less co-linearity among the variables. Panel data is also better for the study of the
dynamics of adjustments over a period of time, because it allows us to construct and test more complicated
behavioral models (Baltagi, 1995).

3.2. Period of study

Basel I was enforced in 1992 in the G10 countries and was later introduced in other regions of the world.
However, in most developing countries, its implementation has been much more recent. The U.A.E., Sudan,
Kuwait, Bahrain, Jordan and Malaysia are among the earliest countries that established Islamic banks, and were
among the rst to follow the Basel capital requirements. Due to differences in the implementation of the Basel
Accord in Islamic countries, which come mostly from developing and emerging markets, this research covers a
sample from 1999 through 2009. Since the establishment of IBs is different in most countries of the world, this
study faces limitations in terms of data availability. Due to data availability issues, an unbalanced panel
comprised of different groups across the countries is utilized over the entire sample period. Furthermore, an
unbalanced panel solves the issue of missing data in certain periods. The distribution of banks used in this study,
classied by region and country, is presented in Table 1.

M. Abdul Karim et al. / Pacic-Basin Finance Journal 28 (2014) 5875

63

Table 1
Classication of banks by region and countries.
Region

Country

African

Algeria
Egypt
Tunisia
Bahrain
Jordan
Kuwait
Qatar
Saudi Arabia
UAE
Yemen
Turkey
Bangladesh
Indonesia
Malaysia

Middle Eastern

South Asia

Total

Number of banks
Islamic banks

Conventional banks

Total

1
2
1
10
2
3
3
2
6
4
4
3
2
9
52

2
20
8
8
9
6
5
9
14
4
22
25
36
18
186

3
22
9
18
11
9
8
11
20
8
26
28
38
37
238

4. Empirical results
4.1. Descriptive analysis
The extreme outliers are trimmed from the data in order to improve the robustness of the ndings. The
nal panel data set used in this study includes a total of 52 Islamic banks and 186 conventional banks from
14 countries. The structure of the sample used in the panel is shown in Table 2, while Table 3 shows the
classication of the sample by country and bank.
The unbalanced panel data of 52 IBs and 186 CBs from 14 OIC countries is tabulated below in Table 3.
4.2. Mean value of CAR
This section provides the mean analysis of the CAR used to contrast CAR variance between IBs and CBs.
This study employs two different measurements of the capital adequacy ratio, which takes into account
differences in the denominator and numerator of the capital calculation. The two measurements used in
determining the capital requirement of the bank for this study are:
i Total equity capital over total asset (leverage ratio); and
ii Basel capital (Tier 1 and Tier 2) over risk weighted assets (risk-based capital ratio).

Table 2
Structure of panel data of the sample banks.
Record of years on each bank

3
4
5
6
7
8
9
10
11
Total

Number of banks

Number of observations

Conventional banks

Islamic banks

Total

Conventional banks

Islamic banks

Total

0
1
1
6
2
10
10
38
118
186

6
7
1
0
4
3
3
6
22
52

6
8
2
6
6
13
13
42
140
236

0
4
5
36
14
80
90
380
1298
1907

18
28
5
0
28
24
27
60
242
432

18
32
10
36
42
104
117
440
1540
2339

64

M. Abdul Karim et al. / Pacic-Basin Finance Journal 28 (2014) 5875

Table 3
Structure of sample by country and type of bank.
Country

Number of bank

Algeria
Bahrain
Bangladesh
Egypt
Indonesia
Jordan
Kuwait
Malaysia
Qatar
Saudi
Tunisia
Turkey
UAE
Yemen
Total

Total

Islamic bank

Conventional bank

3
18
28
22
38
11
9
27
8
11
9
26
20
8
238

2
9
3
2
2
2
3
9
3
2
1
4
6
4
52

1
9
25
20
36
9
6
18
5
9
8
22
14
4
186

The reason for using two different measurements of the capital adequacy ratio is to determine whether the
source of funds and risk elements play a role in determining the impact of capital adequacy on bank behavior.
The mean CAR calculated for all IBs and CBs for each year is shown in Fig. 1. The graph shows that both CB
and IB CARs fall between 15% and 22%. These ratios are far above the minimum requirement of 8% imposed by
the Basel Accord. The graph also indicates a similarity in the trends of the mean ratios of the IBs and the CBs,
implying no signicant difference in the mean CAR between the two banking systems. This conclusion is also
supported by an independent, paired t-test, which shows insignicant results. The test fails to reject the null
hypothesis that there is no signicant difference between the two groups (p-value of 0.38501 and t value of
0.77002). This nding is not consistent with the notion that the risk-weighted ratio for IBs must be higher
than CBs in order to account for the expected additional investment risk assumed by IBs in their businesses.
There are three additional tests presented in this section. The rst test includes all 52 IBs and 186 CBs.
The banks are divided into two sub-categories, based on their capitalization level. To account for
differences between the higher and lower capitalized banks for each country, this study applies the capital
ratio mean values for every country as a baseline. The capital ratio mean value of all samples used is shown
in Table 4 below.
The mean value is used as the baseline, because this study deals with panel data across countries that have
different economic backgrounds and implementations of the Basel Accord. The nature of the economic

25
20
15
Mean CAR IBs

10

Mean CAR CBs

5
0

Fig. 1. Plot graph of mean CAR for CBs and IBs in the selected countries.

M. Abdul Karim et al. / Pacic-Basin Finance Journal 28 (2014) 5875

65

Table 4
Descriptive data on mean value of leverage ratio and risk weighted ratio on each country.
Countries/descriptive data

Conventional banks

Islamic banks

Leverage ratio (CR)

Risk-weighted ratio (CAR)

Leverage ratio (CR)

Risk-weighted ratio (CAR)

Algeria
Mean
Max
Min

11.5419
25.82
4.61

17.5815
25.8
12.1

10.0882
15.81
7.14

14.3570
24
10.72

Bahrain
Mean
Max
Min

15.0481
45.19
4.22

22.1629
57.8
9.7

25.5599
88
9.11

28.0790
65.4
11

Bangladesh
Mean
Max
Min

6.6135
24.85
1.87

11.545
36.68
2.6

6.115
9.83
2.76

9.1456
13.69
4

Egypt
Mean
Max
Min

9.7245
23.75
3.08

16.7914
43.9
6.91

5.1691
8.96
2.97

12.1391
18.08
8.11

Indonesia
Mean
Max
Min

12.5907
64.22
1.3

21.5931
79.5
4

11.3165
43.71
5.6

16.3420
63.18
8.95

Jordan
Mean
Max
Min

11.7095
20.66
4.49

18.7487
32.3
8.1

10.6777
22.89
5.17

14.1591
18.25
11.15

Kuwait
Mean
Max
Min

11.8873
16.61
7.7

16.1352
26.1
5.2

15.1354
23.24
9.24

19.9965
32.1
13.6

Malaysia
Mean
Max
Min

9.51
33.65
3.57

16.4955
55.4
3.6

8.2965
19.79
3.19

15.2984
23.27
8.8

Qatar
Mean
Max
Min

13.7539
25.59
7.01

20.5671
43
11.9

16.6672
43.26
7.07

18.3152
46.7
11.67

Saudi
Mean
Max
Min

11.7394
27.05
2.95

18.2152
41.5
11.5

16.2473
26.81
11.2

26.7467
54
18.4

Tunisia
Mean
Max
Min

9.4552
17.48
3.3

12.3186
22.22
6.45

25.3736
33.95
15.71

21.2436
27.9
9.85

Turkey
Mean
Max
Min

13.4665
52.75
2.95

20.11
67
5.91

10.24054
17.36
3.2

14.3038
21.7
9
(continued on next page)

66

M. Abdul Karim et al. / Pacic-Basin Finance Journal 28 (2014) 5875

Table 4 (continued)
Countries/descriptive data

Conventional banks

Islamic banks

Leverage ratio (CR)

Risk-weighted ratio (CAR)

Leverage ratio (CR)

Risk-weighted ratio (CAR)

UAE
Mean
Max
Min

17.7345
36.77
7.99

20.6065
42.9
10.8

17.7854
39.76
6.34

20.1606
47.2
7.03

Yemen
Mean
Max
Min

7.3516
64.20
3

22.02
79.5
8.61

12.5279
88
4.77

15.0260
65.4
8.75

background of each country will affect the ratio applicable to each country. Since the economic background is
different among the countries, the qualities of deposits and loans of the banks in each country as well as their
regulatory environments will also be different. Therefore, to reduce any biases involved in the study, the mean
value of each country is used to generalize the level of capitalization for all banks in that country. Any banks that
record a capital ratio below the mean value of the country are categorized as lower capitalized, while those that
record their capital ratio above the mean value are considered highly capitalized.
It is noted from Table 4 that most countries in this sample are holding a larger amount of capital than
necessary, which can be seen from the very high CAR and CR ratios (the highest is 79.50 for CAR and 64.22 for
CR). This result is consistent with Haselman and Wachtel's (2007) nding that the higher CARs in most
developing countries are intended to compensate for higher market and economic risks in the banking systems
of these economies.
4.3. Overall description for all variables
The overall description of variables used in this study is presented in Tables 5 and 6. Table 5 presents the
descriptive statistics for both leverage and risk weighted ratios of IBs and Table 6 shows the similar descriptive
statistics of CBs. There are 28 lower capitalized IBs and 24 highly capitalized IBs, and they are equally distributed
for both risk weighted and leverage ratio measurements. From Table 6, there are 106 lower capitalized CBs and
80 highly capitalized CBs from the risk-weighted category.
Deposit growth is higher than loan growth for lower capitalized IBs as well as for all CBs. However, in column
2 of Table 5, deposit growth is higher than loan growth. There are similar ndings with regard to the highly
capitalized IBs in column 10 of Table 5. The higher growth in loans compared with that of deposits for IBs can be
explained by unrestricted investment. Despite the fact that the unrestricted investment account constitutes a
majority of the deposits for IBs, there is no clear standard on how to treat this account on the balance sheet, i.e.
whether to record it as an on-balance sheet or off-balance sheet item. Since some banks did not record the
unrestricted investment account on the balance sheet, the calculation of total deposit growth may be biased and
could eventually show lower growth, compared to that of loans.
Both IBs and CBs in the tables below show higher loan growth in highly capitalized banks than that of lower
capitalized banks. It can be interpreted that higher capitalized banks are providing more loans than lower
capitalized banks. Another interesting point from the tables is that equity growth for IBs is higher than that of
CBs (column 2, row 9 of Tables 5 and 6). This can be explained by the prot sharing principle used in IBs through
the mudarabah account.
4.4. Regression estimates
This study denotes the cross sectional bank sample by i = 1,2,3,,N and the time series period by t =
1, 2, 3,,T. The total T is equal to 11 and it is relatively small, compared with the larger N of 52 IBs and 186
CBs for the unbalanced panel. In order to determine whether a pooled or panel data regression model is
most appropriate, a BreuschPagan test is conducted. The p-value rejects the null hypothesis of a pooled
regression; thus, the estimation is conducted using a panel data model.

Variables

All banks

Islamic banks

Mean

Dependent variables
Deposits
0.1806
Loans
0.2055
Independent variables
CR
15.3379
CMCR
1.3915
CAR
18.5215
CMCAR
1.6266
EQT
0.0505
SIZE
20.8926
LQDT
0.3176
FEE
0.3507
INFL
6.2561
LnGDP
0.0554
Observation
432

Low capitalized banks

High capitalized banks

Max

Min

Std Dev.

Mean

Max

Min

Std Dev.

Mean

Max

Min

Std Dev.

3.5641
6.4560

0.8629
0.5632

0.3207
0.4570

0.1655
0.1528

3.5641
3.3123

0.6780
0.4692

0.2924
0.2689

0.2043
0.2884

2.2810
6.4560

0.8629
0.5632

0.3604
0.6433

88.0000
79.4672
65.4000
126.7457
1.6009
24.5428
2.0540
3.6228
64.8681
0.2371

2.7600
9.1037
4.0000
6.7254
0.1797
17.5650
0.0011
7.4963
4.8632
0.0587

11.7428
5.8407
9.7533
7.8290
0.1344
1.3891
0.2221
0.4901
8.2226
0.0403

11.8172
0.4297
15.4188
0.6536
0.0330
20.9680
0.2606
0.3459
6.8101
0.0530
264

33.9500
17.7174
32.2700
59.5272
1.0267
24.5428
0.8561
2.3429
64.8681
0.2371

2.7600
1.5662
4.0000
1.5469
0.1797
17.5650
0.0099
7.4963
4.8632
0.0587

6.6287
1.6177
5.7216
3.7935
0.0827
1.4788
0.1774
0.5551
9.5621
0.0400

22.0469
2.9699
23.9585
3.1646
0.0780
20.7742
0.4073
0.3582
5.3856
0.0591
168

88.0000
79.4672
65.4000
126.7457
1.6009
23.5826
2.0540
3.6228
44.9641
0.2371

4.7700
9.1037
7.0300
6.7254
0.1531
17.7046
0.0011
0.5973
4.8632
0.0504

15.3019
8.9873
12.3084
11.4761
0.1861
1.2301
0.2539
0.3668
5.4087
0.0405

Notes: Deposits, Loans CR, CAR, CMCR, CMCAR, EQT, SIZE, LQDT, FEE, INFL, and LnGDP refer to change in deposits, change in loans, capital adequacy ratio (using leverage ratio), capital
adequacy ratio (using risk-weighted ratio), interactive terms for CR (CR EQT/TAt 1), interactive term for CAR (CAR EQT/TAt 1), change in equity, bank size, liquidity ratio, fee income, and
ination rate and change in real GDP.

M. Abdul Karim et al. / Pacic-Basin Finance Journal 28 (2014) 5875

Table 5
Descriptive data for variable used in Islamic banks (using both leverage and risk weighted capital ratios).

67

68

Variables

All banks

Low capitalized banks

High capitalized banks

Conventional banks

Mean

Max

Min

Std Dev.

Mean

Max

Min

Std Dev.

Mean

Max

Min

Std Dev.

Dependent variables
Deposits
Loans

0.1418
0.1055

14.7270
2.8929

0.7686
0.5311

0.4146
0.1955

0.1286
0.0988

3.2930
2.8351

0.7687
0.5311

0.2483
0.1877

0.1605
0.1154

14.7270
2.8930

0.4200
0.3588

0.5655
0.2053

Independent variables
CAR
CMCAR
EQT
SIZE
LQDT
FEE
INFL
LnGDP
Observation

18.3772
0.4124
0.0224
21.4397
0.3805
0.3573
7.3005
0.0511
1907

99.4
34.4764
1.8334
25.1613
1.3270
50.7826
64.8681
0.2371

2.6
8.7567
0.7952
17.4054
0.0274
38.0000
4.8632
0.0587

10.0089
1.1717
0.0681
1.6803
0.1826
1.4753
8.8020
0.0334

13.9962
0.2389
0.0157
21.7878
0.3607
0.3326
7.1506
0.0509
1093

53.2000
8.1745
0.4058
25.0730
0.9266
3.1854
64.8681
0.2371

2.6000
8.6127
0.7953
17.5064
0.0438
38.0000
4.8632
0.0587

4.4620
0.6719
0.0464
1.5713
0.1676
1.1919
8.5241
0.0324

23.4372
1.1930
0.0364
20.7108
0.3974
0.4077
7.1315
0.3150
814

79.5000
67.0000
1.8335
25.1613
1.3270
50.7826
64.8681
20.3851

4.0000
8.7568
0.1813
0.0968
0.0107
2.5894
4.8632
0.0587

11.1541
5.3528
0.1076
2.9439
0.1979
1.8006
8.5394
2.2530

Notes: Deposits, Loans CR, CAR, CMCR, CMCAR, EQT, SIZE, LQDT, FEE, INFL, and LnGDP refer to change in deposits, change in loans, capital adequacy ratio (using leverage ratio), capital
adequacy ratio (using risk-weighted ratio), interactive terms for CR (CR EQT/TAt 1), interactive term for CAR (CAR EQT/TAt 1), change in equity, bank size, liquidity ratio, fee income, and
ination rate and change in real GDP.

M. Abdul Karim et al. / Pacic-Basin Finance Journal 28 (2014) 5875

Table 6
Descriptive data for variable used in conventional banks (using both leverage and risk weighted capital ratios).

M. Abdul Karim et al. / Pacic-Basin Finance Journal 28 (2014) 5875

69

4.4.1. Estimated Generalized Least Square (EGLS) model

Table 7 shows the regression results for all 52 IBs and 186 CBs after accounting for all control variables.
The table presents the results using both change in deposits and change in loans as the dependent
variables. The two different methods of computing the capital ratio are also employed.
The CARs used for this study are taken from the previous years' ratios and not the current years' ratios. This is
due to the assumption that the decisions made concerning the current year's capital ratio were determined in
the previous period. This delay is due to a signicant time lag in loan and deposit growth. The banks, therefore,
raise capital in anticipation for future lending or borrowing. The banks' internal variables used in this study are
also lagged one period in order to avoid potential endogeneity bias.
The estimates from the EGLS Fixed Effect Model using White's correction presented in Table 8 are improved
after correcting for heteroskedasticity and autocorrelation problems, as evidenced by improved DW (ranged
between 1.90 and 2.29) and R-squared (ranged between 38% and 80%) values for both CBs and IBs.
In order to get a clearer picture of the impact of higher or lower capitalized banks and their behavior, it is
better to segregate the sample into two groups using the capitalization mean value of each country. Table 8
provides the results for the lower-capitalized banks for both IBs and CBs.
A test for the impact of CAR on highly capitalized banks is conducted using the same EGLS Fixed Effect Model
used in previous regression, and the results are presented in Table 9.
Table 7
Summary of EGLS regression model (for all banks with bank specic and macroeconomic factor).
Indp
variables

Dependent variables
Conventional banks
Leverage ratio

C
CR(1)
CMCR

LQDT(1)
FEE(1)
INFL
GDP
R2
Adj R2
F Stat
DW

Risk-based
capital ratio

Deposits

Loans

Deposits

Loans

Deposits

Loans

Deposits

Loans

0.1750
(1.9481)
0.0044
(6.8398)
0.0214
(2.0570)

0.2329
(1.7674)

0.1477
(1.6291)

0.9361
(4.3763)
0.0036
(3.0264)
0.0007
(0.1237)

0.8028
(4.3884)
0.0053
(5.3979)
0.0281
(6.7722)

0.9472
(4.0745)

0.7275
(3.9156)

0.0026
(4.4858)
0.0205
(2.6789)
0.4988
(3.0043)
0.0071
(1.1879)
0.0870
(2.9505)
0.0022
(1.1983)
0.0013
(2.9608)
0.4732
(6.4721)
0.3527
0.2704
4.2862
2.0369

0.0009
(2.8049)
0.0210
(3.4775)
0.3682
(2.5373)
0.0073
(1.7574)
0.0907
(5.0468)
0.0044
(2.0554)
0.0016
(6.0915)
0.7283
(12.0388)
0.4699
0.4025
6.9726
1.9108

0.0009
(1.1167)
0.0067
(1.0905)
0.1739
(0.7032)
0.0384
(3.4383)
0.1634
(4.1624)
0.0099
(2.6237)
0.0027
(1.6374)
0.2804
(1.8191)
0.5453
0.4612
6.4844
2.0616

0.0038
(5.2431)
0.0183
(3.5331)
1.2729
(6.7480)
0.0369
(4.1693)
0.1661
(3.6659)
0.0069
(3.7678)
0.0051
(3.8625)
0.2603
(2.4167)
0.7941
0.7561
20.8572
2.2937

CMCAR

SIZE(1)

Leverage ratio

0.2826
(2.2040)
0.0094
(9.6668)
0.0089
(0.6848)

CAR(1)

EQT

Islamic banks
Risk-based
capital ratio

0.8171
(3.5309)
0.0130
(2.2216)
0.0298
(1.0576)
0.0017
(0.9067)
0.0013
(3.0154)
0.3934
(5.5842)
0.3769
0.2977
4.7585
2.0145

0.4954
(2.7912)
0.0064
(1.5519)
0.1376
(8.2054)
0.0044
(2.1208)
0.0016
(7.2876)
0.6387
(11.5933)
0.4843
0.4187
7.3852
1.9037

0.4518
(2.0182)
0.0392
(3.7952)
0.1948
(5.1423)
0.0106
(2.5866)
0.0027
(1.7259)
0.1848
(1.2452)
0.6065
0.5337
8.3342
2.0993

1.5588
(8.8386)
0.0422
(4.8842)
0.2330
(6.8625)
0.0054
(3.0331)
0.0054
(4.2691)
0.2226
(2.1491)
0.8195
0.7861
24.5402
2.2224

Number in parentheses shows the t value.

CR, CAR, CMCR, CMCAR, EQT, SIZE(1), LQDT(1), FEE, INFL, and GDP are referring to capital adequacy ratio (using leverage ratio),
capital adequacy ratio (using risk-weighted ratio), interactive terms using leverage ratio, interactive terms using risk-weighted ratio,
change in equity, one year lag of bank size, one year lag of liquidity, fee income, ination rate and change in real GDP.
Signicance at 1%.
Signicance at 5%.
Signicance at 10%.

70

M. Abdul Karim et al. / Pacic-Basin Finance Journal 28 (2014) 5875

Table 8
Summary of EGLS regression model (for low capitalized banks with bank specic and macroeconomic factor).
Indp variable

Dependent variables
Conventional banks
Leverage ratio

C
CR(1)
CMCR

Risk-based capital ratio

Leverage ratio

Risk-based capital ratio

Deposits

Loans

Deposits

Loans

Deposits

Loans

Deposits

Loans

0.6697
(3.9379)
0.0128
(9.0091)
0.0713
(2.6333)

0.3383
(2.4890)
0.0077
(6.4914)
0.0975
(5.0901)

0.5188
(2.9935)

0.231
(1.7163)

1.316
(5.367)
0.010
(5.001)
0.064
(4.34)

0.651
(3.617)
0.0051
(3.9260)
0.040
(2.466)

1.145
(4.898)

0.692
(3.509)

0.0039
(3.0690)
0.0536
(3.3183)
0.2973
(1.2040)
0.0205
(2.6451)
0.103
(2.597)
0.0062
(2.2606)
0.0014
(2.1402)
0.5088
(5.2593)
0.3162
0.2271
3.5483
2.0309

0.0032
(3.1045)
0.0469
(3.5454)
0.1665
(1.0676)
0.011
(1.7561)
0.051
(1.8741)
0.0005
(0.2642)
0.0022
(3.6946)
0.7504
(8.2178)
0.3883
0.3085
4.8700
1.8438

0.0073
(3.7167)
0.0459
(4.1465)
1.5995
(6.0283)
0.0539
(4.6873)
0.1084
(3.0917)
0.0088
(2.2711)
0.0024
(1.6569)
0.0441
(0.2380)
0.5200
0.4356
6.1590
1.8562

0.0004
(0.2403)
0.0036
(0.2122)
0.9322
(3.1788)
0.0308
(3.2175)
0.0319
(0.8809)
0.0071
(3.7604)
0.0031
(1.6901)
0.4420
(2.9646)
0.7509
0.7071
17.1388
2.4212

CAR(1)
CMCAR
EQT
SIZE(1)
LQDT
(1)
FEE(1)
INFL
GDP
R2
Adj R2
F Stat
DW

Islamic banks

0.4466
(1.5386)
0.0305
(3.9384)
0.0430
(1.2634)
0.0066
(2.2119)
0.0010
(1.5205)
0.137
(1.649)
0.3691
0.2867
4.4783
1.9587

0.1814
(1.3094)
0.016
(2.5906)
0.0802
(3.6035)
0.0001
(0.0448)
0.0014
(2.8969)
0.1333
(1.5305)
0.4148
0.3384
5.4270
1.8086

1.712
(6.151)
0.063
(5.08)
0.041
(1.28)
0.008
(2.19)
0.002
(1.60)
0.069
(0.421)
0.491
0.402
5.490
1.823

1.5634
(5.4243)
0.0314
(3.5565)
0.0155
(0.9820)
0.0060
(3.1424)
0.0033
(2.0096)
0.3977
(2.9981)
0.7557
0.7127
17.5881
2.4015

Number in parentheses shows the t value.

CR, CAR, CMCR, CMCAR, EQT, SIZE(1), LQDT(1), FEE, INFL, and GDP refer to capital adequacy ratio (using leverage ratio), capital
adequacy ratio (using risk-weighted ratio), interactive terms using leverage ratio, interactive terms using risk-weighted ratio,
change in equity, one year lag of bank size, one year lag of liquidity, fee income, ination rate and change in real GDP.
Signicance at 1%.
Signicance at 5%.
Signicance at 10%.

4.5. Relationship between capital requirement and bank behavior

Berger and Udell (1994) refer to banks' shrinking their lending as a supply side problem, whereby banks
are not willing to lend, because of a reduction in the loan supply caused by reductions in their capital.
Therefore, it is expected that deposits and loans react in accordance with an increase in capital.
The importance of capital supply is tested using the t-test analysis presented in Tables 7, 8 and 9. The results
show that deposit and loan growth in both IBs and CBs react positively to changes in capital (CAR or CR). Most of
the coefcients for 1 and 1 for Deposits and Loans are signicant and positively associated with a change in
CAR.
The signicant, positive coefcients on Deposits and Loan provide evidence that bank loans react in the
same manner as bank capital, and this supports the supply-side theory that credit behavior is particularly
inuenced by the level of bank capital. The positive coefcient on Deposits suggests that an increase in bank
deposits is also inuenced by changes in CAR. This result implies that deposits are the main source of funding for
the banks. The positive, signicant impact of CAR on lending and borrowing behavior of banks is consistent with
the documented evidence [Peek and Rosengren (1995), Chiuri et al. (2002), Yudistira (2002) and Schmitz
(2007)].

M. Abdul Karim et al. / Pacic-Basin Finance Journal 28 (2014) 5875

71

Table 9
Summary of EGLS regression model (for high capitalized banks with bank specic and macroeconomic factor).
Indp
variables

C
CR(1)
CMCR

Dependent variables
Conventional banks
Leverage ratio

Risk-based capital ratio

Leverage ratio

Risk-based capital ratio

Deposits

Loans

Deposits

Loans

Deposits

Loans

Deposits

Loans

0.0383
(0.190)
0.007
5.875
0.009
(0.07)

0.4806
(3.540)
0.0045
5.0638
0.020
(1.776)

0.1738
(0.849)

0.5600
(4.265)

1.1675
(3.0749)
0.0035
2.722
0.0033
(1.569)

0.9583
(2.052)
0.0040
(3.008)
0.0038
(8.328)

1.4271
(3.755)

1.0380
(2.385)

0.0025
(3.7843)
0.0193
(5.9212)
0.2400
(4.794)
0.0057
(0.6037)
0.087
(1.919)
0.0019
(1.119)
0.0003
(0.272)
0.4226
(4.865)
0.404
0.323
5.014
1.988

0.0013
(3.1709)
0.0213
(5.9196)
0.1758
(2.532)
0.0181
(3.018)
0.1081
(3.853)
0.0074
(5.6141)
0.0004
(0.725)
0.5852
(7.289)
0.545
0.483
8.846
1.937

0.001
(0.565)
0.0028
(10.559)
0.217
(1.479)
0.0630
(3.5471)
0.1507
(2.0841)
0.0720
(1.2691)
0.0088
(3.168)
0.3349
(1.122)
0.622
0.517
5.945
2.406

0.004
(4.112)
0.0013
(3.943)
0.7502 (6.049)

CAR(1)
CMCAR
EQT
SIZE(1)
LQDT(1)
FEE(1)
INFL
GDP
R2
Adj R2
F Stat
DW

Islamic banks

0.835
(3.16)
0.0025
(0.27)
0.019
(0.43)
0.006
(0.38)
0.0002
(0.199)
0.395
(3.679)
0.398
0.318
4.915
1.986

0.3996
(2.0055)
0.0189
(3.019)
0.1897
(5.793)
0.0062
(4.932)
0.0002
(0.506)
0.6208
(6.710)
0.545
0.483
8.868
1.971

0.1821
(1.185)
0.0541
(3.053)
0.156
(2.08)
0.104
(1.862)
0.009
(3.326)
0.4004
(1.306)
0.534
0.405
4.144
2.441

0.750
(6.69)
0.05
(2.57)
0.375
(4.112)
0.0272
(0.938)
0.014
(7.216)
0.2689
(1.249)
0.882
0.849
27.126
2.126

0.0607 (2.93)
0.3823 (3.799)
0.092 (1.514)
0.015 (6.681)
0.2251 (0.847)
0.746
0.676
10.622
2.178

Number in parentheses shows the t value.

CR, CAR, CMCR, CMCAR, EQT, SIZE(1), LQDT(1), FEE, INFL, and GDP refer to capital adequacy ratio (using leverage ratio), capital
adequacy ratio (using risk-weighted ratio), interactive terms using leverage ratio, interactive terms using risk-weighted ratio,
change in equity, one year lag of bank size, one year lag of liquidity, fee income, ination rate and change in real GDP.
Signicance at 1%.
Signicance at 5%.
Signicance at 10%.

4.6. Capital crunch hypothesis

Kim et al. (2002) dene a capital crunch as a serious lack of liquidity in the market due to banks'
avoidance of credit extension in order to satisfy capital requirements. When there is a shortage of bank
capital, there is a leftward shift in bank credit supply, which subsequently causes a contraction of bank
lending. Peek and Rosengren (1995) report that the capital crunch in the U.S. during the nancial crisis of
1990 and 1991 was caused by capital requirement constraints. Similarly, Chiuri et al. (2002) blame the
contraction in bank credit on higher minimum capital requirements imposed by regulations.
In order to identify whether the existence of a capital crunch explains the behavior of a bank, we test whether
the impact of changes in bank capital has a larger impact on lower capitalized banks than higher capitalized
banks. The results are inconclusive for both IBs and CBs, as evidenced by the interaction terms in rows 4 and 6 of
Tables 7, 8, and 9. The positive coefcient on the interaction term for CBs, for all banks, lower capitalized banks,
and higher capitalized banks species that these banks are not constrained by capital requirements, because the
overall capital ratio of most banks is much higher than the required level. Additionally, even though there is a
negative relationship with Deposits for highly capitalized CBs, it is not signicantly different from zero.
Therefore, the capital crunch hypothesis is not supported for CBs in the sample.

72

M. Abdul Karim et al. / Pacic-Basin Finance Journal 28 (2014) 5875

The negative coefcient for IBs (all banks and lower capitalized banks), on the other hand, supports the
capital crunch hypothesis in that the level of capitalization affects the banks' supply of loans. This result shows
that IBs are more capital constrained, particularly when capital is low and the banks have to control the riskiness
of their portfolios. However, the higher capitalized banks show a positive coefcient on both Deposits and
Loans. This result suggests that, when banks are well capitalized, they are not constrained by regulators' CAR
requirements, and their lending and borrowing are not impacted by drastic changes in the level of capital.
Overall, it can be concluded that the capital crunch hypothesis is not supported for CBs, because most of the
banks' capital is well above the minimum capital requirements. The inability of capital crunch theory to fully
explain these banks' behaviors can be explained by the economic landscape of most of the countries in the
sample. During the period of 1999 to 2009, most of the OIC countries in the sample did not experience any
serious internal economic turmoil or systemic banking problems.
From another perspective, the results show that the lending and borrowing of IBs can be explained by
the capital crunch hypothesis. This shows that stricter capital enforcement is important for the Islamic
banking system, because they are more sensitive to changes in the quality of their portfolio. IBs provide
not only commercial banking services, but also investment banking services based on prot-sharing
investment accounts.
4.7. The impact of other bank specic and macroeconomic variables
Bank borrowing and lending activities do not seem to be affected only by the supply of credit, but also by the
demand for credit. Therefore, this study also uses individual bank specic variables that directly inuence loan
supply. To control for the demand shocks, we also include each country's monetary policy and GDP changes.
The results from columns 69 of Tables 7, 8 and 9 show a signicant negative effect of size on the IBs'
Deposits and Loans. The negative and signicant coefcient on bank size on Deposits can be interpreted as
larger banks holding a relatively small share of their assets in the form of loans. They may attract a relatively
larger share of non-deposit or wholesale funding as their main source of short-term funding. This provides
evidence that IBs in the selected sample are well diversied in their assets and are not dependent on their
deposits and lending only. The highly capitalized IBs consistently support the theory that, as they grow larger,
their assets become more diversied. This is consistent with the fact that IBs provide a variety of Islamic
nancing products such as Murabahah and Bai Bithamin Ajil. The negative and signicant coefcient for the SIZE
variable, proxied by the log of assets, is conrmed by Peek and Rosengren (1995), Schmitz (2007) and Kunt and
Huizinga (2011). They conclude that deposit and loan growth are slower for larger banks.
However, there is mixed reaction between Deposits and Loan for CBs. Most of the results show a negative
relationship between bank size and deposit and loan growth. As an exception, a positive relationship is found
between loan changes and RWA for all banks as well as the leverage ratio for highly capitalized CBs. The positive
coefcient can be explained by the too big to fail hypothesis, whereby higher capitalized CBs will provide more
lending as they grow larger.
The liquidity variable (LQDT) is used as a control variable by showing that bank behavior is driven by the level
of capital and not by the bank's liquidity position. Therefore, it is expected to have a negative impact on bank
deposits and a positive impact on loan growth for both bank types. Both groups show that their behavior is
driven by the level of capital, because both groups show a negative coefcient for Deposits and a positive
coefcient for Loan. The negative impact can also be explained by the fact that banks will favor using liquid
assets over their liabilities (deposits) in order to meet capital requirements.
Another control variable is fee income (FEE), which is represented by the ratio of non-interest operating
income to the sum of net interest revenue and other operating income. The FEE variable shows that Deposits is
positively affected by the proportion of other bank earnings for both IBs and CBs. A signicant, positive impact of
FEE on the growth of deposits can be explained by bank stability. Banks with positive deposit growth are likely
less credit constrained and are, thus, in a better position to explore other off-balance sheet activities, compared
with credit constrained banks, which may be limited in pursuing other operating income.
Bank lending is expected to be procyclical and is very likely to be correlated with economic activities
(Goodhart et al., 2004). This study addresses this issue in the context of the impact of capital adequacy
requirements on the procyclical characteristics of the banks. A signicant positive impact shows the potential
growth of the economy. This indicates that the country's economic growth plays a role in the borrowing and
lending activities of the banks, which will lead to further potential increases in the supply of bank credit and loan

M. Abdul Karim et al. / Pacic-Basin Finance Journal 28 (2014) 5875

73

demand. A procyclical effect that causes nancial institutions to tighten credit in downturns and ease credit in
booms is justied by the positive impact it has towards changes in real GDP. During times of slower economic
growth, banks experience a slower growth in bank lending, due to lower interest rates as well as constraints in
capital. However, GDP growth does not have an impact on the deposit and loan growth of higher capitalized IBs.
A well-capitalized bank will have more opportunities to expand their portfolio and are less vulnerable to any
changes in the macroeconomic situation.
Another macroeconomic variable used in this study is monetary policy, which is proxied by the ination rate
(INFL). High ination is usually caused by excess supply of money in the market and low interest rates. The effect
of INFL on bank deposits and loan growth depends on whether it is anticipated or not, or how quickly banks react
to the ination rate (Revell, 1979). Under anticipated conditions, banks will adjust prices and interest rates,
because lenders insist on higher interest rates to offset the decline in purchasing power over the life of the loans.
At the same time, depositors will reduce their savings. Therefore, a positive relationship is expected with loan
growth, and a negative relationship with deposit growth is expected.
However, under unanticipated conditions, INFL is expected to have a negative impact on loan supply, because
the uncertainty makes banks more cautious and conservative in providing loans. On the other hand, INFL is
expected to have a positive relationship on deposit growth.
The impact of INFL on both IBs and CBs in this study is mixed, as shown in Tables 7, 8, and 9. There is a mostly
signicant, positive impact of INFL on Deposits and Loan for IBs in all three tests. The positive impact on Loan
suggests that, if an increase in ination is anticipated, banks adjust their interest rates to support a decreasing
rate of growth. On the other hand, the positive impact of INFL on Deposits suggests that depositors keep their
money in IBs for investment purposes, with the expectation that they will get higher returns when the banks
increase lending.
However, INFL has a signicant, negative impact on Deposits and Loan in CBs, as shown in Tables 7 and 8.
This result can be interpreted as CBs tightening their nancing (loans) and supply of funds (deposits), because of
the uncertainty of adverse price movements. The negative impact to loan growth also suggests that most of the
nancing in CBs consists of long-term loans which are at a xed rate and cannot be adjusted quickly enough,
despite the fact that ination is expected to be high.
4.8. Comparative analysis between Islamic and conventional bank
One of the important differences between IBs and CBs is the element of interest, because IBs offer
non-interest bearing products, and they also share prots according to an agreed ratio. Tables 7, 8 and 9 illustrate
the differences in the reaction of deposit and loan growth between IBs and CBs.
One of the obvious differences between the two groups of banks is the adjusted R-squared in the estimation
model for IBs and CBs. The huge difference between the two groups may be caused by other unknown variables,
which may better explain CB lending behavior, that have not been taken into account in this study.
The independent t-test of mean CAR between IBs and CBs provides a p-value of 0.38501 and two tail t value
of 0.77002. This shows that there is no signicant difference between the two groups. Based on this result, it can
be concluded that the level of capitalization in IBs is similar to that of CBs. Additionally, we can reject the idea
that IBs need higher CAR, compared with CBs.
5. Summary and conclusions
This study investigates bank behavior, particularly the deposit and lending activities of CBs and IBs. The
results show that bank capital requirements have an impact on the deposit and lending behaviors of the 186
conventional banks and 52 Islamic banks in the sample. There is a strong positive relationship between bank
capital and deposit and loan growth for both IBs and CBs. However, using interaction terms, we nd mixed
results when testing the impact of capital requirements. Even though the tests show the importance of capital
supply to bank borrowing and lending activities, the capital crunch hypothesis appears only relevant to IBs. The
negative relationship with capital requirements supports the idea that, when regulators impose higher
capitalization requirements, the assets and liabilities of lower capitalized IBs will suffer greater contraction. The
ndings also show a positive relationship in all three tests of CBs. This shows that most CBs are typically not
constrained by capital requirements, because their CAR is higher than the minimum requirement.

74

M. Abdul Karim et al. / Pacic-Basin Finance Journal 28 (2014) 5875

References
Adebayo, R.I., Hassan, M.K., 2013. Ethical principles of Islamic nancial institutions. Journal of Economic Development and
Cooperation 34 (2).
Adrian, T., Shin, H.S., 2008. Financial intermediaries, nancial stability & monetary policy. FRB Kansas City Working Paper Prepared
for FRB Kansas City Symposium at Jackson Hole.
Aggarwal, R.K., Yousef, T., 2000. Islamic bank and investment nancing. J. Money Credit Bank. 32 (1), 93120.
Allen, F., Santomero, A.M., 1997. The theory of nancial intermediation. Journal of Banking and Finance 21, 14611485.
Athanasoglou, P.P., Brissimis, S.N., Delis, M.D., 2008. Bank-specic, industry-specic and macroeconomic determinants of bank
protability. Journal of International Financial Markets Institutions and Money Institutions and Money 18 (2), 121136.
Baltagi, B.H., 1995. Econometric Analysis of Panel Data. John Wiley & Sons Ltd., West Sussex, England.
Barrell, R., Gottschalk, S., 2006. "The Impacts of Capital Adequacy Requirements on Emerging Markets," NIESR Discussion Papers 269.
National Institute of Economic and Social Research.
Berger, A.N., 1995. The role of capital in nancial institution. Journal of Banking and Finance 19 (34), 393430.
Berger, A.N., Udell, G.F., 1994. Did risk-based capital allocate bank credit and cause a credit crunch in the U.S.? Journal of Money,
credit and Banking 26 (3), 585628.
Bernanke, B., Gertler, M., 1987. Banking and Macroeconomic Equilibrium. In: Barnett, William, Singleton, Kenneth (Eds.), New
Approaches to Monetary Economics. Cambridge University Press, Cambridge.
Bernanke, B.S., Gertler, M., 1989. Agency Costs, Collateral and Business Fluctuations. American Economic Review 79 (1), 1431.
Bernanke, B.S., Lown, C.S., 1991. The credit crunch. Brooking Papers on Economic Activity 2, 205233.
Campbell, T.S., Kracaw, W.A., 1980. Information production, market signalling and the theory of nancial intermediation. Journal of
Finance 35 (4), 863882.
Cannata, F., Quagliariello, M., 2009. The role of Basel II in the subprime nancial crisis: guilty or not guilty? CAREFIN Research Paper.
3.
Chiuri, M.C., Ferri, G., Majnoni, G., 2002. The macroeconomic impact of bank capital requirements in emerging economies: past
evidence to assess the future. Journal of Banking and Finance. 26 (5), 881905.
Chong, B.S., Liu, M.-H., 2009. Islamic banking: interest-free or interest based? Pac. Basin Financ. J. 17, 125144.
Diamond, D.W., Dybvig, P., 1983. Bank runs, deposit insurance, & liquidity. The Journal of Political Economy. 91 (3), 401419.
Diamond, D.W., Rajan, R.G., 2000. A theory of bank capital. J. Financ. LV (No.6), 24312465.
Farook, S., Hassan, M.K., Clinch, G., 2012. Prot distribution management by Islamic banks: an empirical investigation. Quarterly
Review of Economics and Finance Volume 52, 333347.
Ferri, G., Kang, T.S., 1999. The credit channel at work: lessons from the nancial crisis in Korea. Economic Notes 28 (2), 195221.
Freixas, X., Rochet, J.-C., 1997. Microeconomic in Banking.Massachusetts Institute of Technology, Massachusetts: Cambridge.
Frost, S.M., 2004. The Bank Analyst's HandbookMoney, Risk and Conjuring Tricks. John Wiley & Sons, Ltd, Chichester, West Sussex.
Gambacorta, L., Mistrulli, P.E., 2004. Does bank capital affect lending behavior? J. Financ. Intermed. 13, 436457.
Gangopadhyay, S., Singh, G., 2000. Avoiding bank runs in transition economies: the role of risk neutral capital. Journal of Banking and
Finance. 24 (4), 625642.
Goodhart, C., Hofmann, B., Segoviano, M., 2004. Bank regulation and macroeconomic uctuations. Oxf. Rev. Econ. Policy 20 (4),
591615.
Grais, W., Kulathunga, A., 2007. Capital structure and risk in Islamic nancial services. In: Archer, S., Karim, R.A. (Eds.), Islamic
Finance: The Regulatory Challenge. S. Archer. & R.A. Karim. Oxford University Press, New York, pp. 68119.
Haselman, R., Wachtel, P., 2007. Risk taking by bank in transition countries. Comp. Econ. Stud. 49, 411429.
Hassan, M.K., Chowdhury, M.A.M., 2010. Basel II and Islamic banking regulation. American Journal of Islamic Social Sciences (AJISS)
27, 74101 (Winter).
Hassan, M.K., Dicle, M.F., 2005. Basel II and regulatory framework for Islamic banks. Journal of Islamic Economics, Banking and
Finance 1 (1).
Hassan, M.K., Dicle, M.F., 2007. Basel II and corporate governance of Islamic banks. In: Ali, Nazim (Ed.), Integrating Islamic Finance
into the Mainstream: Regulation, Standardization and Transparency. Islamic Finance Project, Islamic Legal Studies Program,
Harvard Law School.
Hassan, M.K., Hussain, M.E., Kayed, R.N., 2011. Impact of capital regulations on capital ratios and risk taking of Islamic banks. Review
of Islamic Economics 15 (No. 1), 521.
Kim, D.W., Lee, Y.S., Park, K.S., 2002. Credit crunch and shocks to rms: Korean experience under the Asian nancial crisis. Emerging
Market Review. 3, 195210.
Kunt, A.D, Huizinga, H., 2011. Do We Need Big Banks? Evidence on Performance, Strategy and Market Discipline. Policy Research
Working Paper (no 5576). The World Bank Development Research Group, Finance and Private Sector Development Team.
Leland, H.E., Pyle, D.H., 1977. Informational asymmetries, nancial structure and nancial intermediaries. J. Financ. 32 (2),
371387.
Mester, L.J., 1992. Traditional and nontraditional banking: an information-theoretic approach. Journal of banking and Finance. 16,
545566.
Muljawan, D., Dar, H.A., Hall, M.J., 2004. A capital adequacy framework for Islamic banks: the need to reconcile depositors' risk
aversion with managers risk taking. Appl. Financ. Econ. 14, 429441.
Peek, J., Rosengren, E., 1995. The capital crunch: neither a borrower nor a lender be. J. Money Credit Bank. 27 (no.3), 625638.
Pellegrina Dalla, L., 2007. Capital Adequacy Ratios, Efciency and Governance: A comparison between Islamic and Western Banks.
Working Paper Universit degli Studi di Milano- Bicocca, Dipartimento di Statistica.
Rashid, M., Abdeljawad, I., Nagalim, S.M., Hassan, M.K., 2012. Customer-centric corporate social responsibility: a framework for
Islamic banks on ethical efciency. Management Research Review 36 (Issue 4), 359378.
Revell, J., 1979. Ination and Financial Institutions. The Financial Times Ltd., London.
Santomero, A.M., 1984. Modeling the banking rm. J. Money Credit Bank. 16 (4), 576602.
Schmitz, B., 2007. The impact of Basel I capital regulation on bank deposits and loan: empirical evidence for Europe. Money Market
and Finance Research Group 39th Annual Conference. No 42 (http:www.essex.ac.uk/afm/mmf/index.html).

M. Abdul Karim et al. / Pacic-Basin Finance Journal 28 (2014) 5875

75

Sharpe, S.A., 1995. Bank capitalization, regulation, and the credit crunch: a critical review of the research ndings. No 95-20. Finance
and Economics Discussion Series.
Smolo, E., Hassan, M.K., 2010. Capital adequacy requirements for Islamic nancial institutions. Islamic Finance: Instruments and
Markets.Bloomsbury Press, London, UK.
Vandenbrink, D., 2005. Capital markets in Asia: changing roles for economic development. Institute of Southeast Asian Studies.
van den Heuvel, S., 2004. The bank capital channel of monetary policy. Working Paper.The Wharton School, University of Pennsylvania.
Yudistira, D., 2002. The impact of bank capital requirement in Indonesia. Economic Working Paper Archive at WUSTL.