14-Rajhi & Hassairi, 2013) - Islamic

___________________ Région et Développement n° 37-2013___________________
ISLAMIC BANKS AND FINANCIAL STABILITY:

A COMPARATIVE EMPIRICAL ANALYSIS BETWEEN
MENA AND SOUTHEAST ASIAN COUNTRIES
Wassim RAJHI*, Slim A. HASSAIRI**
Abstract - The aim of this paper is to investigate whether Islamic banks are
more stable than conventional banks. To measure the financial stability, we
compute the z-score for a sample of banks in 16 countries where Islamic and
conventional banks coexist over the period 2000-2008. We use a robust estima-
tion for analyzing data that are contaminated with outliers and leverage points
in the data. We use also a quantile estimation to allow us to address the question
whether the factors that cause high fragility are systematically different from the
factors that cause medium or low fragility. This empirical analysis explores
causes of insolvency risk between Middle East and North Africa and South East
Asian countries. Finally, by controlling for various factors and by favoring a
comparative analysis between the regions, this article is an extension of the
study begun by Čihák and Hesse (2010).
Key-words - ISLAMIC BANKING, FINANCIAL STABILITY, Z-SCORE,
MENA, SOUTHEAST ASIA.
JEL Classification - G21, G32, G33
We gratefully acknowledge valuable comments from Charles Lai-Tong, Catherine Kir-

by, Laurent Weill and the anonymous referee.
*
LEAD, Université de Toulon ; Email : [email protected].
**
ISCAE, Institut Supérieur de Comptabilité et d'Administration des Entreprises ;
Email : [email protected].
150 Wassim Rajhi, Slim A. Hassairi
1. INTRODUCTION
Emerging financial markets often include Islamic financial services. The
International Organization of Securities Commissions predicts that as much as
half of the savings of 1.2-1.6 billion Muslims will be in Islamic financial institu-
tions in 2015 (Imam and Kpodar, 2010). These services can be expected to en-
hance growth in these markets because of the positive relationship between
market liquidity and economic growth (Rajan and Zingales, 1998; Levine and
Zervos, 1998; Levine, 2005).1 Financial development, including emerging Is-
lamic financial services, needs to be orderly and will be adversely affected by
adverse macroeconomic or financial shocks. It is therefore interesting to analyse
how and why an Islamic banking system might be affected by and respond to
these shocks differently from conventional banking systems. This empirical
study identifies the key differences between Islamic and conventional banks.
The literature focusing on Islamic finance from the viewpoint of distress
and insolvency is not very abundant. A large body of descriptive literature dis-
cusses risks in Islamic banking (Sundararajan and Errico, 2002; Grais and Ku-
lathunga, 2006; Grais and Pellegrini, 2006) but does so in theoretical terms,
while empirical papers focus on Islamic and conventional banking efficiency
(Yudistira, 2004; Hamim, Naziruddin and Syed, 2006; Abdul-Majid, Saal and
Battisti, 2010). The role of Islamic banks in financial stability has been ana-
lyzed in a consistent, cross-country, empirical fashion in 16 countries between
1993 and 2004 (Čihák and Hesse, 2010).2 These authors tried to identify wheth-
er market share of Islamic banks has a significant impact on the financial
strength of conventional banks, arguing that Islamic financial institutions pose a
risk to the financial system that in many regards differs from those posed by the
conventional financial system. They used the asset structure (ratio of total loans
to total assets) as a proxy for the credit risk. We note the absence of a variable
such as a proxy for interest rates risks in their analysis. This article is an exten-
sion of the study begun by Čihák and Hesse (2010).
We study the factors determining failures for both conventional and Is-
lamic banks in MENA and in Southeast Asian countries. For each of the regions
in our sample, we estimate a separate robust and quantile regression for the
period 2000-2008. We show that despite risk management, credit risk measured
by the ratio of loan loss provisions to net interest income decreases the z-score
in small Islamic banks in MENA countries and large Islamic banks in Southeast
Asian countries. The suggested explanation is that Islamic banks in these coun-
tries tend to have problems monitoring credit risks. Our results demonstrate that
income diversification is a factor of insolvency for small and large Islamic
banks in Southeast Asian countries. We find that in Southeast Asian countries,
large Islamic and conventional banks cannot coexist in a competitive market
without creating crowding out effects. Using the z-score as a measure of bank-
1
It is therefore an opportunity to promote economic growth for Muslims worldwide.
2
Bahrain, Bangladesh, Brunei, Egypt, Gambia, Indonesia, Jordan, Kuwait, Malaysia, Mauritania,
Pakistan, Qatar, Saudi Arabia, Tunisia, Turkey, United Arab Emirates and Yemen.
Région et Développement 151
specific stability, we find that in general Islamic banks’ z-scores are on average
higher than those of conventional banks (indicating higher stability than con-
ventional banks). This is not true however for small Islamic banks.
The structure of this paper is as follows. Section 2 discusses our method-
ology, and introduces the variables and data used in the paper (characterized in
more detail in appendix 1). Section 3 presents the empirical results. Section 4
summarizes the conclusions, and suggests topics for further research.
2. METHODOLOGY AND REVIEW OF LITERATURE

2.1. Banking Stability Measure
We define a measure of the distance-to-default, the z-score, which
measures the market value of a bank’s assets in relation to the book value of its
liabilities. The z-score indicates the distance from insolvency combining ac-
counting measures of profitability, leverage and volatility.3 Our primary de-
pendent variable is the z-score as a measure of individual bank risk. The z-score
has become a popular measure of bank soundness (Beck, Demirgüç-Kunt and
Merrouche, 2010; Čihák and Hesse, 2010). The z-score is inversely related to
the probability of a bank’s insolvency, i.e., the probability that the value of its
assets will become lower than the value of the debt. A higher z-score corre-
sponds to a lower risk of insolvency.
Inspired by Čihák and Hesse (2010), we attempt to establish a relation be-
tween z-score and a number of relevant macro and bank-specific indicators. For
that purpose and in order to get more insight into insolvency risk developments,
we estimate a robust and quantile regression in MENA and Southeast Asian
countries.4 Because bank size is an important factor in some of the existing pa-
pers on bank soundness, we subdivide banks into large and small Islamic and
conventional banks. The z-score is calculated by using individual banks. De-
pendent variable, z-score of bank i at time t, is:
eqi , j , t
( E ( ROAA ) i , j , t  )
tai , j , t
Z i , j ,t  (1)
ROAAi , j , t
3
We use the z-score and not a form of probit/logit to see how Islamic banking failure differs from
conventional banking failure. It was not the focus of our study given that one typically has only a
small number of bank failures and especially for Islamic banks. The z-score measures are being
applied at aggregate (sectoral, regional or country) levels, rather than individual bank ones.
Strobel (2010) demonstrates that aggregate bank insolvency risk measures that are constructed
using the mean of individually calculated z-score measures are inherently biased. The other ad-
vantage of using the z-score as a measure of bank stability is that it is based on accounting rather
than on market data.
4
Islamic jurisprudence is based on different schools that may vary from country to country (the
Shiah branch and the Sunni branch, which in turn includes the Madhahib, Shafie, Hanafi, Hanbali
and Maliki traditions). A typical example is the controversy regarding the Bay al Dayn (debt
trading), which is rejected by some Middle East scholars on the grounds that it involves riba
(interest). Islam prohibits the trading of debts, unless it is at par value.
where E(ROAA)i,j,t stands for expected return on average assets, σ(ROAA)i,j,t

denotes standard deviation of return on assets; (ROAA) as a proxy for return
volatility; eqi,j,t is bank’s equity and reserves; tai,j,t is bank’s assets. The z-score
measures the number of standard deviations a return realization has to fall in
order to deplete equity, under the assumption of normality of banks’ returns.
2.2. Regression Analysis
Our approach is mainly to test whether Islamic banks are more or less
stable than conventional banks, using regressions of z-scores as a function of a
number of variables. We aim to define an insolvency model of the form:
( ) - - ∑ ∑ - ∑ - - (2)
Where the dependent variable is the z-score (Zi,j,t) for bank i in country j
at time t; BI,J,t-1 is a vector of bank-specific variables; IJ,t-1 contains time-varying
industry-specific variables TS and TSIJ,t-1 are the type of banks and the interaction
between the type and some industry-specific variables; MJ,t-1 denotes macroeco-
nomic variables. Finally, i,j,t is the residual.
2.3. Control Variables
The risk of insolvency may be due to factors either internal or external to

the bank, such as macro-economic factors, market structures, institutional fac-
tors such as regulations and governance indicators (Čihák and Hesse, 2007;
2010). The list of explanatory variables used in regressions incorporates a num-
ber of possible risks and bank characteristics discussed in the literature. These
are divided into three groups: (i) bank specific variables from bank balance
sheets and profit and loss accounts, (ii) national banking sector data and (iii)
macroeconomic variables. To distinguish the impact of bank type on the z-
score, we include a dummy variable that takes the value of 1 if the bank in ques-
tion is an Islamic bank and 0 otherwise (i.e., if it is a conventional bank).5 In
order to control for bank specific characteristics, the regression includes a num-
ber of other control variables to separate the impact on financial stability of the
Islamic nature of a bank from the impact of other bank-level characteristics,
from macroeconomic and other system-level influences.
We include the bank’s asset size logged, capital-asset-ratio loans over as-
sets, credit risk, cost-income ratio and liquidity risk. The logarithm of total as-
sets is used as a proxy for bank size, whereas share of loans to assets describes
bank asset structure.6 As a robustness check, we add the ratio of loan loss provi-
sions to net interest income (credit risk). We expected that a higher ratio would
5
If Islamic banks were relatively weaker than conventional banks, the dummy variable would
have a negative sign in the regression explaining z-score.
6
The asset structure can be interpreted as the profitability potential (as the return on loans is in
general higher than on other investments such as government bonds or deposits) as well as the
credit risk.
increase the bank insolvency risk. However, this sign might be ambiguous too,
as higher loan loss provisions could reflect banks’ precautionary reserve build-
ing as well as high non-performing loans. Both proxies represent the risk profile
of the bank. We interacted a credit risk variable with the Islamic bank dummy
because there are differences between Islamic and conventional banks.
For cost efficiency, the cost to income ratio has been used as an indicator
of bank cost efficiency. This is a simple measure indicating how well banks
manage their total costs (such as overhead expenses) relative to their income;
higher values indicate more inefficiency. The cost to income ratio is an efficien-
cy measure similar to operating margin. Unlike the operating margin, lower is
better. The cost income ratio is most commonly used in the financial sector.
Liquidity risk is measured as a ratio of liquid assets to deposits and short-term
funding.7 Lack of liquidity can lead to large losses in asset/liability portfolios
and it can generate financial distress and insolvency. In general, liquidity re-
serves promote financial soundness; on the other hand, excess liquidity under-
mines efficiency and profitability.8
Originally developed by Laeven and Levine (2007), income diversity

captures the degree to which banks diversify from traditional lending activities
(those generating net interest incomes) to other activities. In Islamic banks, the
net interest income is generally defined as the sum of the positive and negative
income flows associated with the Profit and Loss Sharing (PLS) arrangements.
Total operating income of a bank can be classified as interest income and non-
interest income (other operating income). Net interest income measures income
from traditional interest earning activities. Other operating income measures
non-interest earning activities and consists of commissions, fees, as well as
trading and asset management income. A higher value of variable income diver-
sity corresponds to a higher degree of diversification. When a bank’s income
diversity equals one, it fully diversifies its earnings between interest income and
non-interest income. When a bank’s income diversity equals zero, it is special-
ized and exclusively engages in either earning interest income or non-interest
income. Consequently, an insight into how specific bank characteristics such as
levels of revenue diversification interact with bank insolvency risk is of particu-
lar interest. We interacted the income diversity variable with the Islamic bank
dummy because there are differences between Islamic and conventional banks.
In addition to bank-by-bank data, we incorporate country-specific control

variables, using a number of variables that take the same value for all banks in a
given year and country. We follow Schaeck and Čihák (2010) by using the in-
flation rate, the official exchange rate and real gross domestic product growth,
since macroeconomic developments are likely to affect the quality of banks‘
assets, as well as the level of bank capitalization. First, we use real GDP growth
7
The choice of a liquidity indicator was constrained, as it was the only uniform indicator availa-
ble in the BankScope database.
8
Given their limited access to an interbank market or hedging instruments, Islamic banks are
more prone to liquidity risk than conventional banks.
rate where we expect higher growth to reflect better conditions for financial
stability. An important variable influencing z-score is the economic activity in
the country. Next, we use inflation and official exchange rates, assuming that
price and exchange rate stability contribute to the profitability and stability of
the banking sector.
We adjust for the impact of the macroeconomic cycle by including LI-

BOR.9 Islamic banks often benchmark the pricing of their instruments to LI-
BOR. The value of assets such as a deferred sale and lease transaction will vary
with the distance between the price at which they were issued and market
changes in the benchmark.10 LIBOR changes affect an Islamic bank income
statement in the same way they do with a conventional bank depending on the
share of the balance sheet linked to the benchmark (Chapra and Ahmed,
2002). Finally, interest rate risk measured by 6-month LIBOR was included in
the model to control effect on bank stability. We interacted this with the Islamic
bank dummy.
Čihák and Hesse (2010) find that the market share of Islamic banks does
not have a significant impact on the financial strength of commercial and Islam-
ic banks. Beck, Demirgüç-Kunt and Merrouche (2010) argue that the conven-
tional banks in countries with a higher market share of Islamic banks were less
stable. Market share measured as bank assets over total banking sector assets is
used as the proxy for market power (Berger, 1995). At the systemic level, we
examine the conventional banks’ impact on Islamic banks and the hypothesis
that the presence of conventional banks lowers systemic stability. For this rea-
son, we have calculated the market share of conventional banks for each year
and country and interacted it with the Islamic bank dummy. A negative sign for
the interaction of the conventional market share and the Islamic bank dummy
would indicate that a higher share of conventional banks reduces their sound-
ness (reduces their z-scores).
We try to capture the possible impact of banking sector concentration on

risk-taking behavior by including the Herfindahl-Hirschman index in the mod-
el.11 Regarding country-specific factors, we include the concentration level,
9
The level of domestic interest rates can also influence banks’ risk (Dell’ Ariccia and Marquez,
2006; Maddaloni and Peydró, 2011; Borio and Zhu, 2012).
10
In profit and loss sharing arrangements, the rate of return on financial assets is unknown before
to undertaking the transaction. In purchase-resale transactions, a mark-up is determined based on
a benchmark rate of return, typically a return determined in international markets such as LIBOR.
Islamic financial institutions’ balance sheets are exposed to variations of rates of return linked to
LIBOR. An increase in LIBOR leads to an increase in the mark-up charged on new transactions
and expected returns by investment account holders, compared to the earnings from long-term
investments.
11
The Herfindahl-Hirschman index is an economic concept widely applied in competition law,
antitrust and technology management. It is the sum of squared market shares (in terms of total
assets) of all banks in the country. The index can have values from 0 to 1, moving from a huge
number of banks to a system with only one bank. An increase in this index indicates a decrease in
competition and an increase of market power, whereas decreases indicate the opposite.
measured by this index for each country and year, to account for cross-country
variation in financial stability caused by differences in market concentration.
Included in the institutions are the process by which governments are se-
lected, monitored and replaced, the capacity of the government to effectively
formulate and implement sound policies and the respect of citizens and the state
for the institutions that govern economic and social interactions. We expect that
banks are more likely to grow in environments where the indicators of govern-
ance are built on a solid foundation and might have an effect on banking risk.
Furthermore, we controlled for the effect of the institutional environment by
using the governance indicators compiled by Kaufmann, Kraay and Mastruzzi
(2010).12 This indicator captures cross-country differences in institutional de-
velopments that might have an effect on banking risks.
All explanatory variables are lagged one year (-1) to capture the possible
past effects of these variables on the banks’ individual risk. To do that, we esti-
mate the same regressions for different bank sizes between MENA and South-
east Asian countries. We had no a priori assumptions on the sign of coefficients
on these variables as they can all affect bank solvency positively or negatively.
To avoid the impact of outliers, we start by a robust estimation for ana-
lyzing data that are contaminated with outliers and leverage points in the data.
The robust estimation technique assigns lower weights to observations with
large residuals, thereby making the estimation less sensitive to outliers, fol-
lowed by a quantile estimation to allow us to address the question whether the
factors that cause high fragility are systematically different from the factors that
cause medium or low fragility (Čihák and Hesse, 2007). Several interpretation
issues are discussed in detail.
2.4. The Data

Our balance sheets and profit and loss accounts data stem from Bureau
van Dijk’s BankScope database. The BankScope data cover 80-90 percent of
the banking systems in terms of total assets.13 Our sample covers 467 conven-
tional banks and 90 Islamic banks for the period 2000-2008 in the following 16
countries: 6 Southeast Asian countries and 10 MENA countries (alphabetically
ordered): Bahrain, Bangladesh, Brunei, Egypt, Indonesia, Jordan, Kuwait, Ma-
laysia, Pakistan, Qatar, Saudi Arabia, Singapore, Tunisia, Turkey, United Arab
12
The Worldwide Governance Indicators (WGI) project reports governance indicators for 213
economies over the period 1996-2009. Indicators of governance include voice and accountability,
political stability, government effectiveness, regulatory quality, the rule of law, and control of
corruption.
13
To decide whether a bank is commercial or Islamic, we have used the BankScope classification
as a starting point. BankScope defines Islamic banks as being members of the "International
Association of Islamic Banks" plus 20 non-member banks are considered to be "Islamic" by Fitch
Ratings. In the BankScope database, some Islamic banks are mistakenly categorized as conven-
tional banks and some as investment banks. The data set does not differentiate between conven-
tional banks with Islamic windows and other conventional banks.
Emirates and Yemen. We have estimated the regressions without Pakistan. It

had no significant impact on our main results. We exclude banking systems that
are entirely Islamic - Iran and Sudan - from the sample. We have included in
conventional banks savings banks and cooperative banks. Some conventional
banks have opened dedicated Islamic branches conducting business according
to Islamic banking principles. We therefore focus only on the comparison of
fully-fledged Islamic banks and conventional banks.
Table 1. Number of Observations by Specialisation and Sub-Samples
Specialisation Observations Percent

Full Sample
Conventional banks 2326 84.89
Islamic banks 414 15.11
Total 2740 100
Large Banks
Total 1464 100
Small Banks
Total 1276 100
Analysis of the sub-samples might show the possible impact of differ-

ences in risks of Islamic banks versus conventional banks. Since bank size is an
important factor in bank soundness, we subdivide banks into large and small
Islamic banks and large and small conventional banks (using total assets of US$
1 billion as the cut-off point between small and large banks). The threshold has
been used in previous research on small and large banks (Čihák and Hesse,
2007; 2010). In total, we have up to 2740 observations (414 Islamic banks and
2326 conventional banks). Our sample covers conventional banks (84.89 %)
and Islamic banks (15.11%) as shown in table 1. 14.21 % of the Islamic banks
and about 85.79 % conventional banks fall into the large bank category. 16.14
% of the Islamic banks and 83.86 % of the conventional banks fall into the
small bank category.
3. EMPIRICAL RESULTS
3.1. Descriptive Statistics and Pairwise Comparisons
As a preliminary step in the analysis, we present a comparison of z-scores
for various subgroups. Table 2 below illustrates that the treatment of outliers is
important.14 If the outliers were not included, the result for large conventional
14
Table 2 includes outliers.
banks would be reversed (because there are some large conventional banks with
extremely high z-scores). In the presence of outliers, ordinary least square esti-
mation can be biased. Because the least squares predictions are dragged towards
the outliers, and because the variance of the estimates is artificially inflated, the
result is that outliers can be masked. A robust regression is an important tool for
analyzing data that are contaminated with outliers and leverage points in the
data. A preliminary look at the z-scores suggests high variability in the sample,
with a z-score varying from - 12.47 to 1614.3. The average is 25.62 for conven-
tional banks with a maximum of 1614.3. The average for Islamic banks is of
33.61 with a maximum of 1002.7.15 Whatever the sample, Islamic banks’ z-
scores are on average higher than those of conventional banks (indicating higher
stability than conventional banks).
Table 2. Descriptive Statistics and T-test Value of Z-score by Specialisation
T-test T-test T-test
Sample Full Sample Large Banks Small Banks
Value Value Value
Commercial Islamic Commercial Islamic Commercial Islamic
Specialisation
Banks Banks Banks Banks Banks Banks
Observations 2287 402 1239 204 1048 198
Mean 25.627 33.612 24.418 31.586 27.058 35.699
Minimum -12.475 -7.691 -6.054 -1.750 -12.475 -7.691
Maximum 1614.3 1002.7 1107.3 208.6 1614.3 1002.7
Kurtosis 433.445 108.223 368.999 11.799 368.571 73.140
Skewness 18.813 9.1216 17.112 3.344 17.826 7.936
Standard Deviation 59.988 69.415 47.024 38.054 72.378 91.144
Methods of Equality of Means
Pooled (Equal Variance) (-2.40)** (2.07)** (-1.47)
Satterthwaite (Unequal Variance) (-2.17)** (-2.41)** (-1.26)
Folded F (Equality of Variances) [1.34]*** [1.53]*** [1.59]***
F Value []. T Value (). P values: * Significant at 10%; ** significant at 5%; *** significant at 1%.
The statistical tests in the table 2 indicate whether the mean for z-score is
the same for conventional and Islamic banks. The statistical test for the equality
of means is reported for both equal and unequal variances. Both tests indicate
evidence of a significant difference between z-scores, except for small Islamic
banks (for the pooled and for the Satterthwaite tests). These tests assume that
the observations are normally distributed. The equality of variances test does
indicate a significant difference in the variances. The results indicate that there
is a statistically significant difference between the mean z-score for convention-
al and Islamic banks. In other words, Islamic banks have a statistically signifi-
15
These comparisons of z-scores between banks are useful, but may overlook some additional
factors that explain bank-to-bank variation in z-scores. We will therefore examine this issue more
formally using robust and quantile regression analysis.
cant higher mean than conventional banks, except for small Islamic banks. Our
results are different from those found by Čihák and Hesse (2010).16
3.2. Regression Analysis
To separate the financial stability impact of the Islamic nature of a bank

from the impact of other bank-level characteristics, and from macroeconomic
and other system-level influences, we turn to regression analysis, following the
methodology described in section 1. We run several specifications between
MENA and Southeast Asian countries. The results for the robust estimation
technique, which assigns lower weights to observations with large residuals,
thereby making the estimation less sensitive to outliers, are shown in the tables
3 to 11, while table 12 shows results of quantile estimation technique.
The regressions do not confirm the result from the comparison of z-scores
in table 2. We find that large Islamic banks tend to be more stable than large
conventional banks, while small Islamic banks tend to be less stable than small
conventional banks. The sign of the Islamic dummy variable is predominantly
positive in Southeast Asian countries, significant in the regressions for large
banks at the 1 percent level in the specifications (1) (2) and (4). The regressions
show that in MENA countries small Islamic banks are less stable; the sign of
the Islamic dummy variable is negative at the 1 percent level in the specifica-
tions (3) and (4). Nevertheless, in Southeast Asian countries, the results for
small Islamic banks become less clear-cut, reflecting the opposite signs in the
specifications (1) and (2).
The bank’s asset size logged is on average positively related to bank sta-
bility at the 1 and 10 percent levels in the full sample and large banks in MENA
and Southeast Asian countries. The slope coefficient is predominantly positive
at the 1 and 5 percent levels; except for small banks in MENA and Southeast
Asian countries, where the results are negative at the 1 and 5 percent levels. The
bank’s asset size logged is on average negatively related to bank stability for
small banks.
Whatever the sample, our econometric results indicate that a higher cost-
to-income ratio has a consistently negative link to the z-scores in all specifica-
tions at the 10, 5 and 1 percent levels. Our results show that higher liquidity
increases bank stability; the results are significant at the 1, 5 and 10 percent
levels. The impact of the liquidity risk does explain much of the variation in z-
scores, so we can clearly drawn conclusions about the association between li-
quidity and stability.
16
Their results indicate that: (i) small Islamic banks tend to be financially stronger than
small conventional banks, (ii) large conventional banks tend to be financially stronger
than large Islamic banks and (iii) small Islamic banks tend to be financially stronger
than large Islamic banks.
Table 3. Robust Estimation (Full Sample - All Countries)

Least Trimmed Squares Parameter - Dependent Variable: Z-score
Region/Country All Countries

Estimate no. (1) (2) (3) (4)
1.6303 2.4049 17.1106 2.4980
Islamic Bank Dummy (0.3002) (0.1345) (0.0048)*** (0.7054)
0.9413 0.8909 0.2303 0.7884
Log Asset (-1) (<.0001)*** (<.0001)*** (0.2604) (<.0001)***
0.0543 0.0437 0.0317 0.0300
Loans/ Assets (-1) (0.0012)*** (0.0107)** (0.0817)* (0.0982)*
-0.0008 -0.0009 0.0009 -0.0015
Credit Risk (- 1) (0.3800) (0.3331) (0.3882) (0.1362)
Credit Risk * Islamic Bank Dummy -0.0001 0.0000 -0.0269 0.0010
(-1) (0.9703) (0.9793) (0.1037) (0.5937)
-0.1290 -0.1386 -0.0770 -0.8754
Income Diversity (-1) (0.5314) (0.4970) (0.7070) (0.0054)***
Income Diversity * Islamic Dummy 1.4184 1.0468 1.7899 2.2368
(-1) (0.5512) (0.6571) (0.4608) (0.4441)
-0.0917 -0.0920 -0.0948 -0.1101
Cost / Income Ratio (-1) (<.0001)*** (<.0001)*** (<.0001)*** (<.0001)***
Liquid Assets Customer / Short Term Fund 0.0653 0.0663 0.0552 0.0702
(-1) (<.0001)*** (<.0001)*** (<.0001)*** (<.0001)***
-8.4326 -18.3769 -10.0749
Herfindahl-Hirschman Index (-1) (0.0161)** (<.0001)*** (0.0051)***
0.0187 0.0819 0.0386
Market Share Conventional Banks (- 1) (0.5333) (0.0224)** (0.2758)
Market Share Conventional Banks*Islamic -0.1829 -0.0187
Bank Dummy (- 1) (0.0081)*** (0.7926)
5.1854
Governance (<.0001)***
0.1175
LIBOR (- 1) (0.4894)
LIBOR* Islamic Bank Dummy 0.1542
(-1) (0.7886)
0.1960
Real GDP Growth (-1) (0.0952)*
-0.1154
Inflation (-1) (0.0063)***
-0.1204
Official Exchange Rate (-1) (<.0001) ***
4.3001 4.8679 12.5860 6.5026
Constant (0.1372) (0.2279) (0.0092)*** (0.1527)
Observations 2059 2059 1856 1924
R-Squared 0.1456 0.1509 0.1652 0.1841
Robust P values in parentheses * Significant at 10%; ** significant at 5%; *** significant at 1%.
Estimates of regression coefficients are reported in the following model:
( ) ∑ ∑ ∑
This table reports the regression results for the robust estimation with sample splits between
MENA and Southeast Asia countries. The dependent variable is the z-score, a high index means a
low bank insolvency risk exposure; thus the relationship between independent variables and bank
insolvency risk exposure is reversed from the sign in this table. If Islamic banks were relatively
weaker than conventional banks, the dummy variable would have a negative sign in the regres-
sion explaining z-scores. A negative sign for the interaction of independent variables and the
Islamic bank dummy would indicate that a higher value reduces their soundness (reduces their z-
scores).
Table 4. Robust Estimation (Full Sample - Middle East and North Africa)
Region/Country Middle East and North Africa
Estimate no. (1) (2) (3) (4)
-0.2882 0.9799 -6.8376 -18.1131
Islamic Bank Dummy (0.8505) (0.5189) (0.2701) (0.0048)***
1.1800 1.2074 1.1314 1.0764
Log Asset (-1) (<.0001)*** (<.0001)*** (<.0001)*** (<.0001)***
0.0712 0.0606 0.0651 0.0509
Loans/ Assets (-1) (0.0004)*** (0.0036)*** (0.0054)*** (0.0235)**
-0.0068 -0.0068 -0.0058 -0.0102
Credit Risk (- 1) (0.0179)** (0.0183)** (0.0483)** (0.0029)***
Credit Risk * Islamic Dummy 0.0065 -0.0171 -0.0138 0.0006
(-1) (0.0379)** (0.2309) (0.3382) (0.9720)
-0.0442 -0.0482 -0.4109 -0.5798
Income Diversity (-1) (0.5079) (0.4685) (0.1517) (0.0385)**
Income Diversity * Islamic Bank Dummy 7.7296 6.2538 5.6403 12.3567
(-1) (0.0004)*** (0.0039)*** (0.0110)** (<.0001)***
-0.1141 -0.1104 -0.0945 -0.0750
(-1) (<.0001)*** (<.0001)*** (<.0001)*** (<.0001)***
-5.2253 -4.2052 -4.3887
Herfindahl-Hirschman Index (-1) (0.2504) (0.4007) (0.3583)
-0.0371 -0.0787 -0.0653
Market Share Conventional Banks (- 1) (0.2624) (0.0492)** (0.0980)*
Market Share Conventional Banks*Islamic 0.1210 0.1270
Bank Dummy (- 1) (0.0956)* (0.0665)*
1.7158
Governance (0.0733)*
-0.1379
LIBOR (- 1) (0.5240)
(-1) (0.0272)**
0.1170
Real GDP Growth (-1) (0.3008)
-0.0591
Inflation (-1) (0.1466)
-0.0469
Official Exchange Rate (-1) (0.1222)
-1.2870 2.8543 5.5769 8.4458
Constant (0.7373) (0.5123) (0.2775) (0.0762)*
Observations 1024 1024 921 891
R-Squared 0.2047 0.2330 0.2378 0.3161
Table 5. Robust Estimation (Full Sample - Southeast Asia and Pakistan)

Region/Country Southeast Asia and Pakistan
Estimate no. (1) (2) (3) (4)
27.0732 27.2524 64.4143 65.8398
Islamic Bank Dummy (<.0001)*** (<.0001)*** (<.0001)*** (0.0003)***
1.1848 1.0209 -0.9108 0.6976
Log Asset (-1) (<.0001)*** (0.0001)*** (0.0037)*** (0.0118)**
0.0085 0.0083 -0.0093 0.0066
Loans/ Assets (-1) (0.7455) (0.7561) (0.7247) (0.8025)
0.0001 0.0001 0.0001 0.0001
Credit Risk (- 1) (0.8852) (0.9211) (0.9544) (0.9384)
Credit Risk * Islamic Bank Dummy --0.2765 -0.2658 -0.3544 -0.1842
(-1) (0.0431)** (0.0502)* (0.0071)*** (0.1872)
-3.5375 -3.7024 -4.7155 -3.8694
Income Diversity (-1) (0.0003)*** (0.0001)*** (<.0001)*** (<.0001)***
Income Diversity * Islamic Bank Dummy -40.1238 -40.7372 -22.3115 -31.2994
(-1) (<.0001)*** (<.0001)*** (0.0021)*** (<.0001)***
-0.0754 -0.0691 -0.0850 -0.0621
(-1) (<.0001)*** (<.0001)*** (<.0001)*** (<.0001)***
-5.4796 -1.6679 -7.6589
-0.0036 0.4005 0.2414
Market Share Conventional Banks (- 1) (0.9624) (<.0001)*** (0.0106)**
Bank Dummy (- 1) (0.0035)*** (0.1020)
11.4280
0.3013
LIBOR (- 1) (0.2374)
LIBOR* Islamic Bank Dummy -3.4249
(-1) (0.0307)**
0.5922
Real GDP Growth (-1) (0.0482)**
-0.5984
Inflation (-1) (<.0001)***
-0.3368
Official Exchange Rate (-1) (<.0001)***
4.9788 8.3199 5.1899 -11.2119
Constant (0.2645) (0.3787) (0.6392) (0.2955)
Observations 1035 1035 935 1033
R-Squared 0.1979 0.2075 0.3218 0.2453
Table 6. Robust Estimation (Large Banks - All Countries)


Estimate no. (1) (2) (3) (4)
6.4977 7.2694 16.0866 24.1546
Islamic Bank Dummy (0.0009)*** (0.0005)*** (0.1596) (0.0373)**
3.3443 3.4005 2.5492 3.5719
Log Asset (-1) (<.0001)*** (<.0001)*** (<.0001)*** (<.0001)***
0.1110 0.1273 0.0454 0.1382
Loans/ Assets (-1) (<.0001)*** (<.0001)*** (0.1090) (<.0001)***
0.0009 0.0011 0.0013 0.0008
Credit Risk (- 1) (0.4812) (0.4023) (0.2770) (0.5656)
Credit Risk * Islamic Bank Dummy 0.0189 0.0050 0.0727 0.0370
(-1) (0.6597) (0.9092) (0.1075) (0.4496)
1.5793 1.4175 -0.0213 -0.8662
Income Diversity (-1) (0.1954) (0.2500) (0.7810) (0.3695)
Income Diversity * Islamic Dummy -4.8437 -4.8741 1.6256 -3.5978
(-1) (0.1069) (0.1096) (0.5567) (0.3105)
-0.1059 -0.1088 -0.0982 -0.1404
(-1) (<.0001)*** (<.0001)*** (<.0001)*** (0.0155)**
8.2635 -6.0067 9.0315
Herfindahl-Hirschman Index (-1) (0.0643)* (0.2247) (0.0504)*
0.1103 0.2686 0.2229
Market Share Conventional Banks (- 1) (0.0233)** (<.0001)*** (0.0002)***
Bank Dummy (- 1) (0.1882) (0.0304)**
8.0925
-0.1059
LIBOR (- 1) (0.6493)
(-1) (0.0748)*
0.4321
Real GDP Growth (-1) (0.0023)***
-0.1583
Inflation (-1) (0.0077)***
-0.0812
Official Exchange Rate (-1) (0.0427)**
-36.9669 -49.6152 -43.2630 -63.6639
Constant (<.0001)*** (<.0001)*** (<.0001)*** (0.0001)***
Observations 1186 1186 1092 1069
R-Squared 0.2796 0.2865 0.3622 0.3570
( ) ∑ ∑ ∑
scores).
Table 7. Robust Estimation (Large Banks - Middle East and North Africa)
Estimate no. (1) (2) (3) (4)
1.5897 0.0145 15.6499 -7.7177
Islamic Bank Dummy (0.4188) (0.9942) (0.2353) (0.5692)
1.6320 1.6370 2.3091 1.9113
Log Asset (-1) (<.0001)*** (<.0001)*** (<.0001)*** (<.0001)***
0.0232 0.0222 -0.0651 -0.0034
Loans/ Assets (-1) (0.4624) (0.4851) (0.0942)* (0.9334)
-0.0093 -0.0100 -0.0088 -0.0093
Credit Risk (- 1) (0.0586)* (0.0398)** (0.0858)* (0.0572)*
Credit Risk * Islamic Bank Dummy 0.0134 -0.0406 0.0898 0.0878
(-1) (0.7433) (0.3342) (0.0565)* (0.0586)*
-0.0977 -0.0098 -0.0087 -1.4669
(-1) (0.2748) (0.0711)* (0.1543) (0.0820)*
-0.1301 -0.1131 -0.1237 -0.1800
Liquid Assets Customer / Short Term Fund 0.0230 0.0328 -0.0336 0.0260
(-1) (0.0514)* (<.0001)*** (0.2494) (0.4500)
-4.2430 -7.2067 -1.6226
0.0461 0.3146 0.1316
Market Share Conventional Banks (- 1) (0.4782) (0.0003)*** (0.1193)
Bank Dummy (- 1) (0.2153) (0.9348)
8.1699
-0.2124
LIBOR (- 1) (0.4490)
LIBOR* Islamic Dummy 1.7556
(-1) (0.0190)**
0.4305
-0.0392
Inflation (-1) (0.5130)
-0.0317
-3.1913 -6.3434 -34.0406 -16.0096
Constant (0.6795) (0.5131) (0.0034)*** (0.1889)
R-Squared 0.2369 0.2629 0.2851 0.3484
Table 8. Robust Estimation (Large Banks - Southeast Asia and Pakistan)

Estimate no. (1) (2) (3) (4)
29.8441 30.1486 3.0197 110.7513
Islamic Bank Dummy (<.0001)*** (<.0001)*** (0.8933) (0.0003)***
5.9120 5.6824 2.7381 4.5289
Log Asset (-1) (0.0007)*** (<.0001)*** (<.0001)*** (<.0001)***
0.3316 0.3406 0.2426 0.2677
Loans/ Assets (-1) (<.0001)*** (<.0001)*** (<.0001)*** (<.0001)***
-0.0003 -0.0002 0.0004 -0.0003
Credit Risk (- 1) (0.8684) (0.9057) (0.7966) (0.8159)
Credit Risk * Islamic Bank Dummy -0.3969 -0.3874 -0.4224 -0.5910
(-1) (0.0523)* (0.0584)* (0.0265)** (0.0053)***
-0.4271 -0.5849 -2.1337 -1.1236
Income Diversity (-1) (0.7330) (0.6412) (0.1304)*** (0.3570)
Income Diversity * Islamic Bank Dummy -25.4954 -25.0508 -15.4124 -26.9590
(-1) (0.0083)*** (0.0117)** (0.1197) (0.0138)**
-0.0331 -0.0348 -0.0359 -0.0345
Cost / Income Ratio (-1) (0.0004)*** (0.0002)*** (<.0001)*** (0.0002)***
(-1) (<.0001)*** (<.0001)*** (0.0001)*** (<.0001)***
24.8823 16.9407 13.8123
Herfindahl-Hirschman Index (-1) (0.0022)*** (0.0964)* (0.0943)*
0.1630 -0.1666 0.2001
Market Share Conventional Banks (- 1) (0.0953)* (0.3724) (0.0906)*
Market Share Conventional Banks*Islamic 0.1926 -1.1498
Bank Dummy (- 1) (0.4273) (0.0135)**
8.3201
-0.1222
LIBOR (- 1) (0.7446)
(-1) (0.1654)
0.4950
-0.6775
Inflation (-1) (0.0002)***
-0.3033
Official Exchange Rate (-1) (0.0027)***
-95.3330 -111.162 -20.5422 -89.1692
Constant (<.0001)*** (<.0001)*** (0.3652) (<.0001)***
R-Squared 0.4792 0.4875 0.5415 0.5193
Table 9. Robust Estimation (Small Banks - All Countries)

Least Trimmed Squares Parameter - Dependent Variable : Z-score
Estimate no. (1) (2) (3) (4)
-8.4574 -5.8727 -1.8487 -0.4078
Islamic Bank Dummy (0.00941)*** (0.0669)* (0.7953) (0.9571)
-1.8820 -1.9998 -1.8038 -1.5661
Log Asset (-1) (0.0001)*** (<.0001)*** (0.0007)*** (0.0009)***
0.0037 -0.0445 -0.0423 -0.0490
Loans/ Assets (-1) (0.8573) (0.0342)** (0.0638)* (0.0228)**
0.0002 -0.0036 -0.0097 -0.0145
Credit Risk (- 1) (0.8786) (0.0076)*** (0.0015)*** (<.0001)***
Credit Risk * Islamic Bank Dummy -0.0290 -0.0214 0.0103 -0.0051
(-1) (0.0712)* (0.1636) (0.0019)*** (0.8017)
0.0172 0.0071 -0.0258 -0.0688
(-1) (0.0009)*** (0.0046)*** (0.0213)** (0.0091)***
-0.0960 -0.1238 -0.0885 -0.1063
(-1) (<.0001)*** (<.0001)*** (<.0001)*** (<.0001)***
-25.6624 -31.5252 -28.4439
Herfindahl-Hirschman Index (-1) (<.0001)*** (<.0001)*** (<.0001)***
0.0744 0.0959 0.0883
Market Share Conventional Banks (- 1) (0.0345)** (0.0268)** (0.0278)**
Bank Dummy (- 1) (0.4774) (0.8148)
0.6638
Governance (0.4742)
0.4390
LIBOR (- 1) (0.0454)**
(-1) (0.0718)*
-0.4716
Real GDP Growth (-1) (0.0202)**
-0.0043
Inflation (-1) (0.9375)
-0.1266
Official Exchange Rate (-1) (0.0019)***
43.0742 45.0180 41.8684 39.3616
Constant (<.0001)*** (<.0001)*** (<.0001)*** (<.0001)***
R-Squared 0.1659 0.2022 0.2192 0.2512
( ) ∑ ∑ ∑
scores).
Table 10. Robust Estimation (Small Banks - Middle East and North Africa)
Estimate no. (1) (2) (3) (4)
-3.3683 -3.1425 -22.2046 -24.3112
Islamic Bank Dummy (0.2347) (0.2850) (0.0014)*** (0.0004)***
-1.4981 -1.6265 -1.4968 -1.8234
Log Asset (-1) (0.0207)** (0.0219)** (0.0566)* (0.0099)***
0.0300 0.0232 0.0906 0.0376
Loans/ Assets (-1) (0.2854) (0.4624) (0.0062)*** (0.2121)
-0.0121 -0.0128 -0.0044 -0.0141
Credit Risk (- 1) (0.0046)*** (0.0032)*** (0.1819) (0.0014)***
Credit Risk * Islamic Bank Dummy -0.0330 -0.0328 -0.0238 -0.0213
(-1) (0.0184)** (0.0197)** (0.1016) (0.2183)
0.1755 -0.3329 -0.3230 -0.2060
(-1) (<.0001)*** (<.0001)*** (0.0860)* (0.0012)***
-0.1130 -0.1407 -0.1158 -0.0978
(-1) (0.6112) (0.6744) (<.0001)*** (<.0001)***
-5.4040 -0.4194 -2.6636
0.0224 -0.0521 -0.0354
Market Share Conventional Banks (- 1) (0.5825) (0.3239) (0.4310)
Market Share Conventional Banks*Islamic 0.3390 0.2560
Bank Dummy (- 1) (<.0001)*** (0.0009)***
-1.3755
Governance (0.3999)
0.1705
LIBOR (- 1) (0.5969)
(-1) (0.3186)
-0.7859
-0.0726
Inflation (-1) (0.1859)
-0.0731
36.7312 38.1473 35.4897 45.5562
Constant (<.0001)*** (<.0001)*** (0.0013)*** (<.0001)***
R-Squared 0.2391 0.2462 0.2622 0.3412
Table 11. Robust Estimation (Small Banks - Southeast Asia and Pakistan)
Estimate no. (1) (2) (3) (4)
-16.4357 40.1367 109.6306 441.5086
Islamic Bank Dummy (0.0979)* (<.0001)*** (0.7321) (0.3629)
-0.5663 -0.9993 -1.1402 -1.3024
Log Asset (-1) (0.3757) (0.1214) (0.1328) (0.0411)**
-0.0831 -0.1073 -0.1093 -0.1496
Loans/ Assets (-1) (0.0047)*** (0.0003)*** (0.0012)*** (<.0001)***
0.0017 0.0010 -0.0052 -0.0024
Credit Risk (- 1) (0.1153) (0.3796) (0.4206) (0.1236)
Credit Risk * Islamic Bank Dummy 0.6371 0.2059 0.2197 -0.0878
(-1) (0.2320) (0.6194) (0.6137) (0.8372)
-5.4887 -5.9936 -5.7791 -4.8260
Income Diversity (-1) (<.0001)*** (<.0001)*** (0.0004)*** (0.0005)***
Income Diversity * Islamic Bank Dummy 7.3488 -67.9937 -67.4792 -49.5890
(-1) (0.6300) (<.0001)*** (<.0001)*** (0.0001)***
-0.0960 -0.1020 -0.0716 -0.1529
(-1) (<.0001)*** (<.0001)*** (<.0001)*** (<.0001)***
-28.1680 -58.2572 -27.2961
Herfindahl-Hirschman Index (-1) (0.0003)*** (<.0001)*** (0.0010)***
-0.8020 0.1282 -0.8559
Market Share Conventional Banks (- 1) (<.0001)*** (0.4874) (<.0001)***
Bank Dummy (- 1) (0.8307) (0.4114)
4.4665
Governance (0.0017)***
0.2812
LIBOR (- 1) (0.3286)
(-1) (0.0400)**
0.1066
0.3909
Inflation (-1) (0.0185)**
-0.0842
37.3279 127.3751 45.1546 135.7748
Constant (<.0001)*** (<.0001)*** (0.0444)** (<.0001)***
R-Squared 0.2567 0.2897 0.3367 0.3446
___________________ Région et Développement n° 37-2013___________________
Table 12. Quantile Regression Results (Full sample)

Dependent Variable : Z-score Logged
Middle East and Southeast Asia and
North Africa Pakistan
Quantile 0.25 0.5 0.75 0.25 0.5 0.75
(1) (2) (3) (1) (2) (3)
-1.3286 -0.8672 -0.1778 2.0862 4.1768 1.5198
Islamic Bank Dummy (0.1069) (0.1063) (0.7243) (0.3865) (0.0619)* (0.5267)
Log Asset 0.0904 0.0201 0.0142 -0.0035 0.0059 0.0160
(-1) (0.0028)*** (0.3415) (0.5675) (0.9128) (0.7598) (0.4125)
Loans/ Assets 0.0010 0.0013 0.0008 -0.0030 0.0001 -0.0022
(-1) (0.7403) (0.5902) (0.7842) (0.2428) (0.9694) (0.4003)
Credit Risk -0.0022 -0.0018 -0.0006 -0.0015 0.0001 0.0000
(- 1) (0.1013) (0.0165)** (0.5183) (0.9617) (0.6824) (0.9625)
Credit Risk * Islamic Bank 0.0023 0.0018 0.0005 -0.0145 -0.0137 -0.0175
Dummy (-1) (0.1379) (0.0426)** (0.6250) (0.3264) (0.2811) (0.2072)
Income Diversity -0.0982 -0.0571 -0.0159 -0.0444 0.0219 -0.0114
(-1) (0.1816) (0.4167) (0.8705) (0.7974) (0.9016) (0.9434)
-0.0196 -2.1158
Income Diversity * Islamic 0.7549 0.3425 -2.5448 -1.4633
(0.9337) (0.0388)**
Bank Dummy (-1) (0.0373)** (0.1695) (0.0022)*** (0.0804)*
Cost / Income Ratio -0.0129 -0.0088 -0.0043 -0.0029 -0.0090 -0.0050
(-1) (0.0004)*** (<.0001)*** (0.0596)* (<.0001)*** (<.0001)*** (<.0001)***
Liquid Assets Customer/ 0.0030 0.0021 0.0016 0.0023 0.0043 0.0043
Short Term Fund (-1) (<.0001)*** (0.0599)* (0.3214) (0.6079) (0.0001)*** (0.0369)**
Herfindahl-Hirschman Index -1.1520 0.0154 -0.6799 0.0520 -0.1833 0.2984
(-1) (0.0694)* (0.9734) (0.1048) (0.9324) (0.7403) (0.5573)
Market Share Conventional -0.0125 -0.0031 0.0039 0.0122 0.0207 0.0045
Banks (- 1) (0.0102)** (0.3617) (0.2380) (0.1026) (0.0214)** (0.6575)
Market Share Conventional 0.0105 0.0087 0.0042 0.0055 -0.0267 0.0007

Banks*Islamic Dummy (- 1) (0.2224) (0.1067) (0.4172) (0.8169) (0.2652) (0.9800)
-0.0138 0.0169 0.0269 0.0396 0.0238 0.0193
LIBOR (-1) (0.5215) (0.3395) (0.3018) (0.1597) (0.2169) (0.2956)
LIBOR* Islamic Dummy 0.0661 0.0368 -0.0123 -0.2573 -0.1129 -0.0625
(-1) (0.1737) (0.4076) (0.7854) (0.2874) (0.3559) (0.7244)
-0.0001 0.0173 0.0157 0.0221 0.0565 0.0191
Real GDP Growth (-1) (0.9910) (0.0743)* (0.2014) (0.1365) (0.0815)* (0.3590)
-0.0089 -0.0081 -0.0066 -0.0163 -0.0318 -0.0346
Inflation (-1) (0.0633)* (0.0589)* (0.2519) (0.2125) (0.0033)*** (0.0005)***
0.0003 -0.0068 -0.0088 -0.0188 -0.0262 -0.0156
Official Exchange Rate (-1) (0.9439) (0.0386)** (0.0654)* (0.0030)*** (<.0001)*** (0.0010)***
2.8804 3.0214 2.8461 1.6314 0.8691 2.9497
Constant (<.0001)*** (<.0001)*** (<.0001)*** (0.0873)* (0.3747) (0.0091)***
Observations 879 879 879 1014 1014 1014
P values in parentheses * Significant at 10%; ** significant at 5%; *** significant at 1%.
Estimates of quantile regression coefficients are reported in the following model:
( ) ∑ ∑ ∑ .
This table reports the regression results for the quantile estimation with sample splits between
MENA and Southeast Asia countries. The dependent variable is the z-score logged, a high index
means a low bank insolvency risk exposure; thus the relationship between independent variables
and bank insolvency risk exposure is reversed from the sign in this table. If Islamic banks were
relatively weaker than conventional banks, the dummy variable would have a negative sign in the
regression explaining z-scores. A negative sign for the interaction of independent variables and
the Islamic bank dummy would indicate that a higher value reduces their soundness (reduces
their z-scores).
We find that a higher share of loans in the asset structure contributes to

increasing bank insolvency for large banks in MENA countries in specification
(3) at the 10 percent level. The slope coefficient is consistently negative for
small banks in Southeast Asian countries in all specifications and significant at
the 1 percent level. Nevertheless, a higher share of loans contributes to decreas-
ing bank insolvency for large banks in Southeast Asian countries in all specifi-
cation at the 1 percent level. The slope coefficient is consistently positive for
small banks in MENA countries in specification (3) at the 1 percent level.
Our assumption that credit risk contributes to the instability of the Islamic
banking sector is confirmed for large banks in Southeast Asian countries. The
results are negative in specifications (1), (2), (3) and (4) at the 1, 5 and 10 per-
cent levels. The results indicate that credit risk for small Islamic banks in
MENA countries is on average negatively related to bank stability in specifica-
tions (1) and (2) at the 5 percent level. Our results indicate that credit risk for
large Islamic banks in MENA countries is positively related to bank stability in
specifications (3) and (4) at the 10 percent level. Islamic banks have the greatest
exposure to credit risk. This risk would increase if lending were to expand rap-
idly before quality deteriorates. We find a negative relationship between z-score
seems to be inversely related to credit risk, probably due to the absence of en-
hanced screening and monitoring of borrowers. This might suggest that Islamic
banks in a given period reflect non-performing loans rather than good risk man-
agement practice.
Our results are consistent with Čihák and Hesse (2010). These authors
show that the smaller banks tend to focus on lower risk investments, whereas
more PLS financing is conducted by large Islamic banks. The suggested expla-
nation for these characteristics is that Islamic banks tend to have problems mon-
itoring credit risks when the business becomes bigger and therefore more com-
plex. The limits of this credit risk management at a larger scale mean problems
with moral hazard. Thus, potential distress continues to be a poor portfolio risk
management, despite the prudential regulation or a lack of attention to changing
external circumstances that can adversely affect the credit standing of a bank’s
counterparties. Ariss (2010) finds that between 2000 and 2006, Islamic banks
allocated a significantly greater share of their assets to financing compared to
conventional banks, implying a greater exposure to credit risk.17 In turn, Islamic
banks balance the higher portfolio risk with significantly lower financial risk
through higher capitalization levels.
Our econometric results show that the evidence regarding the impact of
revenue diversification on insolvency risk in each region and group is mixed.
There is a significant difference in terms of income diversity between Islamic
and conventional banks (large or small) in each region. With a strongly positive
coefficient, greater income diversity tends to increase z-scores in large and
17
The credit risk management in Islamic banks is more complicated not only due to the nature of
the contract and additional externalities. Islamic banks cannot charge a penalty due to default in
payment whereas conventional banks can charge penalties for overdue interest.
170 Wassim Rajhi
small Islamic banks in MENA countries at the 1, 5 and 10 percent levels, sug-
gesting that a move from lending-based operations to other sources of income
might improve stability in those banks. Greater income diversity tends to de-
crease z-scores for large and small Islamic banks in Southeast Asian countries;
the coefficient is strongly negative and significant in several specifications at
the 1 and 10 percent levels. A move from lending-based operations to other
sources of income might increase the insolvency risk in Islamic banks.
Our use of the GDP as variable is intended to capture the effect of macro-
economic conditions (business cycles) on z-score. As expected, GDP growth is
positive for the full sample in Southeast Asian countries and for large banks in
MENA countries; the results are significant at the 1 and 5 percent levels, sug-
gesting a business cyclical behavior of z-scores in these countries, except for
small banks in MENA countries, where the results are significant and negative
at the 1 percent level. Consumer price inflation leads to banking instability as
shown by its negative coefficient which is significant at the 1 percent level for
large banks in Southeast Asian countries. Nevertheless, the coefficient is posi-
tive and significant at the 1 percent level for small banks in Southeast Asian
countries. The official exchange rate plays a particularly important role in fi-
nancial instability as shown by its significant negative coefficient for large
banks in Southeast Asian countries at the 5 percent level, except in MENA
countries.
Our econometric results indicate that with a negative coefficient; an in-

crease in LIBOR leads to a decrease in z-scores in small Islamic banks in
Southeast Asian countries at the 5 percent level. In a dual system, Islamic
banks’ balance sheets in Southeast Asian countries are exposed to variations of
rates of return linked to LIBOR despite operating on interest-free principles. For
large Islamic banks in MENA countries, our result indicates that the interest rate
risk is significant at the 5 percent level with a positive coefficient. Consequent-
ly, an increase in LIBOR leads to an increase in z-scores in large Islamic banks
in MENA countries.
For large banks in MENA and Southeast Asian countries, a higher pres-
ence of conventional banks in a banking system has a positive impact on z-
scores at the 1 and 10 percent levels. Finally, conventional banks contribute to
the overall stability of a banking system. For small banks in Southeast Asian
countries, we can clearly observe the association between a presence of conven-
tional banks and instability. The coefficients are negative in the specifications
(2) and (4) at the 1 percent level. The interaction between the market share of
conventional banks and the Islamic banks dummy indicates that a higher pres-
ence of conventional banks does have a significant negative impact on z-scores
of large Islamic banks in Southeast Asian countries at the 5 percent level. This
result could be justified by the asynchronous reactions of conventional banks
which certainly affect the overall stability of a banking system by increasing
systemic risk. Our results indicate that in Southeast Asian countries, large Is-
lamic and conventional banks cannot coexist in a competitive market without
crowding out effects. Finally, we find that the presence of conventional banks
has a positive impact on the soundness of small Islamic banks in MENA coun-
tries, where the results are significant in the specifications (3) and (4) at the 1
percent level.
MENA countries recorded a substantial decrease of concentration in

banking sectors during the last decade.18 Looking at the movements in rolling
coefficients, we find that a higher concentration in the banking sector contribut-
ed to instability. For small banks, the impact of the Herfindahl-Hirschman index
is significantly negative in Southeast Asian countries at the 1 percent level.
However, as concentration later decreased, this relation finally became negative
in Southeast Asian countries, suggesting that a lower concentration with strong-
er competition in the banking sector contributed to lower insolvency risk. This
is in line with the literature on banking sector concentration and stability that
finds higher concentration to be associated with lower stability (Boyd, de Nico-
ló and Jalal, 2006; Schaeck, Čihák and Wolfe, 2009). For large banks in South-
east Asian countries, the findings are the opposite, where a higher level of con-
centration increases the z-score in all specifications at the 5 and 10 percent lev-
els. Southeast Asian countries recorded a substantial increase of banking sector
concentration during the last decade.19 Concentration in banking sectors is posi-
tively associated with stability in large banks. This result suggests that large
banks should increase their size through mergers and acquisitions in order to
achieve economies of scale and avoid failures. A higher concentration with
weaker competition contributed to lower insolvency risk. This is in line with
some other studies exploring the same relation in a much broader sample of
banks (Carletti and Hartmann, 2003; Allen and Gale, 2004).
The governance variable tends to be strongly positive in all regressions in

which it is entered, suggesting, as expected, that better governance is correlated
with higher z-scores.20 Our results are significant at the 1 percent level, except
for small banks in MENA countries.21 Thus, the quality of institutions as meas-
ured by the governance variable does not statistically influence the insolvency
for small banks in MENA countries. Unlike studies that have found that institu-
tions matter for financial risk (Čihák and Hesse, 2010), we find that insolvency
appears impervious to the quality of the institutional environment in MENA
countries. It may be that regardless of the institutional environment, the way
Islamic banks are permitted to behave is driven by relatively strict Sharia law,
18
Over the last two decades MENA countries, as well as much of the developing countries, have
experienced a wave of liberalization of the financial sector.
19
An increase in concentration can be related to the process of consolidation and restructuring.
Due to the severe impact of the 1998-2001 East Asian financial crisis, the Bank Negara Malaysia
forced mergers to consolidate the banking sector.
20
We did not think it was necessary to lag the governance variable. It did not have an impact on
the results as it does not change much year to year.
21
Here Sharia Supervisory Boards were extremely cautious in their advice, and sometimes even
obstructive. The quality of institutions does not matter, probably because the often higher hurdle
set by Sharia law trumps the quality of local institutions in most countries (Imam and Kpodar,
2010).
172 Wassim Rajhi
making the institutional environment less important than for banks in Southeast
Asian countries and large banks in MENA countries.
Finally, we introduce a methodological advance using quantile regression
to allow us to address the question of whether the factors that cause high fragili-
ty are systematically different from the factors that cause medium or low fragili-
ty.22 Quantile regression (table 12) can be considered superior to the previously
used estimation techniques since it provides more precise estimates of the im-
pact of the determinants of insolvency.
Given that our sample includes outliers, to control the skewness of the
distribution of the z-score, the model specification includes a logarithm of the z-
score which is normally distributed to smooth out higher values of the z-score.
The logarithm of the z-score is most closely related to the odds of insolvency
rather than the probability of insolvency itself (Strobel, 2011). Consequently,
we use the logarithm of z-score, although this does not help the quantile regres-
sion fit, it helps with statistical inference (Chen, 2005). The model setup is the
same as for the full sample with the same variables included in MENA and
Southeast Asian countries.
The model with the z-score logged shows that in Southeast Asian coun-
tries, the median Islamic banks have a higher z-score than the conventional
banks. The result is significant at the 10 % level. Greater income diversity tends
to increase z-scores on the weakest Islamic banks in MENA countries at the 5
% level. In Southeast Asian countries, greater income diversity tends to de-
crease z-scores on the weakest, median and strongest Islamic banks at the 1, 5
and 10 % levels. The coefficients of the credit risk and its interaction with the
Islamic bank dummy in MENA countries show that credit risk has a positive
effect on the median Islamic banks. The result is significant at the 5 % level.
4. CONCLUSION
These analyses are the first to compare empirically the risk of insolvency
of Islamic banks with conventional banks in MENA and Southeast Asian
countries. The relative financial strength of Islamic banks is assessed empiri-
cally based on evidence covering individual Islamic and conventional banks in
16 banking systems which have a substantial presence of Islamic banking. We
find that Islamic banks’ z-scores are on average higher than those of conven-
tional banks (indicating higher stability than conventional banks), except for
small Islamic banks. Our results are different from those found by Čihák and
Hesse (2010). These authors tried to see if market share of Islamic banks has a
significant impact on the financial strength of conventional banks, arguing that
22
The 50th percentile gives the median least square estimator which minimizes the median square
of residuals rather than the average. In the generalized quantile regression, we estimate an equa-
tion describing a quantile other than the median. Specifically, we estimate the first quartile (25th
percentile) as well as the 75th percentile.
Islamic financial institutions pose a risk to the financial system that in many
regards differs from those posed by the conventional financial system.
We know that Islamic banks may fail. The main contribution that we
hope our work makes is to promote overall a better understanding of various
biases that may affect theses failures. We observe that different causal links
analyzed give us new tools for the apprehension of risks in Islamic financial
institutions. Our findings demonstrate that credit risk and income diversity are
the most common cause of insolvency for Islamic banks. We have shown that
despite risk management, credit risk measured by the ratio of loan loss provi-
sions to net interest income decreases the z-score in small Islamic banks in
MENA countries and large Islamic banks in Southeast Asian countries. The
suggested explanation is that Islamic banks in these countries tend to have prob-
lems monitoring credit risks. Our findings demonstrate that income diversifica-
tion is a cause of insolvency for small and large Islamic banks in Southeast
Asian countries. A move from lending-based operations to other sources of
income might increase the insolvency risk in Islamic banks in Southeast Asian
countries while a move from lending-based operations to other sources of in-
come might improve stability for large and small Islamic banks in MENA coun-
tries. Our econometric results indicate that an increase in LIBOR leads to a de-
crease in z-scores in small Islamic banks in Southeast Asian countries. Islamic
banks in Southeast Asian countries are exposed to variations of rates of return
linked to LIBOR despite operating on interest-free principles. Our results indi-
cate that in Southeast Asian countries, large Islamic and conventional banks
cannot coexist in a competitive market without crowding out effects.
Islamic banking could take several measures to improve its stability and
compete with conventional counterparts. As is usually the case in social scienc-
es, the empirical tests conducted in our analysis have limitations. This study
could be completed by incorporating other variables such as stock indexes and
the age or experience level of banks, certainly more difficult to quantify, but
which must be taken into account. By focusing on the ownership structure (con-
trol and cash flow rights), we could examine whether this structure affects bank-
ing risks (Laeven and Levin, 2009; Sanya and Wolfe, 2010).
174 Wassim Rajhi
ANNEX 1.
Definition of Variables
Variable Description
Dependent Variable:
Insolvency Risk Exposure eqi , j , t
( E ( ROAA ) i , j , t  )
tai , j , t
Z i , j ,t 
ROAAi , j , t
where E(ROAA)i,j,t stands for expected return on average
assets, σ(ROAA)i,j,t denotes standard deviation of return
on assets (ROAA) as a proxy for return volatility eqi,j,t is
bank’s equity and reserves tai,j,t is bank’s assets.
Explanatory Variables :
Bank Specific Log Assets Logarithm of the total assets of a bank (In U.S. thousand
dollars).
Assets Structure Share of total credits in bank assets.
Cost to Income Ratio Ratio of cost to income.
Credit Risk Loan loss provisions to net interest income.
Credit risk*Islamic Bank Interaction of credit risk and Islamic bank dummy.
Dummy
Income Diversity 1 - │ (Net interest income - Other operating income) /
Total operating income│.
Income Diversity*Islamic Interaction of income diversity and Islamic bank dummy.
Bank Dummy
Liquidity Risk Liquid assets as percentage of customer and short term
funding.
Banking Sector Concentration Herfindhal-Hirschman Index.
Market Share of Conven- Market share of conventional banks in a country per year.
tional banks
Market Share of Conven- Interaction of market share of conventional banks and
tional Banks*Islamic Bank Islamic bank dummy.
Dummy
Macroeconomic GDP Growth Growth rate of nominal GDP, adjusted for inflation
Variables (Percent).
Inflation Year-on-year change of the CPI index (Percent).
Official Exchange Rate The official exchange rate refers to the exchange rate
determined by national authorities or the rate determined
in the legally sanctioned exchange market. It is calculat-
ed as an annual average based on monthly averages
(local currency units relative to the U.S. dollar), trans-
formed in percent.
London interbank offered 6-months LIBOR.
rate
London interbank offered Interaction of LIBOR and Islamic bank dummy.
rate*Islamic Bank dummy
Governance Governance Voice and accountability, political stability, government
effectiveness, regulatory quality, rule of law, control of
corruption (2000-2008).
Dummy Variable Islamic Bank Dummy Equals 1 for Islamic banks, 0 otherwise.
Sources: All microeconomic variables are taken from BankScope database. The Z-score, the Herfindhal-
H rschman ndex and he marke share of conven onal banks are he au hor’s calcula ons based on he
BankScope database. The macroeconomic variables are taken from the IFS (International Financial Statistics
published by the International Monetary Fund (IMF)). The LIBOR variable is taken from the Mortgage-X
L OR ndex. he dummy var able and he n erac ons are he au hor’s calcula ons. he governance varia-
ble is taken from Kaufmann, Kraay and Mastruzzi (2010). Data frequency: annual.
REFERENCES
Abdul-Majid, M., Saal, D. S. and Battisti, G., 2010, "Efficiency in Islamic and Con-
ventional Banking: An International Comparison", Journal of Productivity Analysis,
34, 25-43.
Allen, F. and Gale, D., 2004, "Competition and Financial Stability", Journal of Money,
Credit and Banking, 36(3), 453-480.
Ariss, R. T., 2010, "Competitive Conditions in Islamic and Conventional Banking: A
Global Perspective", Review of Financial Economics, 19(3), 101-108.
Bader, M. K. I., Mohamad, S., Ariff, M. and Hassan, T., 2008, "Cost, Revenue and
Profit Efficiency of Islamic versus Conventional Banks: International Evidence
Using Data Envelopment Analysis", Islamic Economic Studies, 15, 23-76.
Beck, T., Demirgüç-Kunt, A. and Merrouche, O., 2010, "Islamic vs. Conventional
Banking Business Model, Efficiency and Stability", Policy Research Working
Paper 5546, Finance and Private Sector Development Team, World Bank,
Washington, D.C., 44 p.
Berger, A., 1995, "The Relationship between Capital and Earnings in Banking",
Journal of Money, Credit and Banking, 27, 432-456.
Borio, C. and Zhu, H., 2012, "Capital Regulation, Risk-Taking and Monetary Policy: A
Missing Link in the Transmission Mechanism?", Journal of Financial Stability,
8(4), 236-251.
Boyd, J. H. and De Nicoló, G., 2005, "The Theory of Bank Risk Taking and
Competition Revisited", The Journal of Finance, 60(3), 1329-1343.
Boyd, J. H., De Nicoló, G. and Jalal, Abu M., 2006, "Bank Risk-Taking and
Competition Revisited: New Theory and New Evidence", IMF Working Paper
06/297, International Monetary Fund, Washington, D.C., 51 p.
Carletti, E. and Hartmann, P., 2002, "Competition and Stability: What's Special about
Banking?", European Central Bank Working Paper 146, 50 p.
Chapra, U. and Ahmed, H., 2002, "Corporate Governance in Islamic Financial
Institutions", Occasional Paper 6, Islamic Research and Training Institute, Islamic
Development Bank, Jeddah.
Chen, C., 2005, "An Introduction to Quantile Regression and the QUANTREG
Procedure", Statistics and Data Analysis SUGI 27, SAS Institute Inc.
Chen, C., 2007, "Robust Regression and Outlier Detection with the Robustreg
Procedure", Statistics and Data Analysis SUGI 30, SAS Institute Inc.
Chong, B. and Liu, M., 2009, "Islamic Banking: Interest-Free or Interest-Based?",
Pacific-Basin Finance Journal, 17(1), 125-144.
Čihák, M. and Hesse, H., 2007, "Cooperative Banks and Financial Stability", IMF
Working Paper 07/2, International Monetary Fund, Monetary and Capital Markets
Dept., Washington, D.C., 36 p.
Čihák, M. and Hesse, H., 2010, "Islamic Banks and Financial Stability : An Empirical
Analysis", Journal of Financial Services Research, 38(2-3), 19 p.
176 Wassim Rajhi
Dell’Ariccia, G. and Marquez, R., 2006, "Lending booms and lending standards", The
Journal of Finance, 61, 2511-2546.
Delis, M.D. and Kouretas, G.P., 2011, "Interest rates and bank risk-taking", Journal of
Banking and Finance, 35, 840-855.
Grais, W. and Kulathunga, A., 2006, "Capital Structure and Risk in Islamic Financial
Services", Islamic Finance: The Regulatory Challenge, John Wiley and Sons, Sin-
gapore, 418 p.
Grais, W. and Pellegrini, M., 2006, "Corporate Governance and Shariah Compliance in
Institutions Offering Islamic Financial Services", World Bank Policy Research
Working Paper, n° 4054.
Hasan, M. and Dridi, J., 2010, "The Effects of the Global Crisis on Islamic and
Conventional Banks: A Comparative Study", IMF Working Paper 10/201,
International Monetary Fund, Monetary and Financial Systems Dept., Washington,
D.C., 47 p.
Imam, P. and Kpodar, K., 2010, "Islamic Banking: How Has it Diffused?", IMF
Working Paper 10/195, International Monetary Fund, Washington, D. C., 30 p.
Kaufmann, D., Kraay, A. and Mastruzzi, M., 2010, "Governance Indicators for 2000-
2008, The Worldwide Governance Indicators (WGI) project", World Bank Policy
Research, Washington, D.C.
Koenker, R., 2005, Quantile Regression, Econometric Society Monograph,Cambridge
University Press, Cambridge, 349 p.
Laeven, L. and Levine, R., 2007, "Is There a Diversification Discount in Financial
Conglomerates", Journal of Financial Economics, 85(2), 331-367.
Laeven, L. and Levine, R., 2008, "Complex Ownership Structures and Corporate
Valuations", Review of Financial Studies, 21(2), 579-604.
Laeven, L. and Levine, R., 2009, "Bank governance, Regulation and Risk Taking",
Journal of Financial Economics, 93(2), 259-275.
Levine, R. and Zervos, S., 1998, "Stock Markets, Banks and Economic Growth",
American Economic Review, 88(3), 537-558.
Levine, R., 2005, "Finance and Growth: Theory and Evidence", Handbook of Econom-
ic Growth, Volume 1, Part A, 865-934.
Maddaloni, A. and Peydró, J., 2011, "Bank Risk-Taking, Securitization, Supervision,
and Low Interest Rates: Evidence from the Euro Area and U.S. Lending Standards",
Review of Financial Studies, 24, 2121-2165.
Hamim, S. A. M., Naziruddin, A. and Syed, M. Al-H., 2006, "Efficiency of Islamic
Banks in Malaysia: A Stochastic Frontier Approach", Journal of Economic
Cooperation among Islamic Countries, 27(2), 37-70.
Rajan, R. and Zingales, L., 1998, "Financial Dependence and Growth", American
Economic Review, 88(3), 559-586.
Sanya, S. and Wolfe, S., 2010, "Ownership Structure, Revenue Diversification and
Insolvency Risk in European Banks", Available at Social Science Research Net-
work (SSRN).
Schaeck, K. and Čihák, M., 2010, "Banking Competition and Capital Ratios",
European Financial Management, 18, 836-866.
Schaeck, K., Čihák, M. and Wolfe, S., 2009, "Are More Competitive Banking Systems
More Stable", Journal of Money, Credit and Banking, 41(4), 711-734.
Strobel, F., 2010, "Bank Insolvency Risk and Aggregate Z-score Measures: a Caveat",
Economics Bulletin, 30 (4), 2576-2578.
Strobel, F., 2011, "Bank Insolvency Risk, Z-Scores and an Improved Insolvency
Probability Bound", Available at Social Science Research Network (SSRN).
Sundarajan, V. and Errico, L., 2002, "Islamic Financial Institutions and Products in the
Global Financial System : Key Issues in Risk Management and Challenges Ahead",
IMF Working Paper 02/192, International Monetary Fund, Monetary and Exchange
Affairs Department, Washington, D.C., 27 p.
Weill, L., 2009, "Do Islamic Banks Have Greater Market Power?", Comparative
Economic Studies, 53 (2), 291-306.
Yudistira, D., 2004, "Efficiency in Islamic Banking: An Empirical Analysis of
Eighteen Banks", Islamic Economic Studies, 12(1), 1-19.
BANQUES ISLAMIQUES ET STABILITÉ FINANCIÈRE : UNE ANALYSE

COMPARÉE DES PAYS MENA ET DE L’ASIE DU SUD-EST
Résumé - L’ob ec f de ce ar cle es d’évaluer la stabilité financière des

banques islamiques et des banques conventionnelles. Pour mesurer la stabilité
financière, nous calculons sur la période 2000-2008 le z-score pour 16 pays où
les deux types de banques coexistent. A cette fin, nous utilisons un modèle d'es-
timation robuste et quantile afin de comparer les causes de l’ nsolvab l é des
banques islamiques et conventionnelles dans les pays du Moyen Orient et de
l’Afrique du Nord et celles du Sud-Est asiatique. Finalement, en utilisant diffé-
rents facteurs et une analyse comparative entre les régions, cet article tente
d’offrir une ex ens on à l’é ude de Č hák et Hesse (2010).
Mots-clés : BANQUE ISLAMIQUE, STABILITE FINANCIERE, Z-SCORE,

MENA, ASIE DU SUD-EST

14-Rajhi & Hassairi, 2013) - Islamic

Uploaded by

Copyright:

Available Formats

14-Rajhi & Hassairi, 2013) - Islamic

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

14-Rajhi & Hassairi, 2013) - Islamic

Uploaded by

Copyright:

Available Formats

___________________ Région et Développement n° 37-2013___________________

ISLAMIC BANKS AND FINANCIAL STABILITY:

Wassim RAJHI*, Slim A. HASSAIRI**

JEL Classification - G21, G32, G33

We gratefully acknowledge valuable comments from Charles Lai-Tong, Catherine Kir-

2. METHODOLOGY AND REVIEW OF LITERATURE

where E(ROAA)i,j,t stands for expected return on average assets, σ(ROAA)i,j,t

2.2. Regression Analysis

2.3. Control Variables

The risk of insolvency may be due to factors either internal or external to

Originally developed by Laeven and Levine (2007), income diversity

In addition to bank-by-bank data, we incorporate country-specific control

We adjust for the impact of the macroeconomic cycle by including LI-

We try to capture the possible impact of banking sector concentration on

2.4. The Data

Emirates and Yemen. We have estimated the regressions without Pakistan. It

Specialisation Observations Percent

Analysis of the sub-samples might show the possible impact of differ-

3.2. Regression Analysis

To separate the financial stability impact of the Islamic nature of a bank

Table 3. Robust Estimation (Full Sample - All Countries)

Region/Country All Countries

Table 5. Robust Estimation (Full Sample - Southeast Asia and Pakistan)

Table 6. Robust Estimation (Large Banks - All Countries)

Region/Country All Countries

Table 8. Robust Estimation (Large Banks - Southeast Asia and Pakistan)

Table 9. Robust Estimation (Small Banks - All Countries)

Table 12. Quantile Regression Results (Full sample)

Market Share Conventional 0.0105 0.0087 0.0042 0.0055 -0.0267 0.0007

We find that a higher share of loans in the asset structure contributes to

Our econometric results indicate that with a negative coefficient; an in-

MENA countries recorded a substantial decrease of concentration in

The governance variable tends to be strongly positive in all regressions in

BANQUES ISLAMIQUES ET STABILITÉ FINANCIÈRE : UNE ANALYSE

Résumé - L’ob ec f de ce ar cle es d’évaluer la stabilité financière des

Mots-clés : BANQUE ISLAMIQUE, STABILITE FINANCIERE, Z-SCORE,

You might also like

_ Région et Développement n° 37-2013_