Description
In light of the current debate on bank capital requirements, the purpose of this paper is to
investigate the relative impact of capitalization on risk-taking efficiency in Islamic and conventional
banks. The author tests whether changes occurring to the capital structure of such different types of
intermediaries unevenly affect their behaviour in terms of risk-taking efficiency.
Accounting Research Journal
Does capitalization enhance efficient risk undertaking?: A comparison between Islamic
and conventional banks
Lucia Dalla Pellegrina
Article information:
To cite this document:
Lucia Dalla Pellegrina, (2012),"Does capitalization enhance efficient risk undertaking?", Accounting
Research J ournal, Vol. 25 Iss 3 pp. 185 - 207
Permanent link to this document:
http://dx.doi.org/10.1108/10309611211290167
Downloaded on: 24 January 2016, At: 21:16 (PT)
References: this document contains references to 56 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 1362 times since 2012*
Users who downloaded this article also downloaded:
Ismail Cebeci, (2012),"Integrating the social maslaha into Islamic finance", Accounting Research J ournal,
Vol. 25 Iss 3 pp. 166-184 http://dx.doi.org/10.1108/10309611211290158
Rihab Grassa, (2012),"Islamic banks' income structure and risk: evidence from GCC countries", Accounting
Research J ournal, Vol. 25 Iss 3 pp. 227-241 http://dx.doi.org/10.1108/10309611211290185
Nafis Alam, (2013),"Impact of banking regulation on risk and efficiency in Islamic banking", J ournal of
Financial Reporting and Accounting, Vol. 11 Iss 1 pp. 29-50 http://dx.doi.org/10.1108/J FRA-03-2013-0010
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Does capitalization enhance
ef?cient risk undertaking?
A comparison between Islamic and
conventional banks
Lucia Dalla Pellegrina
Department of Economics, University of Milano-Bicocca, Milan, Italy and
Paolo Baf? Center, Bocconi University, Milan, Italy
Abstract
Purpose – In light of the current debate on bank capital requirements, the purpose of this paper is to
investigate the relative impact of capitalization on risk-taking ef?ciency in Islamic and conventional
banks. The author tests whether changes occurring to the capital structure of such different types of
intermediaries unevenly affect their behaviour in terms of risk-taking ef?ciency.
Design/methodology/approach – The paper conducts an empirical analysis using data for the
period 2001-2011 by means of both standard regression methods and stochastic cost frontier
techniques.
Findings – Results provide evidence that more capitalized Islamic banks are associated to less risky
positions in terms of their asset structure. In particular, the latter exhibit higher liquidity standards
and a lower incidence of non-performing loans compared to other banks. This has delayed positive
effects on pro?tability and no substantial impact on ef?ciency. On the other hand, highly capitalized
conventional banks tend to shift from more traditional lending activities to investment in other (pro?t
generating) assets. Such strategy increases pro?tability and ef?ciency, although raising impaired
loans.
Research limitations/implications – This study does not address potential endogeneity concerns
that might affect the variables at stake, hence mainly providing indications in terms of correlation
between phenomena rather than causality.
Practical implications – The analysis has important practical implications when considering
capital adequacy as a regulatory tool for managing the risk of Islamic banks’ activity, following
principles similar to those recommended by the Basel committee.
Originality/value – The original contribution of the paper to the literature consists of comparing the
effects of capitalization in different types of banks, and results can be usefully exploited by
policymakers wishing to tailor banking regulation on the speci?c model of banking they are entitled to
regulate.
Keywords Islam, Banks, Capitalization, Islamic banks, Capital structure, Capital adequacy,
Stochastic ef?ciency frontier
Paper type Research paper
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1030-9616.htm
The author thanks two anonymous referees, the Editor M. Kabir Hassan, Paolo Colla,
Eliana La Ferrara, David Llewellyn, Matteo Manera, Donato Masciandaro, Margherita Saraceno,
participants at the European Financial Management Association Conference on Risk
Management in Financial Institutions, April 23-25, 2009, Audencia School of Management,
Nantes, and participants at the Conference on Islamic Finance and the Financial Crisis,
November 18-19, 2009, FEEM, Milano, for useful comments. The author is also grateful to the
Monash University and the Paolo Baf? Center, Bocconi University. Fabio Dragoni has provided
excellent research assistance. The usual disclaimer applies.
Capitalization
185
Accounting Research Journal
Vol. 25 No. 3, 2012
pp. 185-207
qEmerald Group Publishing Limited
1030-9616
DOI 10.1108/10309611211290167
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
1. Introduction
As a response to the devastating consequences of the recent ?nancial crisis, considerable
effort has been devoted to set down remedies aimed at preventing possible reiteration.
Besides massive government intervention and bank bailouts, a deep revision of banks’
capital structure has been warmly recommended. Such a proposal, in particular,
represents a reaction to excessive leverage which led part of the ?nancial system to
collapse. In this vein, for instance, the current debate on the new Basel III principles
seems to go towards establishing more severe standards for bank capital requirements.
However, a key element in the discussion on capitalization regulatory constraints is
that they may limit banks’ pro?tability. Hence, the focus of the debate shifts (back) to
whether the capital structure affects bank performance and, related to the focus of this
paper, to what extent it differently affects intermediaries of various nature.
Unquestionably, ever since the pioneering work of Modigliani and Miller (1958), the
issue of whether solid capitalization has some in?uence on – not necessarily ?nancial –
?rms’ performance has generated one of most vivid debates in the theory of corporate
governance. Traditional theory on agency costs asserts that, under certain conditions,
a higher amount of equity relative to debt pushes managers towards more ef?cient
behaviour. The rationale for this outcome is explained by a reduction of con?icting
interests between managers, shareholders, and other investors ( Jensen and Meckling,
1976; Myers, 1977).
However, the traditional approach to the optimal capital structure of ?rms is not
univocally assessed. Jensen (1986) himself, for example, observes that less capitalized
institutions shouldbe more inclined to improve their performance because debt ?nancing
reduces hazardous behaviour of managers by lowering free cash ?ow. Moreover, Jensen
observes that managers’ moral hazard is likely to be particularly severe when their
rewards are dependent upon asset dimension – which is not infrequent in banking –
since ?rms’ activity could be expanded beyond its maximum level of ef?ciency.
As for the banking sector, the analysis of the effects of capital standards becomes
extremely important, especially because of their predictive role on default risk[1].
Empirical studies, however, provide ambiguous results in terms of the effects of
capitalization on the actual ef?cient risk-taking position of intermediaries and,
eventually, on their pro?tability (see for example Dahl and Shrieves, 1990; Editz et al.,
1998; Short, 1994, for a survey). When turning to the comparison between different
models of banking, any prediction can become even more fading.
In this paper we explore the role of capitalization on risk-taking behaviour of two
different types of banks: Islamic and conventional. Islamic banks are in fact
characterized by substantial differences compared to conventional ones, especially as
far as – most part of – deposit agreements are concerned. A key element for the
purposes of this paper is that, as opposite to conventional banks’ deposits, Islamic ones
are based on pro?t and loss sharing principles. Hence, changes in the capital structure
of Islamic banks may not end up stimulating effects which are identical to those
expected in the conventional system (i.e. lower risk exposure, although this needs to
be veri?ed in our context). Next, since a revision of the capital structure may induce
different investment strategies in the two types of banks, we also check whether and
how in either banking system these strategies re?ect on pro?tability and ef?ciency.
In order to investigate risk we mainly concentrate on banks’ balance sheet structure,
while considering the return on assets and risk-adjusted pro?t ef?ciency[2] to verify
ARJ
25,3
186
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
the effects on their performance. Estimates of pro?t ef?ciency, in particular, are
obtained through stochastic frontier techniques (Aigner et al., 1977; Battese and Coelli,
1995; Jondrow et al., 1982; Schmidt and Sickles, 1984).
We carry out the empirical analysis on commercial Islamic and conventional banks
in the period 2001-2011 in countries where both types of bank operate. In this way we
can assume that the main feature characterizing the differences in bank behaviour is
the type of bank, and not country-speci?c characteristics, like the economic, social, and
legal environment.
The paper innovates upon similar previous contributions to the literature by
providing a thorough evaluation of risk undertaking bank attitudes associated with
given capital standards in different models of banking, conditional on the institutional
set up. Contrary to previous studies on this topic we do not con?ne our attention on the
mere comparison of ef?ciency across different types of banks, but rather investigate
how different funding strategies (i.e. capital relative to other ?nancial resources) are
associated to different levels of risk undertaking, and how these elements combine in
either an ef?cient or an inef?cient way.
There are, in fact, several studies comparing ef?ciency scores computed either for
Islamic banks set up in different countries or for groups of Islamic and conventional
banks belonging to the same country (Al-Fayoumi and Alkour, 2008; Moktar et al.,
2006; Kader et al., 2007; Al-Jarrah and Molyneux, 2003; Hassan, 2005; Hasan, 2004;
Kamaruddin et al., 2008; Niazi et al., 2010; Qayyum, 2007; Su?an, 2007; Yudistira, 2004).
However, these works often disregard or put little attention to the contribution of
capital standards in explaining the behaviour of each individual model of banking,
especially with regard on bank ef?cient risk undertaking.
We ?nd that the strategies of Islamic banks in association with a higher degree of
capitalization are substantially different compared to those of conventional banks. Our
estimates suggest that, on the one hand, higher capitalized Islamic banks are actually
associated to less risky policies in terms of their assets structure. In particular, they
exhibit higher liquidity standards compared to other banks. We also ?nd that this
reduces non-performing loans, with overall weak delayed positive effects on
pro?tability and no substantial impact on ef?ciency.
On the other hand, more capitalized conventional banks tend to shift from more
traditional lending activities to investment in other assets, probably yielding higher
returns but also raising impaired loans. This is in line with Awdeh et al. (2011), Kahane
(1977), Kim and Santomero (1988), Kohen and Santomero (1980) and Stiglitz (2001),
among others. Overall, ef?ciency and pro?tability positively respond to such policy.
The implications of these results are important from the banking sector’s regulatory
perspective, since proposals welcoming the recapitalization of ?nancial institutions
often tend to disregard their nature and operational context. This analysis shows that
different capital standards are instead associated to different risk-taking strategies,
depending on the speci?c model of banking, suggesting somehow more tailored policy
guidelines (see, for example, Ainley et al., 2007; Hesse et al., 2008; Sundararajan and
Errico, 2002 for a discussion).
The paper is organized as follows: in the next section we discuss the main implications
of Islamic banking principles on bank capital structure and risk undertaking; in Section 3
we describe the dataset; and in Section 4 we illustrate the econometric techniques used for
the estimation. Results are presented in Section 5. Section 6 concludes.
Capitalization
187
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
2. Background discussion
Risk-sharing principles[3] – on which Musharakah and Musharakah credit agreements
are based – and the prohibition of investing in some given activities[4] – such
as gambling, pornography, gun trading, etc. – constitute the twin pillars of Islamic
?nance. Both these peculiarities could imply some different attitudes of Islamic banks in
terms of risk undertaking compared to conventional ?nancial intermediaries.
As far as risk-sharing principles are concerned, the Musharakah agreement is
invoked in the equity structure of banks and is similar to the modern concepts of
partnership and joint stock ownership. On the other hand, the bank manages deposits
to generate pro?ts subject to the rules of Mudarabah. In this perspective, the true
difference with respect to conventional banking consists in the fact that most of Islamic
banks’ depositors[5] are entitled to earn a share of pro?ts on a pre-speci?ed basis – and
also incur in losses – thus being rewarded with a variable return which depends on the
return on the pool of assets in which their funds are invested by the bank. However,
they are normally not allowed to participate to decisions concerning funds allocation.
In the reminder of this section we provide an overview of the main elements which
may affect the attitude towards risk in the two models of banking, with an eye on
speci?c corporate governance tools that different categories of stakeholders can exploit
in order to protect their own interests.
An important feature that may affect the way the capital structure in?uences
ef?cient risk undertaking is that Islamic banks’ ?nanciers are subject to a double
agency problem since, besides holding imperfect information on banking activities
(which, however, is also common to stockholders), they also suffer from imperfect
information on bank reported returns on investments. In either type of banks –
conventional and Islamic – such kind of complication normally pertains to
shareholders only, whereas in Islamic banks it also involves depositors.
It is worth noting, however, that depositors and stockholders are not to be
considered identical in any perspective. From our standpoint, the most relevant
features distinguishing the two are the rights protecting each category of investor and
their attitude towards risk.
First, as for rights protecting each category, only shareholders normally take part to
the decisions concerning Islamic bank management. In other words, they have both
voice and exit rights, while depositors are only endowed with the possibility to exit.
Second, as for their attitude towards risk, depositors participate to a lower share of
banks’ pro?ts and losses, thus normally bearing less risk compared to shareholders.
This, in some sense, reveals their higher risk aversion. These two features, if combined,
suggest that changing the relative share of equity to deposits may not be a neutral
policy from some perspectives because of stakeholder’s incomplete interest alignment.
Our concern is whether risk undertaking can be one of these perspectives.
Given the discussion above, one important element driving possible differences
between Islamic and conventional banks in terms of the speci?c effects of
capitalization on risk attitude is related to monitoring. In fact, since in Islamic banks
depositors share in the risk of bank choices, they could be pushed towards exerting
tighter oversight over bank management compared to what occurs in traditional banks
(C
?
iha´k and Hesse, 2010). This should in principle result in an overall intensi?ed
monitoring activity, also sustained by higher interest alignment between stockholders
and depositors ( Jensen and Meckling, 1976). This may eventually limit managers’
ARJ
25,3
188
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
risk-taking exposure. On the other hand, however, the intensity of shareholders’
oversight on managerial actions could be mitigated by the fact that the former have
more incentives to delegate it to other concerned agents (i.e. depositors).
Another element driving possible differences between Islamic and conventional
banks in terms of the speci?c effects of capitalization on risk attitude is the capability
of rewarding ?xed-return investors. In fact, following Jensen (1986), if the revision of
the capital structure involves a relative increase of debt over equity, then in a
conventional bank free cash ?ow must reduce, and risk with it. However, compared to
conventional banks, less capitalized Islamic banks should have less incentives to limit
risk undertaking by reducing free cash ?ow since they are allowed to shift part of the
risk on depositors (Hesse et al., 2008; Iqbal and Llewellyn, 2002; Sundararajan and
Errico, 2002). For this reason, in Islamic banks lower capitalization may not decrease
risk taking to the same extent as it does in conventional banks.
Overall, the arguments discussed above suggest that risk exposure in Islamic banks
may increase or decrease as a consequence of stakeholders’ weight in the pool of bank
?nanciers. In principle, the effects of a revision of the relative share of investor types on
risk-taking attitudes are not clear-cut when comparing Islamic banks to their
conventional counterparts.
Finally, with regard to the second pillar of Islamic ?nance (the prohibition of
investing in some given activities) predictions seem somehow less puzzling, especially
when considering the relationship between capitalization and bank risk-taking
positions, on the one hand, and pro?tability and ef?ciency, on the other hand.
In general, regardless of whether higher shares of capital increase or decrease the
level of risk, in Islamic banks the consequences on pro?tability may end up being
smoothed due to the fact that they are not allowed to exploit the full range of
risk-return possibilities available for conventional banks. In particular, if we admit –
as it has been largely recognized – that most forbidden activities also re?ect a higher
content of both risk and expected return, in Islamic banks the possibilities of pursuing
the highest risk-return strategies are limited compared to what their conventional
counterparts can achieve. Hence, a revision of capital standards may, under some
circumstances, correspond to a smoothed effect on pro?tability.
In the next sections we empirically investigate whether capitalization is effective in
affecting risk undertaking in either model of banking, and then whether this affects
pro?t ef?ciency to some extent.
3. Data
Our sample of conventional and Islamic banks[6] has been extracted from the database
Bankscope for the period 2001-2011[7]. We include all commercial banks established in
countries where both conventional and Islamic intermediaries operate[8]. In particular,
we do not consider investment banks. Furthermore, banks that failed to supply
information regarding the variables used in our empirical analysis have been excluded
fromthe sample. This process eliminates a considerable number of intermediaries which
were listed in the original database but did not make available any records apart from
their generalities[9]. Overall, the panel is composedby350 banks – 289 conventional and
61 Islamic – for a total of 2,764 observations.
The panel is unbalanced. This choice has been driven by considerations of different
nature. The ?rst concerns the importance of taking speci?c account of the contribution
Capitalization
189
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
of banks that, for very different reasons, either left business or were absorbed by other
ones during the period of analysis.
Second, the length of the time span, in addition to the dynamics that have been
characterizing the bank ownership structure during the last years would imply a
substantial reduction of the number of observations. In fact, due to mergers and
acquisitions, only a restricted set of intermediaries at the end of the time interval are
operating under their initial structure.
Third, it is worth observing that the number of Islamic banks has considerably
grown in recent years. Consequently, the content of the information concerning the
behaviour of the Islamic banking sector would be scarce in case we admitted only
companies that were already operating in 2001.
Aggregate descriptive statistics on the capital structure and other related measures
are reported in Tables I and II, for Islamic and conventional banks, respectively. Some
differences between the two types of intermediaries emerge from this preliminary
inspection.
On average, deposits are 70 and 57 per cent of total liabilities plus own capital
(i.e. equal to total assets), in the period 2001-2011 in conventional and Islamic banks,
respectively. Equity, on its turn, is 15 per cent of total assets in Islamic banks and
16.6 per cent in conventional banks. Residual liabilities are composed by bonds and
other sources of debt.
On the left side of their balance sheet Islamic and conventional banks report similar
shares of gross loans[10] over total assets (49 and 52 per cent, respectively). The share of
impaired loans to gross loans is slightly in favour of conventional banks (8 per cent
against 9 per cent of Islamic banks). Other earning activities also exhibit equal shares
Variable Label Obs Mean SD Min. Max.
Equity/total assets eq_ta 434 15.00 19.62 0 100
Liquid assets/total assets liqass_ta 434 25.22 18.98 0.02 97.70
Fixed-assets/total assets ?xass_ta 421 3.96 10.87 0 88.98
Gross loans/total assets gloans_ta 414 48.68 25.65 0 99.28
Other earning assets/total assets otherea_ta 433 35.61 24.45 0.01 98.90
Deposits/total assets dep_ta 364 56.84 28.84 0 100
Off-balance sheet assets/total assets offbal_ta 352 20.97 28.08 0 212.29
Liquid assets/short-term funding liq_shtmf 362 92.48 168.05 1.19 1,000
Loan loss provisions/gross loans llpr_gloans 333 3.24 14.60 281.64 100
Non-performing loans/gross loans imploans_gloans 228 8.76 15.41 0 83.20
Net income/total assets ROA 434 1.44 6.20 244.35 24.73
Total assets
a
ta 434 3.71 7.38 0.02 58.88
Index of ownership dispersion
c
bvd 434 5.93 3.67 1 11
Net interest revenues/income netirev_income 434 3.71 14.89 241.50 238.51
No. employees/total assets staffnr_ta 434 0.05 0.08 0.01 60
Staff expenditure/No. employees
b
staffexp 434 57.06 64.15 3.71 391.97
Interest rate
d
int 434 0.06 0.21 0 1
Deprecation rate
d
depr 434 0.14 0.24 0 1
Notes:
a
Billion US dollars;
b
thousand US dollars;
c
computed by Bankscope; the index ranges from 1 to
11; higher values indicate higher ownership dispersion; all variables are in percentage except
superscript “a”, “b”, “c”, “d”
Table I.
Descriptive statistics:
Islamic banks
ARJ
25,3
190
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
of total assets in both types of banks (36 per cent). The incidence of ?xed-assets in terms
of total assets in Islamic banks is instead notably higher (4 per cent against less than
2 per cent of conventional banks). The situation is reversed as far as liquidity is
concerned (25 per cent of total assets in Islamic banks against 29 per cent in conventional
banks). Residual assets are constituted by other non-earning activities, which, by
difference, result higher in Islamic banks. Finally, off-balance sheet items are higher in
conventional banks – 29 per cent of total assets – than in Islamic ones (21 per cent)[11].
Another accounting ratio which may provide an idea of a bank’s risk-taking
position is the share of loan loss provisions to gross loans (1 per cent in conventional
banks against 3 per cent in their Islamic counterparts). Finally, liquidity represents
92 per cent of short-term funding in Islamic banks, whereas conventional banks seem
exhibiting a higher degree of risk in these terms, since the ratio is 52 per cent.
Overall, at the aggregate level, conventional and Islamic banks do not show
substantial differences in terms of their balance sheet structure and attitude towards
the risk. On the one hand, Islamic banks retain more ?xed-assets than conventional
intermediaries, thus apparently showing a higher degree of risk. On the other hand,
however, they retain more liquidity in relation to the amount of short-term deposits,
which in some sense counterbalances the two facts above.
Turning to the variables that are used in the empirical analysis to explain risk
attitude measures, we ?rst observe a substantial difference in the dimension of the two
types of intermediaries. Islamic banks record on average 3.71 billion of total assets,
compared to a larger dimension of conventional banks (7.41 billion on average).
Another important difference emerge in terms staff endowment, since the ratio of
employees over total assets is in favour of conventional banks (0.11 per cent against
Variable Label Obs Mean SD Min. Max.
Equity/total assets eq_ta 2,330 16.59 15.66 0 100
Liquid assets/total assets liqass_ta 2,330 29.10 17.97 0.07 98.88
Fixed-assets/total assets ?xass_ta 2,328 1.69 1.95 0 39.35
Gross loans/total assets gloans_ta 2,325 52.23 18.65 0.78 100
Other earning assets/total assets otherea_ta 2,330 36.44 17.84 0.62 100
Deposits/total assets dep_ta 2,311 69.63 16.26 2.91 100
Off-balance sheet assets/total assets offbal_ta 2,085 28.93 28.54 0 314.26
Liquid assets/short-term funding liq_shtmf 2,309 51.75 78.74 0.19 1,000
Loan loss provisions/gross loans llpr_gloans 2,254 1.14 5.02 2100 100
Non-performing loans/gross loans imploans_gloans 1,984 8.41 10.64 0 100
Net income/total assets roa 2,330 1.25 3.01 2100 29.71
Total assets
a
ta 2,330 7.41 14.17 0.02 136.38
Index of ownership dispersion
c
bvd 2,330 4.76 3.54 1 11
Net interest revenues/income netirev_income 2,330 4.91 86.06 2552.03 3,997.09
No. employees/total assets
d
staffnr_ta 2,330 0.11 0.13 0.01 1.29
Staff expenditure/No. employees
b
staffexp 2,330 30.02 65.40 0.43 1,979.25
Interest rate
d
int 2,330 0.07 0.07 0 1
Deprecation rate
d
depr 2,330 0.08 0.14 0 1
Notes:
a
Billion US dollars;
b
thousand US dollars;
c
computed by Bankscope; the index ranges from
1 to 11; higher values indicate higher ownership dispersion; all variables are in percentage except
superscript “a”, “b”, “c”, “d”
Table II.
Descriptive statistics:
conventional banks
Capitalization
191
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
0.05 per cent of Islamic banks), although this ratio does not account for staff members’
quality and skills. Expenditure per employee, which may under some conditions re?ect
the latter aspect, is instead in favour of Islamic banks (57,000 US$ on a yearly basis,
against 30,000 US$ of conventional banks).
Variables referring to other input costs different from personnel expenses are also
summarized in Tables I and II. In particular, ?xed-assets depreciation (computed as
non-interest expenses over ?xed-assets) is 8 per cent in conventional banks, whereas it
is 14 per cent in Islamic banks. Interest rate[12] on deposits is, respectively, 6 and
7 per cent in Islamic and conventional banks.
Finally, in order to verify whether risk may re?ect on bank pro?tability, we account
for the return on assets (ROA) and ef?ciency scores computed trough pro?t ef?ciency
frontier techniques. As for the ROA, descriptive statistics exhibit records of 1.44 and
1.25 in Islamic and conventional banks, respectively.
In the next section we describe the estimated equations and the empirical techniques
used to compute ef?ciency scores.
4. Empirical analysis
4.1 Analysis of risk undertaking positions and measures of pro?tability
We estimate the following regression for each type of bank:
y
it
¼ a
0
þa
1
eq_ta
it
þa
2
X
it
þm
i
þm
t
þ1
it
ð1Þ
where i identi?es the bank, and t refers to the year. y
it
is either one measure of risk
undertaking discussed in the previous section (share of liquidity, loans, other earning
assets, off-balance sheet items, deposits to total assets, loan loss provisions and
non-performing loans over gross loans, and liquidity over short-term funding) or a
measure of pro?tability (the return on assets and its one-year lag).
In particular, a
1
is the estimated parameter of interest which refers to the
effect of the capital ratio (eq_ta) on risk and pro?tability of bank i at time t. X
it
is a vector of covariates that consist in bank size (total assets), staff endowment
relative to the dimension of the bank, a measure of the incidence of traditional lending
activity on the overall activities of the bank (measured by interest revenues over
normalized total income), and an ownership dispersion index computed by
Bankscope. We also include in X
it
the unitary costs of inputs (employees’ annual
salary, ?xed-asset depreciation rate, and interest rates paid on deposits) in order to
control for the convenience of undertaking certain investments or speci?c ?nancing
policies.
In addition, a
0
is a constant term; m
i
are bank ?xed effects[13], while m
t
are year
dummies used to capture the economic cycle and other time features. We omit the
dummy relative to 2005 in order to avoid collinearity[14]. 1
it
is an idiosyncratic error,
such that Eð1
it
jeq_ta
it
; X
it
; m
i
; m
t
Þ ¼ 0. Robust standard errors are used.
Equation (1) is estimated for both the sample of conventional as well as for Islamic
banks through random-effects general least squares, after testing that the error term is
not correlated with the regressors[15].
4.2 Pro?t ef?ciency
Pro?t (in)ef?ciency[16] is interpreted as the diversion between the income achievable
by a ?rm given the faced unitary cost of inputs and its actual observed income.
ARJ
25,3
192
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
It can be measured in several ways. Some diffused techniques, such as data
envelopment models and the free disposable hull analysis, are non-parametric.
Non-parametric methods, however, embody a strong assumption, consisting of the fact
that all deviations from the maximum income are due to inef?ciency. For this reason,
several studies in the banking literature (Maudos et al., 2002; Berger and Mester, 1997;
Berger and Humphrey, 1997; Berger and Bonaccorsi di Patti, 2006) have been conducted
using a stochastic frontier analysis (SFA), which allows for the presence of both a
one-sided inef?ciency component and a random error term. We follow this tradition
estimating the (indirect) relationship betweenbankcapital structure andbankinef?ciency
with SFA.
The most complicated feature to be managed when using stochastic frontier
techniques is to separate the actual inef?ciency component from other purely random
factors affecting bank pro?tability. The availability of panel data, in our case, allows
avoiding any strong hypotheses on the distribution of the inef?ciency component and
its independence from other factors determining bank behaviour, which is instead the
case of cross-section analyses[17].
We assume that the deterministic component of the pro?t ef?ciency frontier for
bank i in year t, depends upon input prices and output quantities, and has a translog
form[18]:
lnðP
ijt
þuÞ ¼ b
0
þ
m
X
a
m
ln y
mijt
þ
n
X
b
n
ln w
nijt
þ
1
2
m
X
h
X
a
mh
ln y
mijt
ln y
hijt
þ
1
2
n
X
k
X
b
nk
ln w
nijt
ln w
kijt
þ
n
X
m
X
g
nm
ln w
nijt
ln y
mijt
þm
i
þ u
ijt
þ v
ijt
ð2Þ
where P
ijt
is each bank’s observed pro?t represented by earnings on variable outputs
minus variable costs expended to produce these outputs, u is a scale-constant added to
every bank’s pro?t so that the natural log is taken of a positive number, w
n
are unitary
input prices, y
m
are output quantities, and m
i
are ?rm-speci?c effects made available by
the panel-structure of the dataset.
As for the inef?ciency component, u
ijt
is a one-sided pro?t inef?ciency term typical
of each bank i in country j at time t, while v
ijt
is a two-sided normal random noise term
with zero mean and a standard deviation s
v
. Following Battese and Coelli (1995)[19] we
assume that the u
ijt
are time-variant, constructed as the interaction between bank
inef?ciency and time.
Furthermore, w
ijt
is made of three input costs (included in X in equation (1)): the
unitary cost of capital to the bank (interest paid on deposits), the unitary cost of
?xed-assets (depreciation) and the unitary cost of labour (average annual salary).
Following the outstanding literature on bank ef?ciency measurement (Berger and
Humphrey, 1997) we include three outputs in vector y
ijt
: net loans, other earning assets,
and off-balance sheet transactions. Non-performing loans are subtracted from loans, in
such a way that a risk-adjusted frontier is obtained from the data.
After imposing the usual linear restrictions characterizing the translog pro?t
function[20], we performed our estimates following the one-step maximum likelihood
introduced by Battese and Coelli (1995)[21].
Capitalization
193
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Such technique also provides an equation in which ef?ciency terms are speci?ed as
a function of a set of explanatory variables. The estimated equation is the following:
j
ijt
¼a
0
þa
1
eq_ta
ijt
*isl
i
þa
2
eq_ta
ijt
*conv
i
þa
3
isl
i
þa
3
X
ijt
þm
i
þm
j
þm
t
þ1
ijt
ð3Þ
where j
ijt
are the ef?ciency components correspondent to the estimated u
ijt
, while m
i
, m
j
and m
t
are, respectively, bank, country, and time speci?c effects. a
0
is a constant term
and 1
ijt
is a random error.
Ratios between equity and total assets are separately accounted for Islamic and
conventional banks as the interaction between the equity-to-assets variable and
individual dummy variables for each type of banking (eq_ta
ijt*
isl
i
and eq_ta
ijt*
conv
i
). As
standard practice when interacting terms, one of the dummies (in this case isl
i
) is also
accounted for as a separate regressor. This is particularly important in order to
distinguish the contribution to ef?ciency provided by an increase of capital ratios in each
individual model of bankingfromother elements of inef?ciencywhichare typical of either
Islamic or conventional intermediaries but are not speci?cally related to capitalization.
Finally, X
ij
is a vector of control variables (logarithm of total assets, number of
employees over total assets, net interest revenues over normalized total income, and the
Bankscope ownership dispersion index) equal to the one included in equation (1), with
the exception of input costs.
5. Results
5.1 Regression output
The output from the estimation of equation (1) is reported in Table III for Islamic banks
and in Table IV for conventional banks. In each speci?cation we consider the same set
of controls while changing the dependent variables.
Results show that differently capitalized Islamic banks adopt different strategies
compared to conventional banks. More in detail, as reported in Table III, a higher capital
ratio corresponds to lower lending activity in Islamic banks (the parameter associated to
eq_ta in the gross loans-to-total assets is negative and strongly signi?cant). This
translates into holding more liquid assets than less capitalized Islamic intermediaries
(the parameter associated to eq_ta in the liquidity-to-total assets equation shows
opposite sign and almost the same magnitude as the one associated to loans).
It is then not surprising that the shift from loans to liquid assets that is observed in
Islamic banks also raises liquidity over short-term funding. Most important,
non-performing loans appear reduced and this is likely to slightly increase returns
afterwards, since the parameter associated to the one-period lagged return on assets
turns positive and, although weakly, signi?cant. In a nutshell, more capitalized Islamic
banks seem to engage in safer policies compared to less capitalized in terms of their
assets structure and this increases pro?tability with some delay.
As reported in Table IV, more capitalized conventional banks also refrain from
engaging in more traditional lending activities (as it occurs in Islamic banks, the
parameter of eq_ta in the regression of gross loans-to-total assets is also signi?cant
although considerably lower than in the case of Islamic banks) at the advantage of
investment in other pro?t-generating activities (the parameter associated to eq_ta turns
positive and signi?cant in the third column of Table IV). Disinvesting in traditional
lending also leads conventional banks to increase liquidity, although to lesser extent
than in their Islamic counterparts (the parameter associated to eq_ta turns signi?cant in
ARJ
25,3
194
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
l
i
q
a
s
s
_
t
a
g
l
o
a
n
s
_
t
a
o
t
h
e
r
e
a
_
t
a
o
f
f
b
a
l
_
t
a
d
e
p
_
t
a
l
i
q
_
s
h
t
m
f
l
l
p
r
_
g
l
o
a
n
s
i
m
p
l
o
a
n
s
_
g
l
o
a
n
s
a
r
o
a
r
o
a
_
1
e
q
_
t
a
0
.
2
3
3
5
9
*
*
*
2
0
.
2
7
1
9
7
*
*
*
0
.
0
4
9
7
4
2
0
.
0
1
1
6
6
2
0
.
1
4
6
4
1
2
.
6
4
2
4
8
*
*
2
0
.
0
3
7
2
3
2
0
.
0
9
7
0
3
*
*
0
.
0
1
8
7
2
0
.
0
4
7
5
1
*
(
0
.
0
7
1
8
0
)
(
0
.
0
7
8
8
3
)
(
0
.
0
8
9
8
8
)
(
0
.
0
3
6
6
7
)
(
0
.
1
0
9
3
1
)
(
1
.
0
9
8
4
6
)
(
0
.
0
3
9
1
5
)
(
0
.
0
4
4
1
9
)
(
0
.
0
2
5
3
0
)
(
0
.
0
2
8
4
6
)
t
a
0
.
1
3
4
4
3
0
.
4
1
9
8
2
*
2
0
.
3
3
5
0
3
0
.
5
2
8
0
9
*
0
.
0
4
6
2
2
2
1
.
4
9
6
2
0
2
0
.
1
5
6
2
7
0
.
0
8
3
9
2
0
.
0
5
2
4
7
*
2
0
.
0
4
5
5
6
(
0
.
1
2
0
3
6
)
(
0
.
2
1
4
5
5
)
(
0
.
2
3
5
7
9
)
(
0
.
2
7
1
3
0
)
(
0
.
2
2
4
2
6
)
(
1
.
1
6
9
6
1
)
(
0
.
1
2
0
8
9
)
(
0
.
1
0
0
2
6
)
(
0
.
0
3
0
8
3
)
(
0
.
0
7
3
9
2
)
b
v
d
0
.
1
9
3
0
1
2
2
.
4
3
6
6
6
*
*
*
0
.
8
1
2
2
2
0
.
5
8
0
4
3
2
0
.
9
0
3
4
4
1
.
3
5
4
9
1
0
.
5
3
8
9
5
2
0
.
7
0
2
7
2
*
0
.
1
6
4
0
6
*
*
0
.
3
7
5
4
4
*
*
(
0
.
6
3
3
2
5
)
(
0
.
8
3
3
4
6
)
(
0
.
7
2
3
9
2
)
(
1
.
5
2
3
6
2
)
(
0
.
9
4
7
6
0
)
(
3
.
0
5
5
8
7
)
(
0
.
4
7
1
0
5
)
(
0
.
4
0
1
6
1
)
(
0
.
0
8
0
1
3
)
(
0
.
1
5
4
6
3
)
n
e
t
i
r
e
v
_
i
n
c
o
m
e
2
0
.
0
0
3
1
4
0
.
0
5
6
2
7
2
0
.
0
1
7
8
0
2
0
.
1
5
4
4
1
*
*
*
0
.
0
3
0
2
6
*
*
0
.
5
0
5
2
0
2
0
.
0
4
2
8
1
2
0
.
0
2
3
3
7
0
.
0
1
0
0
8
0
.
0
3
2
6
6
(
0
.
0
2
9
3
5
)
(
0
.
0
4
0
9
0
)
(
0
.
0
5
4
1
7
)
(
0
.
0
3
0
2
1
)
(
0
.
0
1
5
0
8
)
(
0
.
6
8
5
4
1
)
(
0
.
0
2
9
2
1
)
(
0
.
0
2
4
7
5
)
(
0
.
0
1
1
0
6
)
(
0
.
0
3
6
4
9
)
s
t
a
f
f
n
r
_
t
a
1
,
4
0
9
.
7
6
2
2
,
0
3
6
.
0
0
*
2
2
,
7
5
3
.
6
1
*
*
6
8
1
.
0
1
5
1
,
5
2
7
.
7
7
1
1
,
2
1
1
.
8
1
1
,
2
3
5
.
8
5
5
8
5
8
,
9
1
2
.
0
7
*
*
*
2
8
7
5
.
2
0
2
*
2
4
,
4
6
2
.
0
6
*
*
*
(
1
,
7
1
2
.
4
5
3
4
0
)
(
1
,
1
4
2
.
8
5
6
5
4
)
(
1
,
4
0
1
.
2
6
2
8
5
)
(
2
,
6
1
3
.
9
8
0
8
2
)
(
1
,
7
5
8
.
5
7
7
8
4
)
(
1
3
,
7
0
0
.
1
9
4
1
5
)
(
1
,
2
7
5
.
2
4
7
4
3
)
(
3
,
2
7
9
.
0
8
6
9
3
)
(
4
6
7
.
2
8
5
4
6
)
(
8
1
7
.
4
8
4
6
7
)
s
t
a
f
f
e
x
p
2
0
.
0
3
3
4
7
2
0
.
0
4
2
9
3
0
.
0
7
9
7
3
*
*
2
0
.
0
8
2
8
2
*
2
0
.
0
6
5
2
9
0
.
3
7
8
2
3
0
.
0
1
4
7
9
0
.
0
1
0
2
6
2
0
.
0
0
2
6
6
2
0
.
0
0
5
2
7
(
0
.
0
2
2
0
8
)
(
0
.
0
3
5
8
7
)
(
0
.
0
3
8
1
6
)
(
0
.
0
4
2
9
1
)
(
0
.
0
7
2
7
2
)
(
0
.
3
8
6
2
1
)
(
0
.
0
1
8
7
8
)
(
0
.
0
1
7
6
8
)
(
0
.
0
0
3
2
0
)
(
0
.
0
0
9
1
8
)
d
e
p
r
2
1
8
.
5
8
2
1
1
.
8
6
1
.
6
8
9
2
1
.
8
5
8
1
0
.
9
8
*
9
0
.
5
3
3
.
7
3
2
8
.
9
4
3
2
3
.
8
5
2
2
2
.
8
2
1
(
1
7
.
7
6
)
(
1
2
.
4
2
)
(
1
8
.
4
5
)
(
3
.
2
4
8
)
(
5
.
8
6
4
)
(
7
7
.
7
2
)
(
2
.
8
0
5
)
(
9
.
8
9
3
)
(
2
.
9
8
6
)
(
4
.
7
9
3
)
i
n
t
2
5
.
7
0
2
2
1
5
.
8
0
1
3
.
8
0
2
2
.
9
0
2
2
4
5
.
9
3
*
*
*
2
4
6
.
8
*
*
*
1
.
3
5
7
9
.
5
3
6
.
2
3
4
*
*
*
4
.
7
6
0
*
(
5
.
0
0
3
)
(
1
4
.
7
0
)
(
1
2
.
7
9
1
)
(
7
.
8
2
5
)
(
1
6
.
8
3
)
(
1
1
5
.
9
)
(
4
.
5
6
1
)
(
7
.
0
9
6
)
(
2
.
7
9
2
)
(
2
.
9
8
1
)
2
0
0
1
2
4
.
6
5
0
8
1
0
.
7
9
7
7
5
3
.
2
3
1
3
1
1
.
3
1
4
6
7
2
1
.
5
0
6
8
4
1
9
.
5
3
6
0
2
8
.
9
2
2
4
1
2
2
.
2
4
0
9
5
*
(
3
.
8
6
3
7
1
)
(
5
.
2
6
7
9
5
)
(
6
.
4
3
8
4
2
)
(
3
.
7
5
5
9
7
)
(
4
.
6
0
6
4
7
)
(
4
4
.
5
5
9
9
0
)
(
9
.
1
3
6
9
9
)
(
1
.
2
9
6
3
6
)
2
0
0
2
2
2
.
5
7
3
7
5
2
.
5
1
4
1
5
4
.
4
8
0
0
2
5
.
7
1
5
5
7
2
3
.
4
5
8
3
8
2
5
.
9
9
0
3
9
2
0
.
3
3
3
7
7
2
2
.
2
9
6
1
1
*
*
2
3
.
4
7
6
0
5
(
3
.
5
5
7
3
0
)
(
3
.
7
4
4
4
7
)
(
4
.
9
7
8
1
9
)
(
5
.
7
3
5
2
8
)
(
6
.
5
6
0
4
2
)
(
5
1
.
2
5
5
7
9
)
(
1
.
3
2
7
1
7
)
(
1
.
1
3
1
5
3
)
(
3
.
0
5
1
8
5
)
2
0
0
3
2
2
.
9
9
4
0
5
4
.
3
2
4
1
6
0
.
5
7
4
2
7
7
.
3
2
3
8
3
2
.
0
4
9
6
6
7
.
5
8
1
8
3
2
0
.
5
5
4
6
0
2
2
.
3
9
3
2
7
2
2
.
1
5
7
6
7
*
*
*
2
3
.
8
9
4
3
2
(
2
.
4
9
0
8
6
)
(
2
.
8
8
5
6
7
)
(
4
.
4
2
7
9
7
)
(
7
.
9
2
2
5
2
)
(
5
.
4
7
6
4
3
)
(
3
4
.
2
0
8
8
7
)
(
1
.
9
1
7
0
4
)
(
2
.
3
8
5
3
1
)
(
0
.
7
3
0
4
5
)
(
2
.
9
4
1
4
0
)
2
0
0
4
2
4
.
2
0
5
7
1
*
*
2
.
5
2
3
0
7
2
0
.
7
7
9
6
9
2
.
9
1
3
4
3
3
.
3
3
1
7
6
3
9
.
2
9
5
5
8
0
.
5
7
2
7
5
2
1
.
7
2
5
9
3
2
1
.
7
1
8
0
7
*
*
2
3
.
2
6
4
2
5
(
1
.
9
4
6
8
0
)
(
2
.
3
8
0
5
9
)
(
3
.
3
7
2
6
4
)
(
2
.
2
1
6
8
7
)
(
3
.
4
5
7
9
2
)
(
4
4
.
5
2
6
2
9
)
(
1
.
8
4
2
8
2
)
(
1
.
8
8
8
3
5
)
(
0
.
7
1
5
1
7
)
(
2
.
7
8
2
0
9
)
(
c
o
n
t
i
n
u
e
d
)
Table III.
Effect of capitalization on
risk and pro?tability:
Islamic banks
Capitalization
195
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
l
i
q
a
s
s
_
t
a
g
l
o
a
n
s
_
t
a
o
t
h
e
r
e
a
_
t
a
o
f
f
b
a
l
_
t
a
d
e
p
_
t
a
l
i
q
_
s
h
t
m
f
l
l
p
r
_
g
l
o
a
n
s
i
m
p
l
o
a
n
s
_
g
l
o
a
n
s
a
r
o
a
r
o
a
_
1
2
0
0
6
0
.
8
0
3
5
9
2
3
.
6
2
2
3
1
6
.
2
8
4
8
1
*
*
1
.
4
9
2
1
7
2
5
.
6
0
6
7
8
2
0
.
6
7
8
7
2
1
.
9
7
5
3
7
2
0
.
2
0
4
7
2
2
2
.
2
1
2
7
5
*
*
2
5
.
3
7
2
9
6
*
*
(
2
.
3
1
3
7
4
)
(
2
.
7
8
3
3
7
)
(
3
.
1
0
7
8
3
)
(
2
.
3
0
2
3
2
)
(
3
.
5
0
4
9
7
)
(
3
3
.
9
1
1
3
6
)
(
2
.
4
0
1
3
1
)
(
1
.
2
6
7
2
5
)
(
0
.
9
5
1
5
6
)
(
2
.
4
5
2
9
7
)
2
0
0
7
2
4
.
4
4
7
7
9
*
2
1
.
4
8
2
9
4
1
.
7
2
5
5
8
2
.
7
5
0
6
5
2
0
.
9
8
0
1
2
2
2
5
.
6
4
4
5
0
0
.
3
6
3
9
8
2
0
.
6
5
9
8
1
2
1
.
0
5
3
8
0
2
0
.
8
8
1
4
6
(
2
.
3
4
5
8
6
)
(
2
.
8
6
2
8
5
)
(
2
.
9
5
9
7
8
)
(
3
.
3
1
9
4
4
)
(
3
.
3
7
7
4
1
)
(
1
9
.
3
1
1
2
5
)
(
0
.
8
1
5
5
7
)
(
1
.
1
1
7
9
7
)
(
0
.
7
1
0
7
4
)
(
2
.
4
0
2
2
8
)
2
0
0
8
2
6
.
6
5
9
3
8
*
*
0
.
2
2
5
5
6
2
0
.
7
9
9
0
2
2
.
1
3
0
4
0
2
1
.
7
3
9
0
0
2
3
1
.
9
0
9
7
1
*
3
.
3
8
6
6
6
1
.
3
5
6
7
8
2
2
.
6
3
4
5
0
*
*
2
3
.
2
7
7
8
9
(
2
.
7
2
1
8
1
)
(
3
.
2
3
9
5
6
)
(
3
.
1
1
0
8
1
)
(
4
.
7
6
9
2
4
)
(
3
.
3
7
8
0
3
)
(
1
7
.
9
0
5
3
9
)
(
2
.
2
9
4
2
7
)
(
3
.
0
1
9
1
2
)
(
1
.
1
9
7
9
3
)
(
2
.
3
6
5
7
9
)
2
0
0
9
2
6
.
9
6
0
3
7
*
*
2
1
.
0
8
6
3
1
0
.
0
4
0
7
4
2
0
.
8
4
1
5
0
2
1
.
0
4
5
3
2
2
3
.
0
2
8
7
9
6
.
5
8
3
6
9
*
*
2
0
.
4
9
3
5
2
2
5
.
2
9
1
8
2
*
*
*
2
6
.
8
7
7
8
8
*
*
*
(
2
.
9
0
9
9
4
)
(
3
.
6
8
5
9
9
)
(
3
.
3
7
9
9
1
)
(
4
.
7
0
5
2
8
)
(
3
.
5
9
6
8
2
)
(
2
8
.
5
8
5
0
3
)
(
3
.
3
0
8
4
1
)
(
3
.
4
9
0
2
8
)
(
1
.
7
8
1
7
6
)
(
2
.
3
4
3
7
1
)
2
0
1
0
2
8
.
9
4
8
1
2
*
*
*
2
2
.
8
8
9
3
2
1
.
0
7
3
5
5
2
0
.
7
4
9
5
6
2
0
.
0
7
0
6
8
2
1
3
.
5
4
7
0
5
5
.
4
2
7
4
4
*
*
0
.
7
9
3
7
2
2
6
.
5
9
1
3
9
*
*
*
2
7
.
8
9
3
1
3
*
*
*
(
3
.
2
3
0
0
2
)
(
4
.
1
5
7
8
8
)
(
3
.
8
4
3
4
7
)
(
5
.
3
6
8
0
5
)
(
3
.
4
8
5
0
6
)
(
2
7
.
7
4
4
4
0
)
(
2
.
6
9
0
1
9
)
(
3
.
5
4
2
6
4
)
(
1
.
9
4
0
5
7
)
(
2
.
3
6
3
4
8
)
2
0
1
1
2
9
.
9
1
4
6
0
*
*
*
2
4
.
7
7
1
9
6
1
.
8
1
6
9
5
2
8
.
9
5
7
2
2
*
1
.
8
1
0
5
0
2
2
3
.
8
0
4
0
0
0
.
2
3
7
9
8
0
.
4
5
0
6
3
2
4
.
3
3
1
5
3
*
*
*
2
6
.
2
6
6
4
4
*
*
(
3
.
5
0
6
0
4
)
(
4
.
0
9
8
9
1
)
(
4
.
2
9
7
3
0
)
(
5
.
2
4
1
4
1
)
(
3
.
4
5
4
7
8
)
(
2
3
.
5
1
7
4
8
)
(
2
.
7
1
8
1
4
)
(
3
.
0
1
2
1
3
)
(
1
.
2
9
4
6
5
)
(
2
.
4
8
5
2
2
)
C
o
n
s
t
a
n
t
2
6
.
9
7
2
3
9
*
*
*
6
9
.
7
0
5
0
6
*
*
*
2
6
.
3
3
4
2
7
*
*
*
2
2
.
6
6
6
0
4
*
*
*
6
4
.
9
8
6
0
6
*
*
*
2
7
.
4
7
0
2
5
2
2
.
4
3
3
7
2
5
.
8
9
1
2
6
3
.
6
8
1
2
4
*
*
*
6
.
8
3
1
5
7
*
*
*
(
4
.
4
1
2
6
7
)
(
6
.
0
9
0
3
8
)
(
5
.
7
7
4
6
0
)
(
8
.
0
2
4
3
4
)
(
7
.
5
3
4
4
5
)
(
2
8
.
8
2
5
2
1
)
(
2
.
4
3
5
6
8
)
(
4
.
6
6
0
9
2
)
(
1
.
0
6
8
6
9
)
(
2
.
3
3
3
7
3
)
B
a
n
k
?
x
e
d
e
f
f
e
c
t
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
O
b
s
e
r
v
a
t
i
o
n
s
4
3
4
4
1
4
4
3
3
3
5
2
3
6
4
3
6
2
3
3
3
2
9
6
4
3
4
3
7
7
R
2
0
.
1
4
1
0
0
.
2
3
2
5
0
.
2
6
0
5
0
.
0
8
2
8
0
.
2
3
4
9
0
.
3
4
1
3
0
.
1
7
3
9
0
.
3
5
8
3
0
.
1
2
8
2
0
.
1
6
9
3
N
o
t
e
s
:
S
i
g
n
i
?
c
a
n
t
a
t
:
*
p
,
0
.
1
,
*
*
p
,
0
.
0
5
,
a
n
d
*
*
*
p
,
0
.
0
1
;
O
L
S
e
s
t
i
m
a
t
e
s
,
b
a
n
k
-
s
p
e
c
i
?
c
a
f
f
e
c
t
s
a
r
e
t
r
e
a
t
e
d
a
s
r
a
n
d
o
m
;
r
o
b
u
s
t
s
t
a
n
d
a
r
d
e
r
r
o
r
s
i
n
p
a
r
e
n
t
h
e
s
e
s
;
a
g
r
o
s
s
l
o
a
n
s
r
e
f
e
r
t
o
t
w
o
y
e
a
r
s
p
r
e
v
i
o
u
s
t
o
i
m
p
a
i
r
e
d
l
o
a
n
s
Table III.
ARJ
25,3
196
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
l
i
q
a
s
s
_
t
a
g
l
o
a
n
s
_
t
a
o
t
h
e
r
e
a
_
t
a
o
f
f
b
a
l
_
t
a
d
e
p
_
t
a
l
i
q
_
s
h
t
m
f
l
l
p
r
_
g
l
o
a
n
s
i
m
p
l
o
a
n
s
_
g
l
o
a
n
s
a
r
o
a
r
o
a
_
1
e
q
_
t
a
0
.
1
1
5
3
2
*
*
*
2
0
.
0
8
3
3
5
*
*
0
.
0
8
1
6
7
*
*
0
.
0
0
2
0
9
2
0
.
0
7
0
4
8
*
*
0
.
1
1
2
7
5
2
0
.
0
5
3
3
0
*
*
0
.
0
7
1
5
6
*
*
0
.
0
1
9
3
4
*
*
*
0
.
0
0
7
4
7
*
(
0
.
0
4
1
3
7
)
(
0
.
0
3
7
4
8
)
(
0
.
0
3
6
5
9
)
(
0
.
0
4
5
9
2
)
(
0
.
0
3
1
8
3
)
(
0
.
3
2
0
7
0
)
(
0
.
0
2
3
6
9
)
(
0
.
0
3
5
4
5
)
(
0
.
0
0
6
5
8
)
(
0
.
0
0
4
0
4
)
t
a
2
0
.
0
9
7
7
3
*
*
0
.
0
2
7
9
7
2
0
.
0
3
2
9
2
2
0
.
0
0
0
3
5
2
0
.
0
8
1
2
1
*
*
2
0
.
1
2
7
7
2
2
0
.
0
1
0
5
8
*
2
0
.
0
5
0
8
3
*
*
*
0
.
0
1
0
2
4
*
*
*
0
.
0
0
7
0
9
(
0
.
0
4
8
4
0
)
(
0
.
0
4
5
9
7
)
(
0
.
0
4
7
0
3
)
(
0
.
0
7
9
0
2
)
(
0
.
0
3
6
3
2
)
(
0
.
1
1
4
8
6
)
(
0
.
0
0
5
4
7
)
(
0
.
0
1
6
2
4
)
(
0
.
0
0
3
6
8
)
(
0
.
0
0
5
5
2
)
b
v
d
2
1
.
0
3
5
2
6
*
*
*
0
.
9
3
7
2
5
*
*
*
2
0
.
4
9
2
1
9
*
0
.
4
5
0
8
4
0
.
7
5
3
5
8
*
*
*
2
3
.
2
2
9
4
5
*
*
*
2
0
.
0
4
8
2
3
2
0
.
0
9
3
7
6
0
.
0
8
1
6
0
*
*
*
0
.
0
7
5
6
2
*
*
*
(
0
.
2
2
1
5
7
)
(
0
.
2
8
1
7
7
)
(
0
.
2
5
3
6
9
)
(
0
.
3
3
1
4
7
)
(
0
.
2
3
4
9
7
)
(
1
.
0
4
7
9
6
)
(
0
.
0
3
6
8
1
)
(
0
.
1
5
1
5
2
)
(
0
.
0
2
2
8
4
)
(
0
.
0
2
7
4
0
)
n
e
t
i
r
e
v
_
i
n
c
o
m
e
0
.
0
0
1
3
0
2
0
.
0
0
0
4
3
0
.
0
0
0
2
0
2
0
.
0
0
3
7
2
0
.
0
0
0
9
7
*
*
0
.
0
0
2
0
1
0
.
0
0
0
1
8
0
.
0
0
0
7
4
2
0
.
0
0
0
2
4
2
0
.
0
0
0
2
3
(
0
.
0
0
0
9
3
)
(
0
.
0
0
1
0
2
)
(
0
.
0
0
0
9
9
)
(
0
.
0
0
2
3
9
)
(
0
.
0
0
0
4
1
)
(
0
.
0
0
1
9
2
)
(
0
.
0
0
0
4
3
)
(
0
.
0
0
1
3
9
)
(
0
.
0
0
0
1
8
)
(
0
.
0
0
0
5
0
)
s
t
a
f
f
n
r
_
t
a
5
5
4
.
4
0
2
1
5
2
1
,
2
6
6
.
3
5
5
8
0
*
*
2
1
6
7
.
2
9
5
7
8
2
3
,
1
0
7
.
4
6
4
9
7
*
*
*
6
4
9
.
8
7
1
3
5
2
1
,
1
9
8
.
1
2
8
7
2
0
.
7
9
4
1
1
5
9
3
.
2
0
8
2
0
2
1
9
0
.
7
1
5
*
*
*
2
2
7
4
.
5
7
1
*
*
*
(
6
2
6
.
3
0
7
2
2
)
(
6
3
7
.
6
0
0
2
0
)
(
5
8
7
.
7
5
8
0
9
)
(
5
4
8
.
7
5
5
5
4
)
(
5
5
8
.
2
7
5
6
7
)
(
2
,
2
8
0
.
9
6
9
0
3
)
(
7
1
.
2
9
2
7
9
)
(
4
0
3
.
6
3
6
2
3
)
(
7
0
.
1
7
9
0
7
)
(
6
8
.
5
4
2
5
5
)
s
t
a
f
f
e
x
p
0
.
0
0
5
3
4
2
0
.
0
0
9
4
2
*
*
0
.
0
0
8
0
7
*
*
2
0
.
0
0
0
4
6
2
0
.
0
0
6
4
7
*
0
.
0
4
8
3
1
2
0
.
0
0
2
7
9
2
0
.
0
0
0
2
8
0
.
0
0
2
1
0
*
0
.
0
0
1
6
9
*
(
0
.
0
0
3
7
4
)
(
0
.
0
0
4
4
6
)
(
0
.
0
0
3
9
8
)
(
0
.
0
0
5
8
0
)
(
0
.
0
0
3
3
9
)
(
0
.
0
4
8
6
0
)
(
0
.
0
0
2
5
5
)
(
0
.
0
0
0
9
8
)
(
0
.
0
0
1
1
2
)
(
0
.
0
0
0
8
7
)
d
e
p
r
2
2
.
8
5
4
3
.
8
2
2
6
.
8
6
3
*
2
0
.
0
5
6
2
2
0
.
0
8
5
4
2
2
4
.
2
3
1
.
5
5
8
3
.
7
8
3
*
*
*
0
.
6
4
2
0
.
2
6
8
(
4
.
6
7
9
)
(
3
.
6
8
2
)
(
3
.
5
0
7
)
(
2
.
6
6
8
)
(
2
1
.
6
4
2
)
(
4
5
.
8
1
)
(
2
.
7
8
9
)
(
1
.
0
5
6
)
(
0
.
7
9
8
)
(
0
.
3
7
6
)
i
n
t
9
.
6
7
8
2
4
.
4
8
9
1
.
8
4
2
1
5
.
0
7
2
4
5
.
9
0
*
*
*
5
6
2
.
1
*
*
*
0
.
2
7
5
2
2
.
6
7
9
5
.
4
8
6
*
6
.
9
5
4
*
*
*
(
9
.
5
9
2
)
(
8
.
0
7
4
)
(
7
.
8
9
3
)
(
1
7
.
2
4
)
(
1
1
.
7
7
)
(
1
6
7
.
8
)
(
6
.
8
2
1
)
(
4
.
7
3
0
)
(
3
.
3
5
1
)
(
2
.
7
8
3
)
2
0
0
1
6
.
1
3
6
7
8
*
*
*
2
3
.
5
5
0
0
4
*
*
*
6
.
4
4
1
2
9
*
*
*
2
1
.
4
0
4
9
8
2
0
.
4
5
9
4
3
6
.
5
1
1
7
2
1
.
0
0
8
8
2
*
2
0
.
7
1
2
0
9
*
*
(
1
.
3
0
8
0
7
)
(
1
.
3
1
0
5
8
)
(
1
.
2
4
3
9
5
)
(
1
.
8
8
5
1
6
)
(
0
.
8
9
8
2
5
)
(
3
.
9
6
6
6
1
)
(
0
.
5
7
3
9
0
)
(
0
.
3
0
8
0
2
)
2
0
0
2
6
.
3
1
3
9
4
*
*
*
2
3
.
5
0
3
5
2
*
*
*
7
.
1
7
7
3
7
*
*
*
2
2
.
2
3
0
8
3
0
.
2
0
6
8
2
1
1
.
7
3
0
1
2
*
2
0
.
1
6
5
1
3
2
0
.
3
2
9
0
6
2
0
.
7
4
0
9
1
*
*
*
(
1
.
2
5
2
8
4
)
(
1
.
1
5
5
2
3
)
(
1
.
0
8
0
6
9
)
(
1
.
7
9
2
2
0
)
(
0
.
8
4
2
3
4
)
(
7
.
0
5
3
1
3
)
(
0
.
7
8
5
6
0
)
(
0
.
2
1
3
6
4
)
(
0
.
2
0
2
7
4
)
2
0
0
3
3
.
7
3
7
3
4
*
*
*
2
1
.
3
9
4
2
8
3
.
7
3
8
3
4
*
*
*
2
1
.
5
4
0
8
7
0
.
1
2
8
7
3
1
1
.
9
5
4
7
9
*
0
.
0
1
3
8
1
3
.
1
7
7
6
8
*
*
*
2
0
.
0
0
8
6
4
2
0
.
3
6
8
0
9
*
(
0
.
9
0
8
1
5
)
(
0
.
8
8
8
5
6
)
(
0
.
8
7
8
0
1
)
(
1
.
5
4
4
7
8
)
(
0
.
6
9
2
0
0
)
(
7
.
2
2
2
7
2
)
(
0
.
4
8
7
9
1
)
(
0
.
7
0
8
3
6
)
(
0
.
1
7
8
6
8
)
(
0
.
1
9
9
5
9
)
2
0
0
4
2
.
6
9
2
9
7
*
*
*
2
1
.
6
3
3
5
4
*
*
*
2
.
2
1
0
7
6
*
*
*
2
0
.
1
7
4
6
7
2
0
.
5
4
9
9
8
6
.
0
3
4
8
7
2
0
.
5
1
5
2
9
1
.
5
0
2
6
3
*
0
.
0
4
2
1
9
2
0
.
3
6
0
8
7
*
(
0
.
6
6
8
0
4
)
(
0
.
5
9
9
4
4
)
(
0
.
6
9
8
0
1
)
(
1
.
2
4
4
6
9
)
(
0
.
6
1
3
4
0
)
(
5
.
9
2
9
1
5
)
(
0
.
6
1
6
3
0
)
(
0
.
8
2
6
8
3
)
(
0
.
2
3
9
0
2
)
(
0
.
1
9
6
7
2
)
2
0
0
6
1
.
2
1
4
8
0
*
*
0
.
0
9
4
0
1
0
.
0
5
8
1
7
2
0
.
6
9
2
4
8
2
0
.
4
9
1
1
8
2
0
.
8
7
2
4
4
0
.
0
4
5
9
6
2
0
.
2
7
0
8
7
2
0
.
1
1
0
4
3
2
0
.
2
7
7
5
2
(
0
.
5
6
8
2
2
)
(
0
.
5
4
9
1
9
)
(
0
.
6
2
1
3
9
)
(
2
.
3
2
9
5
4
)
(
0
.
4
8
9
1
1
)
(
3
.
8
4
8
6
0
)
(
0
.
5
2
2
4
7
)
(
0
.
5
6
8
4
9
)
(
0
.
1
6
6
9
8
)
(
0
.
1
9
3
4
5
)
(
c
o
n
t
i
n
u
e
d
)
Table IV.
Effect of capitalization on
risk and pro?tability:
conventional banks
Capitalization
197
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
l
i
q
a
s
s
_
t
a
g
l
o
a
n
s
_
t
a
o
t
h
e
r
e
a
_
t
a
o
f
f
b
a
l
_
t
a
d
e
p
_
t
a
l
i
q
_
s
h
t
m
f
l
l
p
r
_
g
l
o
a
n
s
i
m
p
l
o
a
n
s
_
g
l
o
a
n
s
a
r
o
a
r
o
a
_
1
2
0
0
7
0
.
0
3
0
9
8
0
.
6
2
0
2
7
2
2
.
2
8
5
3
7
*
*
*
1
.
8
5
0
0
3
2
0
.
5
7
1
5
6
2
4
.
2
4
2
8
2
0
.
5
9
3
1
9
2
1
.
2
4
2
0
1
*
2
0
.
2
2
9
8
1
2
0
.
3
6
2
2
2
*
(
0
.
8
5
1
6
7
)
(
0
.
6
8
0
7
5
)
(
0
.
7
8
7
3
0
)
(
2
.
3
4
2
9
3
)
(
0
.
5
6
4
9
8
)
(
3
.
0
8
4
0
8
)
(
0
.
5
4
5
4
8
)
(
0
.
6
5
3
2
4
)
(
0
.
1
6
8
0
1
)
(
0
.
1
9
1
1
8
)
2
0
0
8
2
5
.
5
1
8
0
2
*
*
*
4
.
6
9
7
7
7
*
*
*
2
5
.
4
1
8
3
2
*
*
*
1
.
0
9
8
6
2
0
.
7
1
2
2
9
2
1
5
.
3
4
4
2
7
*
*
*
0
.
7
5
3
1
7
2
0
.
6
9
4
4
0
2
0
.
8
4
5
3
8
*
2
1
.
0
6
2
6
2
*
*
*
(
1
.
0
0
3
1
3
)
(
0
.
8
7
6
6
3
)
(
0
.
9
0
6
2
9
)
(
2
.
3
1
7
7
3
)
(
0
.
5
9
9
8
9
)
(
4
.
7
6
1
4
1
)
(
0
.
6
9
9
0
7
)
(
0
.
7
9
6
2
3
)
(
0
.
4
6
6
2
0
)
(
0
.
1
9
1
4
3
)
2
0
0
9
2
4
.
0
4
3
5
4
*
*
*
2
.
3
3
5
8
4
*
*
2
3
.
0
2
0
7
8
*
*
*
2
3
.
7
7
3
8
8
*
1
.
8
8
4
7
8
*
*
*
2
1
0
.
4
6
1
7
8
*
*
0
.
5
3
4
9
6
2
0
.
6
9
9
7
3
2
0
.
5
0
3
9
9
*
*
*
2
0
.
8
7
8
8
4
*
*
*
(
1
.
0
5
4
9
9
)
(
0
.
9
9
1
1
8
)
(
1
.
0
2
4
0
4
)
(
1
.
9
7
6
7
6
)
(
0
.
6
9
5
8
0
)
(
4
.
9
2
3
9
8
)
(
0
.
5
4
5
5
7
)
(
0
.
7
3
5
8
0
)
(
0
.
1
8
5
2
1
)
(
0
.
1
9
2
2
1
)
2
0
1
0
2
3
.
7
7
1
6
4
*
*
*
3
.
4
8
8
7
6
*
*
*
2
5
.
0
7
4
0
2
*
*
*
2
5
.
2
8
8
5
8
*
*
1
.
6
4
1
1
9
*
*
2
1
0
.
0
2
6
5
7
0
.
0
6
7
2
5
2
1
.
1
6
0
8
8
2
0
.
3
3
8
4
5
*
2
0
.
7
2
8
5
7
*
*
*
(
1
.
0
6
1
9
7
)
(
1
.
0
4
8
7
3
)
(
1
.
1
2
8
8
2
)
(
2
.
2
8
9
0
3
)
(
0
.
7
7
6
0
3
)
(
6
.
2
1
5
8
4
)
(
0
.
5
0
8
6
1
)
(
0
.
7
3
2
5
6
)
(
0
.
1
8
5
2
9
)
(
0
.
1
9
2
5
0
)
2
0
1
1
2
3
.
8
9
4
8
4
*
*
*
3
.
9
7
2
6
3
*
*
*
2
5
.
7
7
2
0
9
*
*
*
2
6
.
8
8
4
6
9
*
*
*
1
.
0
8
2
8
7
2
1
0
.
7
9
6
3
7
*
*
2
0
.
0
8
9
4
0
2
1
.
0
1
2
8
8
2
0
.
4
4
9
4
8
*
*
2
0
.
9
2
2
6
6
*
*
*
(
1
.
0
3
5
6
0
)
(
1
.
0
2
1
2
0
)
(
1
.
1
6
7
5
3
)
(
2
.
0
7
2
2
7
)
(
0
.
8
6
0
3
5
)
(
4
.
5
1
1
6
4
)
(
0
.
4
9
1
7
3
)
(
0
.
7
6
0
4
7
)
(
0
.
1
8
4
8
1
)
(
0
.
2
1
7
6
3
)
C
o
n
s
t
a
n
t
3
2
.
3
0
6
0
0
*
*
*
4
9
.
4
1
1
6
6
*
*
*
3
8
.
3
2
9
3
9
*
*
*
3
1
.
3
5
2
1
5
*
*
*
6
6
.
1
1
6
0
0
*
*
*
7
1
.
8
8
3
9
0
*
*
*
2
.
2
4
4
7
0
*
*
*
7
.
5
2
3
1
7
*
*
*
0
.
9
2
8
0
2
*
*
*
1
.
9
1
3
4
7
*
*
*
(
1
.
9
8
2
4
6
)
(
2
.
1
2
2
0
4
)
(
1
.
9
2
3
8
2
)
(
3
.
1
9
7
1
3
)
(
1
.
9
2
7
8
4
)
(
9
.
9
1
0
0
6
)
(
0
.
7
6
6
5
6
)
(
1
.
2
0
0
0
8
)
(
0
.
2
6
9
5
8
)
(
0
.
2
3
5
3
3
)
B
a
n
k
?
x
e
d
e
f
f
e
c
t
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
O
b
s
e
r
v
a
t
i
o
n
s
2
,
3
3
0
2
,
3
2
5
2
,
3
3
0
2
,
0
8
5
2
,
3
1
1
2
,
3
0
9
2
,
2
5
4
1
,
7
4
2
2
,
3
3
0
2
,
0
2
9
R
2
0
.
1
5
4
4
0
.
1
4
3
8
0
.
1
6
8
6
0
.
0
5
4
0
0
.
1
4
5
1
0
.
1
0
9
9
0
.
2
3
0
6
0
.
1
1
7
5
0
.
1
2
6
5
0
.
1
1
4
8
N
o
t
e
s
:
S
i
g
n
i
?
c
a
n
t
a
t
:
*
p
,
0
.
1
,
*
*
p
,
0
.
0
5
,
a
n
d
*
*
*
p
,
0
.
0
1
;
O
L
S
e
s
t
i
m
a
t
e
s
,
b
a
n
k
-
s
p
e
c
i
?
c
a
f
f
e
c
t
s
a
r
e
t
r
e
a
t
e
d
a
s
r
a
n
d
o
m
;
r
o
b
u
s
t
s
t
a
n
d
a
r
d
e
r
r
o
r
s
i
n
p
a
r
e
n
t
h
e
s
e
s
;
a
g
r
o
s
s
l
o
a
n
s
r
e
f
e
r
t
o
t
w
o
y
e
a
r
s
p
r
e
v
i
o
u
s
t
o
i
m
p
a
i
r
e
d
l
o
a
n
s
Table IV.
ARJ
25,3
198
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
the ?rst column of Table III but it is considerably smaller that the correspondent one
reported in Table III). More capitalized conventional banks, however, accumulate less
reserves (the parameter associated to eq_ta inthe seventh columnof Table IVis negative
and signi?cant) although impaired loans tend to increase (eighth column, Table IV).
From to the negative parameter associated to deposits, we can also infer that
non-traditional assets of conventional banks are likely to be ?nanced through sources
different from customers’ deposits, plausibly bank securities.
Although perhaps intended for handling risk, the strategy of conventional banks as a
consequence of recapitalization seems rather unsafe, although pro?table, especially in the
very short-run. One possible rationale for this behaviour is that increasing the share of
equity over other funding sources calls for higher stock rewards and hence investing in
more pro?table activities. In fact, the return on asset positively responds to a
contemporaneous increase incapitalization, althoughturningalmost negligible afterwards.
Turning to other determinants of banks’ asset structure, we can observe that larger
Islamic banks (in terms of total assets) tend to operate more through traditional lending
(in terms of a higher share of gross loans-to-assets). Off-balance sheets items are also
higher in large Islamic banks. Interesting evidence is that a positive association
between dimension and pro?tability (roa) holds for Islamic intermediaries, a result
which seems in line, for example, with what Bashir (1999) ?nds on Sudanese Islamic
banks. Larger conventional banks seem instead to re?ect a different situation since
they appear more ef?cient and highly pro?table (lower shares of liquidity and reserve
provision, less non-performing loans, higher return on assets).
An important element that consistently affects returns in both types of banks is
ownership dispersion, which does not seem a questionable issue according to the
literature. Overstaffed banks, regardless whether they are Islamic or conventional,
show signi?cantly lower returns on their assets, which is perhaps due to inef?ciencies.
Also, banks granting better staff compensation seem to be associated to more
intensi?ed non-traditional activities (high other earning assets), which may involve
more tailored products for their customers and therefore require higher skills. This
situation seems somehow similar in both types of banks.
The economic cycle also has a clear impact on pro?tability in both Islamic and
conventional banking systems since year dummies in both the ninth and tenth columns
become dramatically negative since the beginning of the crisis. Such a reduction of
pro?tability seems to have involved Islamic banks more severely than conventional
ones. Reasonably for the same facts, Islamic banks have also increased loan loss
provisions in 2009-2010.
Finally, the effects of the crisis seem having brought back conventional banks to
operate more on traditional activities consisting in collecting deposits and lending
(the parameters associated to the share of loans-to-assets is positive since 2008), while
this phenomenon does not seem to have affected the balance sheet structure of Islamic
banks to a large extent.
5.2 Pro?t ef?ciency
Table V reports the stochastic translog pro?t frontier parameters for equation (2)
estimated through the maximum-likelihood model. Overall, parameters have the
expected sign where signi?cant. The only variable which does not signi?cantly affect
pro?ts is the amount of other earning assets (lnotherea). On the other hand, the cost
Capitalization
199
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Pro?t frontier
Dependent variable: log of income
lnloans 0.57739
* * *
(0.07185)
lnotherea 0.11420 (0.08352)
lnoffbal 0.10422
*
(0.05831)
lndepr 20.71826
* * *
(0.05574)
lnint 20.34760
* * *
(0.07756)
lnstaffexp 0.62933
* * *
(0.07659)
1/2lnloans
*
lnloans 0.06076
* * *
(0.01028)
1/2lnotherea
*
lnotherea 20.06053
* * *
(0.01269)
1/2lnoffbal
*
lnoffbal 20.01530
* * *
(0.00510)
lnloans
*
lnotherea 20.07864
* * *
(0.00933)
lnloans
*
lnoffbal 0.01412
* *
(0.00656)
lnotherea
*
lnoffbal 0.00614 (0.00707)
1/2lndepr
*
lndepr 20.07547
* * *
(0.00380)
1/2lnint
*
lnint 20.09690
* * *
(0.00633)
1/2lnstaffexp
*
lnstaffexp 0.01076
*
(0.00628)
lndepr
*
lnint 0.18314
* * *
(0.00504)
lndepr
*
lnstaffexp 20.03220
* * *
(0.00773)
lnint
*
lnstaffexp 0.01067 (0.01164)
lnloans
*
lndepr 0.05568
* * *
(0.00575)
lnloans
*
lnint 20.08719
* * *
(0.00804)
lnloans
*
lnstaffexp 0.03151
* * *
(0.00755)
lnotherea
*
lndepr 20.07008
* * *
(0.00688)
lnotherea
*
lnint 0.13979
* * *
(0.00939)
lnotherea
*
lnstaffexp 20.06971
* * *
(0.00850)
lnoffbal
*
lndepr 20.00333 (0.00487)
lnoffbal
*
lnint 0.00295 (0.00806)
lnoffbal
*
lnstaffexp 0.00038 (0.00614)
Constant 20.04240
* * *
(5.85727)
lnsigma2 2.18229
* * *
(0.07956)
gamma 0.26994 (0.58538)
mu 0.91347
* * *
(0.11557)
eta 20.00467 (0.01005)
LR test one side error 6,583.43
Log likelihood 2239.94689
Regression of correlates with pro?t ef?ciency scores
Dependent variable: pro?t ef?ciency
eq_ta
*
isl 20.00055 (0.00064)
eq_ta
*
conv 0.00122
* * *
(0.00033)
isl 20.00003 (0.00146)
log ta 0.02375
* * *
(0.00333)
bvd 0.00402
* *
(0.00183)
netirev_income 20.00003
*
(0.00001)
staffnr_ta
a
13.11555 (14.69708)
Constant 0.320692
* * *
(0.005477)
LR test one side error 6583.43
Log likelihood 2239.94689
Note: Signi?cant at:
*
p , 0.1,
* *
p , 0.05, and
* * *
p , 0.01; year, bank, and country ?xed effects
included;
a
parameter and standard error divided by 100
Table V.
Effect of capitalization
on pro?t ef?ciency
ARJ
25,3
200
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
of labour (lnstaffexp) is surprisingly positive in the pro?t function, although this
anomaly is not new to the empirical literature on Islamic bank ef?ciency (Srairi, 2010).
Estimates of equation (3) are also reported in Table V. In this case, parameters allow
comparing the impact of the capital structure on ef?ciency in the two banking systems,
as well as drawing some useful conclusions on the ef?cient use of inputs.
Turning to the estimation of equation (3), the most relevant outcome is the
signi?cance of parameter a
2
, which is associated to the interaction term between
equity over total assets and the dummy variable for conventional banks (eq_ta
*
conv).
Its positive sign indicates that, controlling for other elements of ef?ciency related to the
type of banking, increasing capitalization in conventional banks leads to higher pro?t
ef?ciency. According to the regression output discussed above, this possibly suggests
that equity, since requiring higher rewards than other sources of funds, may induce
more pressure on management to generate income.
Such a result does not hold, however, for Islamic banks. In fact, the parameter a
1
,
associated to the interaction term between equity over total assets and the dummy
variable for Islamic banks (eq_ta
*
isl), is not signi?cant. This can be an indication that
differences between depositors and shareholders provide less remarkable effects
compared to conventional banks.
Interesting evidence also concerns the dummy for Islamic banks taken alone.
Indeed, contrary to what is commonly argued, Islamic banks do not seem to be less
ef?cient than their conventional counterparts since the coef?cient associated to the
Islamic bank dummy variable (isl) is not signi?cant.
Important effects on ef?ciency are also provided by both banks’ dimension and
ownership dispersion. The coef?cients associated to the logarithm of total assets
(log ta) and to the ownership dispersion index (bvd) are indeed positive and highly
signi?cant, suggesting that larger banks having a structure closer to public companies
tend to be more ef?cient, which in line with the literature on bank ef?ciency (Berger and
Bonaccorsi di Patti, 2006).
Finally, producing a higher share of income through traditional lending activity
may reduce ef?ciency. The – although weakly – negative sign of the variable
netirev_income, in particular, seems mostly in line with the strategy of conventional
banks, namely shifting to less traditional activities in order to boost pro?tability. Such
a result, which does not look particularly surprising, is likely to be driven by practices
adopted in the pre-crisis period. The assets related to such activities, however, recently
ceased to yield abnormal returns and have been discarded in some cases.
6. Concluding remarks
Pro?t and loss sharing principles, by reformulating the allocation of risk between
shareholders and depositors, may affect the incentives of both depositors to monitor the
conduct of banking activity and managers to undertake risky business. On the one hand,
under Islamic banking principles a number of depositors supply investment accounts and
participate in the bank investment activities through risk-sharing schemes, thus,
in principle, requiring less protection than conventional bank depositors (Ariss and
Sarieddine, 2007). Onthe other hand, the possibilityof shiftingrisks ondepositors raises the
risk of moral hazard, so that all investors should look for a higher degree of safekeeping.
An appropriate capital structure is normally recommended in order to assure
the optimal level of protection to stakeholders from excessive risk undertaking.
Capitalization
201
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
From the perspective of Islamic banks, however, it is not clear whether several forces
acting in different directions are likely to shape the “ideal” recommendations in terms of
bank capital structure similar, for example, to those set up by the Basel Committee for
European banks.
Given this debate, in this paper we have focused on the impact of capitalization on
both Islamic and conventional banks’ risk-taking ef?ciency in the period 2001-2011 in
countries where both bank types are available.
Results, obtained by regressions investigating bank capital structure with attention
to risk undertaking exposure have shown that more capitalized Islamic banks are
likely to pursue safer policies than their conventional counterparts. Increasing liquid
assets is an example. This has the effect of reducing non-performing loans, with weak
bene?ts on returns.
On the other hand, more capitalized conventional banks tend to shift from
traditional lending activities to investment in other pro?t-generating assets. However,
we have observed that this policy raises impaired loans, although not offsetting the
previous effect. Thus, pro?tability boosts in correspondence to an increase in
capitalization, although becoming almost negligible afterwards.
We have also used stochastic frontier techniques to complete the analysis of
ef?cient risk undertaking. We have found that conventional banks show a positive
relationship between capitalization and pro?t ef?ciency, whereas there are no similar
effects in Islamic banks. This suggests that equity, compared to deposits, is likely to
represent a stinging blow to banks to perform, but the similarities between stocks and
deposits in the Islamic banking context may de?nitely smooth this effect.
Finally, the implications of this analysis are important from a regulatory
perspective. First, it provides an empirical evaluation of an issue which does not have
any theoretical univocal answer, since there are valid arguments to suggest either
increasing or decreasing capital standards in Islamic banks. Second, predictions are in
line with regulatory provisions set up in some European countries in relation to the
application of Basel principles to Islamic banks (or branches of Islamic banks)
operating in these countries. It is not a case that, for example, in the UK the FSA
declared that when capital is not an appropriate mitigating tool for Islamic banks, then
other ways of managing risk are to be set up. These include, for example,
recommendations in favour of sound corporate governance mechanisms and other
appropriate systems of control (Ainley et al., 2007).
Notes
1. During economic slumps, low capitalization standards and inef?cient risk undertaking also
appear to be a good proxy for bank failures (Beim and Calomiris, 2001; Estrella et al., 2000).
2. It has been observed that most of the contrasting opinions in the empirical literature may
stem from the use of various measures of performance, going from basic accounting ratios to
more sophisticated ones. On this point, it is increasingly becoming acknowledged that “it is
necessary to recognize explicitly the concept of ef?ciency as a chief measure of performance
in the empirical models linking bank capital to risk, in such a way that it becomes possible to
distinguish between ef?cient and inef?cient risk undertaking” (Hughes and Moon, 1995;
Kwan and Eisenbeis, 1997; Mathuva, 2009; Weill, 2002).
3. Although Islamic bank are increasingly recently shying away from strictly applying
Mudaraba and Musharaka in their lending activity in favor of more standardized products,
ARJ
25,3
202
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
these principles still characterize the structure of the liability side (see for example Hassan
and Lewis, 2007 for a discussion on the evolution of Islamic products).
4. See also Ariff (1988) for an overview of Islamic banking principles.
5. Islamic deposits can range from current accounts to saving and investment deposits, and
other special investment accounts. As in conventional banks, current accounts do not receive
any remuneration and are exempted from losses. In all other cases, deposits are typically
pooled by the bank under a Mudarabah agreement to provide funds to its customers. See
Lewis and Algaud (2001) for details on these and other types of contacts.
6. In this group we consider only pure commercial Islamic banks, i.e. those providing
Shariah-compatible products. Mixed banks with Islamic services are not included.
7. Although a parallel analysis has been conducted limited to the period previous to the recent
?nancial crisis (results, which are not substantially different from the ones reported in this
paper, are available upon request), here we also include years 2009-2011. In order to allow
this extension we accurately dealt with outliers encountered, in particular, in the last three
years of the overall time span.
8. We follow C
?
iha´k and Hesse (2010) including Bahrain, Bangladesh, Brunei, Egypt, Indonesia,
Iran, Jordan, Kuwait, Lebanon, Malaysia, Pakistan, PalestinianTerritory, Qatar, Saudi Arabia,
Sudan, Tunisia, United Arab Emirates, and Yemen. We are forced to exclude Gambia and
Mauritania since only one Islamic bank in each country provided data to Bankscope in the
time span considered. Like C
?
iha´k and Hesse we keep Iran and Sudan in the sample due to the
weight of their Islamic bank sector, although being conscious of the fact that proper
conventional bankingis not allowedinthese countries. We also include Turkey, Iraq andSyria
because of the relevance of both their Islamic and conventional banking systems, although
estimates which exclude these countries do not provide substantially different results
compared to those presented further in this paper. Weill (2011) uses a similar set of countries
(which includes Sudan and Iran) in a study performed on “all the countries where Islamic
banks and conventional banks coexisted during the period 2000-2007” (Weill, 2011).
9. Note that there is no form of selection bias in the sample, except from information
availability at the time of data collection.
10. Including non-performing loans.
11. Islamic banks have not traditionally held off-balance sheet exposures as is instead the case
conventional banks. Furthermore, the cost of holding off-balance sheet items may be higher
in Islamic banks since a number of these off-balance sheet items require more procedures
which in turn lead to higher costs.
12. Interests are implicitly computed for Islamic banks.
13. We also carried out parallel regressions where bank ?xed effects have been replaced by
country ?xed effects. Results are not substantially different from those reported in the next
section, apart from the higher ef?ciency of parameters associated to some control variables
when using country ?xed effects.
14. Year 2005 has been chosen as a residual category due to reasons related to the economic
cycle, so as to ease the interpretation of all other time dummies by comparison with the
upper-limit for economic expansion in our time span.
15. Hausman (1978) test supports the hypothesis of absence of correlation in all regressions
presented in the next section.
16. We use pro?t ef?ciency rather than cost ef?ciency because, as highlighted by the literature
(see Berger and Bonaccorsi di Patti, 2006 for a discussion), when dealing with agency
Capitalization
203
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
problems it is better suited to account for how well managers raise revenues as well as
control costs, so as to maximize corporate value.
17. Schmidt and Sickles (1984), among others, note that cross-sectional models require that the
inef?ciency component be independent from the regressors, although this hypothesis may
often be violated.
18. See for example Greene (1980) and Carvallo and Kasman (2005) for a discussion on the
advantages of using the translog speci?cation.
19. See also Aigner et al. (1977), Jondrow et al. (1982) and Lee and Schmidt (1993).
20. See Kumbhakar and Lovell (2000) for further details.
21. See also Coelli (1996).
References
Aigner, D.J., Lovell, C.A.K. and Schmidt, P. (1977), “Formulation and estimation of stochastic
frontier production function models”, Journal of Econometrics, Vol. 6 No. 1, pp. 21-37.
Ainley, M., Mashayekhi, A., Hicks, R., Rahman, A. and Ravalia, A. (2007), Islamic Finance in the
UK: Regulation and Challenges, Financial Services Authority, London.
Al-Fayoumi, N.A. and Alkour, E.M. (2008), “Cost and pro?t ef?ciency in the Jordan commercial
banks: parametric and non-parametric approaches for estimating ef?ciency”, Dirasat,
Vol. 35, pp. 21-38.
Al-Jarrah, I. and Molyneux, P. (2003), “Cost ef?ciency, scale elasticity and scale economies in
Arabian banking”, paper presented at 10th Annual Conference of Economic Research
Form for the Arab Countries, Iran and Turkey in Morocco, December 16-18.
Ariff, M. (1988), “Islamic banking”, Asian-Paci?c Economic Literature, Vol. 2 No. 2, pp. 48-64.
Ariss, R.T. and Sarieddine, Y. (2007), “Challenges in implementing capital adequacy guidelines to
Islamic banks”, Journal of Banking Regulation, Vol. 9 No. 1.
Awdeh, A., El-Moussawi, C. and Machrouh, F. (2011), “The effect of capital requirements on
banking risk”, International Research Journal of Finance and Economics, Vol. 66, pp. 133-46.
Bashir, A.H.M. (1999), “Risk and pro?tability measures in Islamic banks: the case of two
Sudanese banks”, Islamic Economic Studies, Vol. 6 No. 2, pp. 1-24.
Battese, G.E. and Coelli, T.J. (1995), “A model for technical inef?ciency effects in stochastic
frontier production function for panel data”, Empirical Economics, Vol. 20, pp. 325-32.
Beim, D.O. and Calomiris, C.W. (2001), Emerging Financial Markets, McGraw-Hill, NewYork, NY.
Berger, A. and Mester, L.J. (1997), “Inside the black box: what explains differences in the
ef?ciencies of ?nancial institutions?”, Journal of Banking and Finance, Vol. 21, pp. 895-947.
Berger, A.N. and Bonaccorsi di Patti, E. (2006), “Capital structure and ?rm performance: a new
approach to testing agency theory and an application to the banking industry”, Journal of
Banking and Finance, Vol. 30 No. 4, pp. 1065-102.
Berger, A.N. and Humphrey, D.B. (1997), “Ef?ciency of ?nancial institutions: international
survey and directions for future research”, Center for Financial Institutions
Working Papers 97-05, Wharton School Center for Financial Institutions, University of
Pennsylvania, Philadelphia, PA.
Carvallo, O. and Kasman, A. (2005), “Cost ef?ciency in the Latin American and Caribbean banking
systems”, International Financial Market Institutions and Money, Vol. 15, pp. 55-72.
C
?
iha´k, M. and Hesse, H. (2010), “Islamic banks and ?nancial stability: an empirical analysis”,
Journal of Financial Services Research, Vol. 38 No. 2, pp. 95-113.
ARJ
25,3
204
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Coelli, T. (1996), “A guide to frontier version 4.1: a computer program for stochastic frontier
production and cost function estimation”, CEPA Working Paper.
Dahl, D. and Shrieves, R.E. (1990), “The impact of regulation on bank equity infusions”, Journal
of Banking and Finance, Vol. 6, pp. 1209-28.
Editz, T., Michael, I. and Perraudin, W. (1998), “The impact of capital requirements on UK bank
behaviour”, Federal Reserve Bank of New York Policy Review, Vol. 4 No. 3, pp. 15-22.
Estrella, A., Park, S. and Peristiani, S. (2000), “Capital ratios as predictors of bank failure”,
Federal Reserve Bank of New York Economic Policy Review, Vol. 6 No. 2, pp. 33-52.
Greene, W.H. (1980), “On the estimation of a ?exible frontier production model”, Journal of
Econometrics, Vol. 13, pp. 101-15.
Hasan, Z. (2004), “Measuring the ef?ciency of Islamic banks: criteria, methods and social
priorities”, Islamic Economic Review, Vol. 8 No. 2, pp. 1-30.
Hassan, K.M. (2005), “The cost, pro?t and X-ef?ciency of Islamic banks”, Proceeding of the
Economic Research Forum 12th Annual Conference, Cairo, Egypt, working paper.
Hassan, K.M. and Lewis, M.K. (2007), Handbook of Islamic Banking, Edward Elgar, Cheltenham.
Hausman, J. (1978), “Speci?cation tests in econometrics”, Econometrica, Vol. 46, pp. 1251-72.
Hesse, H., Jobst, A. and Sole, J. (2008), “Trends and challenges in Islamic ?nance”, World
Economics, Vol. 9 No. 2, pp. 175-93.
Hughes, J. and Moon, C.G. (1995), “Measuring bank ef?ciency when managers trade return for
reduced risk”, working paper, Department of Economics, Rutgers University,
New Brunswick, NJ.
Iqbal, M. and Llewellyn, D.T. (2002), Islamic Banking and Finance: New Perspective on
Pro?t-Sharing and Risk, Edward Elgar, Cheltenham.
Jensen, M.C. (1986), “Agency costs of free cash ?ow, corporate ?nance and takeovers”, American
Economic Review, Vol. 76, pp. 323-9.
Jensen, M.C. and Meckling, W.H. (1976), “Theory of the ?rm: managerial behavior, agency costs
and ownership structure”, Journal of Financial Economics, Vol. 4, pp. 305-60.
Jondrow, J., Lovell, C.A.K., Materov, I.S. and Schmidt, P. (1982), “On estimation of technical
inef?ciency in the stochastic frontier production function model”, Journal of Econometrics,
Vol. 19 Nos 2/3, pp. 233-8.
Kader, J.M., Asaporta, A.J. and Al-Maghaireh, A. (2007), “Comparative ?nancial performance of
Islamic banks vis-a`-vis conventional banks in the UAE”, Proceeding of the Annual Student
Research Symposium and the Chancellor’s Undergraduate Research Award, available at:
http://sra.uaeu.ac.ae/CURA/Proceedings
Kahane, M.C. (1977), “Capital adequacy and the regulation of ?nancial intermediaries”, Journal of
Banking and Finance, Vol. 1, pp. 207-18.
Kamaruddin, B.H., Safa, M.S. and Mohd, R. (2008), “Assessing production ef?ciency of Islamic
banks and conventional bank Islamic windows in Malaysia”, International Journal of
Business and Management, Vol. 1 No. 1, pp. 31-48.
Kim, D. and Santomero, A.M. (1988), “Risk in banking and capital regulation”, Journal of
Finance, Vol. 43, pp. 1219-33.
Koehn, M. and Santomero, A.M. (1980), “Regulation of bank capital and portfolio risk”, Journal of
Finance, Vol. 35, pp. 1235-44.
Kumbhakar, S.C. and Lovell, C.A.K. (2000), Stochastic Frontier Analysis, Cambridge University
Press, Cambridge.
Capitalization
205
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Kwan, S. and Eisenbeis, R.A. (1997), An Analysis of Inef?ciencies in Banking: A Stochastic
Frontier Approach, Federal Reserve Bank of San Francisco, San Francisco, CA.
Lee, Y.H. and Schmidt, P. (1993), “A production frontier model with ?exible temporal variation in
technical inef?ciency”, in Fried, H.O., Lovell, C.A.K. and Schmidt, S.S. (Eds),
The Measurement of Productive Ef?ciency: Techniques and Applications, Oxford
University Press, New York, NY.
Lewis, M.K. and Algaud, L.M. (2001), Islamic Banking, Edward Elgar, Cheltenham.
Mathuva, D.M. (2009), “Capital adequacy, cost income ratio and the performance of commercial
banks: the Kenyan scenario”, International Journal of Applied Economics and Finance,
Vol. 3, pp. 35-47.
Maudos, J., Pastor, J.M., Perez, F. andQuesada, J. (2002), “Cost andpro?t ef?ciencyinEuropeanbanks”,
Journal of International Financial Markets, Institutions and Money, Vol. 12 No. 1, pp. 33-58.
Modigliani, F. and Miller, M.H. (1958), “The cost of capital, corporation ?nance, and the theory of
investment”, American Economic Review, Vol. 48, pp. 261-97.
Moktar, H.S., Abdullah, N. and Al-Habshi, S.M. (2006), “Ef?ciency of Islamic banks in Malaysia:
a stochastic frontier approach”, Journal of Economic Cooperation Among Islam Countries,
Vol. 27 No. 2, pp. 37-70.
Myers, S.C. (1977), “Determinants of corporate borrowing”, Journal of Financial Economics, Vol. 5,
pp. 147-75.
Niazi, G.S.K., Raoof, A., Rehman, R. and Shahid, H. (2010), “Ef?ciencies comparison of Islamic
and conventional banks of Pakistan”, International Research Journal of Finance and
Economics, Vol. 49.
Qayyum, A. (2007), Financial Sector Reforms and the Ef?ciency of Banking in Pakistan,
South Asian Network of Economic Research Institutes (SANEI), Dhaka, available at:
www.saneinetwork.net
Schmidt, P. and Sickles, R.C. (1984), “Production frontiers and panel data”, Journal of Business
and Economic Statistics, Vol. 2 No. 4, pp. 367-74.
Short, H. (1994), “Ownership, control, ?nancial structure, and the performance of ?rms”, Journal
of Economic Surveys, Vol. 8 No. 3, pp. 203-49.
Srairi, S. (2010), “Cost and pro?t ef?ciency of conventional and Islamic banks in GCC countries”,
Journal of Productivity Analysis, Vol. 34 No. 1, pp. 45-62.
Stiglitz, J. (2001), “Principles of ?nancial regulation: a dynamic portfolio approach”, The World
Bank Research Observer, Spring, pp. 1-18.
Su?an, F. (2007), “The ef?ciency of Islamic banking industry in Malaysia: foreign vs domestic
banks”, International Journal of Systems and Ethics, Vol. 23 No. 3, pp. 174-92.
Sundararajan, V. and Errico, L. (2002), “Islamic ?nancial institutions and products in the global
?nancial system: key issues in risk management and challenges ahead”, IMF Working
Paper No. 02/192, International Monetary Fund, Washington, DC.
Weill, L. (2002), “Leverage and corporate performance: a frontier ef?ciency analysis on European
countries”, Universite´ Robert Schuman, Institut d’Etudes Politiques.
Weill, L. (2011), “Do Islamic banks have greater market power?”, Comparative Economic Studies,
Vol. 53 No. 2, pp. 291-306.
Yudistira, D. (2004), “Ef?ciency in Islamic banking: an empirical analysis of eighteen banks”,
Islamic Economic Studies, Vol. 12 No. 1, pp. 1-19.
ARJ
25,3
206
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Further reading
Hussein, K.A. (2004), Operational Ef?ciency in Islamic Banking: The Sudanese Experience,
Islamic Development Bank, Jeddah.
About the author
Lucia Dalla Pellegrina is an Assistant Professor at the University of Milan-Bicocca where she
teaches Macroeconomics and Advanced Macroeconomics. She is Fellow of the Paolo Baf? Center
for Economic and Financial Research, Bocconi University and a faculty member of the PhD in
Economics at the University of Pavia where she teaches Law and Finance. Lucia Dalla Pellegrina
can be contacted at: [email protected]
Capitalization
207
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_606921860.pdf
In light of the current debate on bank capital requirements, the purpose of this paper is to
investigate the relative impact of capitalization on risk-taking efficiency in Islamic and conventional
banks. The author tests whether changes occurring to the capital structure of such different types of
intermediaries unevenly affect their behaviour in terms of risk-taking efficiency.
Accounting Research Journal
Does capitalization enhance efficient risk undertaking?: A comparison between Islamic
and conventional banks
Lucia Dalla Pellegrina
Article information:
To cite this document:
Lucia Dalla Pellegrina, (2012),"Does capitalization enhance efficient risk undertaking?", Accounting
Research J ournal, Vol. 25 Iss 3 pp. 185 - 207
Permanent link to this document:
http://dx.doi.org/10.1108/10309611211290167
Downloaded on: 24 January 2016, At: 21:16 (PT)
References: this document contains references to 56 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 1362 times since 2012*
Users who downloaded this article also downloaded:
Ismail Cebeci, (2012),"Integrating the social maslaha into Islamic finance", Accounting Research J ournal,
Vol. 25 Iss 3 pp. 166-184 http://dx.doi.org/10.1108/10309611211290158
Rihab Grassa, (2012),"Islamic banks' income structure and risk: evidence from GCC countries", Accounting
Research J ournal, Vol. 25 Iss 3 pp. 227-241 http://dx.doi.org/10.1108/10309611211290185
Nafis Alam, (2013),"Impact of banking regulation on risk and efficiency in Islamic banking", J ournal of
Financial Reporting and Accounting, Vol. 11 Iss 1 pp. 29-50 http://dx.doi.org/10.1108/J FRA-03-2013-0010
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Does capitalization enhance
ef?cient risk undertaking?
A comparison between Islamic and
conventional banks
Lucia Dalla Pellegrina
Department of Economics, University of Milano-Bicocca, Milan, Italy and
Paolo Baf? Center, Bocconi University, Milan, Italy
Abstract
Purpose – In light of the current debate on bank capital requirements, the purpose of this paper is to
investigate the relative impact of capitalization on risk-taking ef?ciency in Islamic and conventional
banks. The author tests whether changes occurring to the capital structure of such different types of
intermediaries unevenly affect their behaviour in terms of risk-taking ef?ciency.
Design/methodology/approach – The paper conducts an empirical analysis using data for the
period 2001-2011 by means of both standard regression methods and stochastic cost frontier
techniques.
Findings – Results provide evidence that more capitalized Islamic banks are associated to less risky
positions in terms of their asset structure. In particular, the latter exhibit higher liquidity standards
and a lower incidence of non-performing loans compared to other banks. This has delayed positive
effects on pro?tability and no substantial impact on ef?ciency. On the other hand, highly capitalized
conventional banks tend to shift from more traditional lending activities to investment in other (pro?t
generating) assets. Such strategy increases pro?tability and ef?ciency, although raising impaired
loans.
Research limitations/implications – This study does not address potential endogeneity concerns
that might affect the variables at stake, hence mainly providing indications in terms of correlation
between phenomena rather than causality.
Practical implications – The analysis has important practical implications when considering
capital adequacy as a regulatory tool for managing the risk of Islamic banks’ activity, following
principles similar to those recommended by the Basel committee.
Originality/value – The original contribution of the paper to the literature consists of comparing the
effects of capitalization in different types of banks, and results can be usefully exploited by
policymakers wishing to tailor banking regulation on the speci?c model of banking they are entitled to
regulate.
Keywords Islam, Banks, Capitalization, Islamic banks, Capital structure, Capital adequacy,
Stochastic ef?ciency frontier
Paper type Research paper
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1030-9616.htm
The author thanks two anonymous referees, the Editor M. Kabir Hassan, Paolo Colla,
Eliana La Ferrara, David Llewellyn, Matteo Manera, Donato Masciandaro, Margherita Saraceno,
participants at the European Financial Management Association Conference on Risk
Management in Financial Institutions, April 23-25, 2009, Audencia School of Management,
Nantes, and participants at the Conference on Islamic Finance and the Financial Crisis,
November 18-19, 2009, FEEM, Milano, for useful comments. The author is also grateful to the
Monash University and the Paolo Baf? Center, Bocconi University. Fabio Dragoni has provided
excellent research assistance. The usual disclaimer applies.
Capitalization
185
Accounting Research Journal
Vol. 25 No. 3, 2012
pp. 185-207
qEmerald Group Publishing Limited
1030-9616
DOI 10.1108/10309611211290167
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
1. Introduction
As a response to the devastating consequences of the recent ?nancial crisis, considerable
effort has been devoted to set down remedies aimed at preventing possible reiteration.
Besides massive government intervention and bank bailouts, a deep revision of banks’
capital structure has been warmly recommended. Such a proposal, in particular,
represents a reaction to excessive leverage which led part of the ?nancial system to
collapse. In this vein, for instance, the current debate on the new Basel III principles
seems to go towards establishing more severe standards for bank capital requirements.
However, a key element in the discussion on capitalization regulatory constraints is
that they may limit banks’ pro?tability. Hence, the focus of the debate shifts (back) to
whether the capital structure affects bank performance and, related to the focus of this
paper, to what extent it differently affects intermediaries of various nature.
Unquestionably, ever since the pioneering work of Modigliani and Miller (1958), the
issue of whether solid capitalization has some in?uence on – not necessarily ?nancial –
?rms’ performance has generated one of most vivid debates in the theory of corporate
governance. Traditional theory on agency costs asserts that, under certain conditions,
a higher amount of equity relative to debt pushes managers towards more ef?cient
behaviour. The rationale for this outcome is explained by a reduction of con?icting
interests between managers, shareholders, and other investors ( Jensen and Meckling,
1976; Myers, 1977).
However, the traditional approach to the optimal capital structure of ?rms is not
univocally assessed. Jensen (1986) himself, for example, observes that less capitalized
institutions shouldbe more inclined to improve their performance because debt ?nancing
reduces hazardous behaviour of managers by lowering free cash ?ow. Moreover, Jensen
observes that managers’ moral hazard is likely to be particularly severe when their
rewards are dependent upon asset dimension – which is not infrequent in banking –
since ?rms’ activity could be expanded beyond its maximum level of ef?ciency.
As for the banking sector, the analysis of the effects of capital standards becomes
extremely important, especially because of their predictive role on default risk[1].
Empirical studies, however, provide ambiguous results in terms of the effects of
capitalization on the actual ef?cient risk-taking position of intermediaries and,
eventually, on their pro?tability (see for example Dahl and Shrieves, 1990; Editz et al.,
1998; Short, 1994, for a survey). When turning to the comparison between different
models of banking, any prediction can become even more fading.
In this paper we explore the role of capitalization on risk-taking behaviour of two
different types of banks: Islamic and conventional. Islamic banks are in fact
characterized by substantial differences compared to conventional ones, especially as
far as – most part of – deposit agreements are concerned. A key element for the
purposes of this paper is that, as opposite to conventional banks’ deposits, Islamic ones
are based on pro?t and loss sharing principles. Hence, changes in the capital structure
of Islamic banks may not end up stimulating effects which are identical to those
expected in the conventional system (i.e. lower risk exposure, although this needs to
be veri?ed in our context). Next, since a revision of the capital structure may induce
different investment strategies in the two types of banks, we also check whether and
how in either banking system these strategies re?ect on pro?tability and ef?ciency.
In order to investigate risk we mainly concentrate on banks’ balance sheet structure,
while considering the return on assets and risk-adjusted pro?t ef?ciency[2] to verify
ARJ
25,3
186
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
the effects on their performance. Estimates of pro?t ef?ciency, in particular, are
obtained through stochastic frontier techniques (Aigner et al., 1977; Battese and Coelli,
1995; Jondrow et al., 1982; Schmidt and Sickles, 1984).
We carry out the empirical analysis on commercial Islamic and conventional banks
in the period 2001-2011 in countries where both types of bank operate. In this way we
can assume that the main feature characterizing the differences in bank behaviour is
the type of bank, and not country-speci?c characteristics, like the economic, social, and
legal environment.
The paper innovates upon similar previous contributions to the literature by
providing a thorough evaluation of risk undertaking bank attitudes associated with
given capital standards in different models of banking, conditional on the institutional
set up. Contrary to previous studies on this topic we do not con?ne our attention on the
mere comparison of ef?ciency across different types of banks, but rather investigate
how different funding strategies (i.e. capital relative to other ?nancial resources) are
associated to different levels of risk undertaking, and how these elements combine in
either an ef?cient or an inef?cient way.
There are, in fact, several studies comparing ef?ciency scores computed either for
Islamic banks set up in different countries or for groups of Islamic and conventional
banks belonging to the same country (Al-Fayoumi and Alkour, 2008; Moktar et al.,
2006; Kader et al., 2007; Al-Jarrah and Molyneux, 2003; Hassan, 2005; Hasan, 2004;
Kamaruddin et al., 2008; Niazi et al., 2010; Qayyum, 2007; Su?an, 2007; Yudistira, 2004).
However, these works often disregard or put little attention to the contribution of
capital standards in explaining the behaviour of each individual model of banking,
especially with regard on bank ef?cient risk undertaking.
We ?nd that the strategies of Islamic banks in association with a higher degree of
capitalization are substantially different compared to those of conventional banks. Our
estimates suggest that, on the one hand, higher capitalized Islamic banks are actually
associated to less risky policies in terms of their assets structure. In particular, they
exhibit higher liquidity standards compared to other banks. We also ?nd that this
reduces non-performing loans, with overall weak delayed positive effects on
pro?tability and no substantial impact on ef?ciency.
On the other hand, more capitalized conventional banks tend to shift from more
traditional lending activities to investment in other assets, probably yielding higher
returns but also raising impaired loans. This is in line with Awdeh et al. (2011), Kahane
(1977), Kim and Santomero (1988), Kohen and Santomero (1980) and Stiglitz (2001),
among others. Overall, ef?ciency and pro?tability positively respond to such policy.
The implications of these results are important from the banking sector’s regulatory
perspective, since proposals welcoming the recapitalization of ?nancial institutions
often tend to disregard their nature and operational context. This analysis shows that
different capital standards are instead associated to different risk-taking strategies,
depending on the speci?c model of banking, suggesting somehow more tailored policy
guidelines (see, for example, Ainley et al., 2007; Hesse et al., 2008; Sundararajan and
Errico, 2002 for a discussion).
The paper is organized as follows: in the next section we discuss the main implications
of Islamic banking principles on bank capital structure and risk undertaking; in Section 3
we describe the dataset; and in Section 4 we illustrate the econometric techniques used for
the estimation. Results are presented in Section 5. Section 6 concludes.
Capitalization
187
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
2. Background discussion
Risk-sharing principles[3] – on which Musharakah and Musharakah credit agreements
are based – and the prohibition of investing in some given activities[4] – such
as gambling, pornography, gun trading, etc. – constitute the twin pillars of Islamic
?nance. Both these peculiarities could imply some different attitudes of Islamic banks in
terms of risk undertaking compared to conventional ?nancial intermediaries.
As far as risk-sharing principles are concerned, the Musharakah agreement is
invoked in the equity structure of banks and is similar to the modern concepts of
partnership and joint stock ownership. On the other hand, the bank manages deposits
to generate pro?ts subject to the rules of Mudarabah. In this perspective, the true
difference with respect to conventional banking consists in the fact that most of Islamic
banks’ depositors[5] are entitled to earn a share of pro?ts on a pre-speci?ed basis – and
also incur in losses – thus being rewarded with a variable return which depends on the
return on the pool of assets in which their funds are invested by the bank. However,
they are normally not allowed to participate to decisions concerning funds allocation.
In the reminder of this section we provide an overview of the main elements which
may affect the attitude towards risk in the two models of banking, with an eye on
speci?c corporate governance tools that different categories of stakeholders can exploit
in order to protect their own interests.
An important feature that may affect the way the capital structure in?uences
ef?cient risk undertaking is that Islamic banks’ ?nanciers are subject to a double
agency problem since, besides holding imperfect information on banking activities
(which, however, is also common to stockholders), they also suffer from imperfect
information on bank reported returns on investments. In either type of banks –
conventional and Islamic – such kind of complication normally pertains to
shareholders only, whereas in Islamic banks it also involves depositors.
It is worth noting, however, that depositors and stockholders are not to be
considered identical in any perspective. From our standpoint, the most relevant
features distinguishing the two are the rights protecting each category of investor and
their attitude towards risk.
First, as for rights protecting each category, only shareholders normally take part to
the decisions concerning Islamic bank management. In other words, they have both
voice and exit rights, while depositors are only endowed with the possibility to exit.
Second, as for their attitude towards risk, depositors participate to a lower share of
banks’ pro?ts and losses, thus normally bearing less risk compared to shareholders.
This, in some sense, reveals their higher risk aversion. These two features, if combined,
suggest that changing the relative share of equity to deposits may not be a neutral
policy from some perspectives because of stakeholder’s incomplete interest alignment.
Our concern is whether risk undertaking can be one of these perspectives.
Given the discussion above, one important element driving possible differences
between Islamic and conventional banks in terms of the speci?c effects of
capitalization on risk attitude is related to monitoring. In fact, since in Islamic banks
depositors share in the risk of bank choices, they could be pushed towards exerting
tighter oversight over bank management compared to what occurs in traditional banks
(C
?
iha´k and Hesse, 2010). This should in principle result in an overall intensi?ed
monitoring activity, also sustained by higher interest alignment between stockholders
and depositors ( Jensen and Meckling, 1976). This may eventually limit managers’
ARJ
25,3
188
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
risk-taking exposure. On the other hand, however, the intensity of shareholders’
oversight on managerial actions could be mitigated by the fact that the former have
more incentives to delegate it to other concerned agents (i.e. depositors).
Another element driving possible differences between Islamic and conventional
banks in terms of the speci?c effects of capitalization on risk attitude is the capability
of rewarding ?xed-return investors. In fact, following Jensen (1986), if the revision of
the capital structure involves a relative increase of debt over equity, then in a
conventional bank free cash ?ow must reduce, and risk with it. However, compared to
conventional banks, less capitalized Islamic banks should have less incentives to limit
risk undertaking by reducing free cash ?ow since they are allowed to shift part of the
risk on depositors (Hesse et al., 2008; Iqbal and Llewellyn, 2002; Sundararajan and
Errico, 2002). For this reason, in Islamic banks lower capitalization may not decrease
risk taking to the same extent as it does in conventional banks.
Overall, the arguments discussed above suggest that risk exposure in Islamic banks
may increase or decrease as a consequence of stakeholders’ weight in the pool of bank
?nanciers. In principle, the effects of a revision of the relative share of investor types on
risk-taking attitudes are not clear-cut when comparing Islamic banks to their
conventional counterparts.
Finally, with regard to the second pillar of Islamic ?nance (the prohibition of
investing in some given activities) predictions seem somehow less puzzling, especially
when considering the relationship between capitalization and bank risk-taking
positions, on the one hand, and pro?tability and ef?ciency, on the other hand.
In general, regardless of whether higher shares of capital increase or decrease the
level of risk, in Islamic banks the consequences on pro?tability may end up being
smoothed due to the fact that they are not allowed to exploit the full range of
risk-return possibilities available for conventional banks. In particular, if we admit –
as it has been largely recognized – that most forbidden activities also re?ect a higher
content of both risk and expected return, in Islamic banks the possibilities of pursuing
the highest risk-return strategies are limited compared to what their conventional
counterparts can achieve. Hence, a revision of capital standards may, under some
circumstances, correspond to a smoothed effect on pro?tability.
In the next sections we empirically investigate whether capitalization is effective in
affecting risk undertaking in either model of banking, and then whether this affects
pro?t ef?ciency to some extent.
3. Data
Our sample of conventional and Islamic banks[6] has been extracted from the database
Bankscope for the period 2001-2011[7]. We include all commercial banks established in
countries where both conventional and Islamic intermediaries operate[8]. In particular,
we do not consider investment banks. Furthermore, banks that failed to supply
information regarding the variables used in our empirical analysis have been excluded
fromthe sample. This process eliminates a considerable number of intermediaries which
were listed in the original database but did not make available any records apart from
their generalities[9]. Overall, the panel is composedby350 banks – 289 conventional and
61 Islamic – for a total of 2,764 observations.
The panel is unbalanced. This choice has been driven by considerations of different
nature. The ?rst concerns the importance of taking speci?c account of the contribution
Capitalization
189
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
of banks that, for very different reasons, either left business or were absorbed by other
ones during the period of analysis.
Second, the length of the time span, in addition to the dynamics that have been
characterizing the bank ownership structure during the last years would imply a
substantial reduction of the number of observations. In fact, due to mergers and
acquisitions, only a restricted set of intermediaries at the end of the time interval are
operating under their initial structure.
Third, it is worth observing that the number of Islamic banks has considerably
grown in recent years. Consequently, the content of the information concerning the
behaviour of the Islamic banking sector would be scarce in case we admitted only
companies that were already operating in 2001.
Aggregate descriptive statistics on the capital structure and other related measures
are reported in Tables I and II, for Islamic and conventional banks, respectively. Some
differences between the two types of intermediaries emerge from this preliminary
inspection.
On average, deposits are 70 and 57 per cent of total liabilities plus own capital
(i.e. equal to total assets), in the period 2001-2011 in conventional and Islamic banks,
respectively. Equity, on its turn, is 15 per cent of total assets in Islamic banks and
16.6 per cent in conventional banks. Residual liabilities are composed by bonds and
other sources of debt.
On the left side of their balance sheet Islamic and conventional banks report similar
shares of gross loans[10] over total assets (49 and 52 per cent, respectively). The share of
impaired loans to gross loans is slightly in favour of conventional banks (8 per cent
against 9 per cent of Islamic banks). Other earning activities also exhibit equal shares
Variable Label Obs Mean SD Min. Max.
Equity/total assets eq_ta 434 15.00 19.62 0 100
Liquid assets/total assets liqass_ta 434 25.22 18.98 0.02 97.70
Fixed-assets/total assets ?xass_ta 421 3.96 10.87 0 88.98
Gross loans/total assets gloans_ta 414 48.68 25.65 0 99.28
Other earning assets/total assets otherea_ta 433 35.61 24.45 0.01 98.90
Deposits/total assets dep_ta 364 56.84 28.84 0 100
Off-balance sheet assets/total assets offbal_ta 352 20.97 28.08 0 212.29
Liquid assets/short-term funding liq_shtmf 362 92.48 168.05 1.19 1,000
Loan loss provisions/gross loans llpr_gloans 333 3.24 14.60 281.64 100
Non-performing loans/gross loans imploans_gloans 228 8.76 15.41 0 83.20
Net income/total assets ROA 434 1.44 6.20 244.35 24.73
Total assets
a
ta 434 3.71 7.38 0.02 58.88
Index of ownership dispersion
c
bvd 434 5.93 3.67 1 11
Net interest revenues/income netirev_income 434 3.71 14.89 241.50 238.51
No. employees/total assets staffnr_ta 434 0.05 0.08 0.01 60
Staff expenditure/No. employees
b
staffexp 434 57.06 64.15 3.71 391.97
Interest rate
d
int 434 0.06 0.21 0 1
Deprecation rate
d
depr 434 0.14 0.24 0 1
Notes:
a
Billion US dollars;
b
thousand US dollars;
c
computed by Bankscope; the index ranges from 1 to
11; higher values indicate higher ownership dispersion; all variables are in percentage except
superscript “a”, “b”, “c”, “d”
Table I.
Descriptive statistics:
Islamic banks
ARJ
25,3
190
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
of total assets in both types of banks (36 per cent). The incidence of ?xed-assets in terms
of total assets in Islamic banks is instead notably higher (4 per cent against less than
2 per cent of conventional banks). The situation is reversed as far as liquidity is
concerned (25 per cent of total assets in Islamic banks against 29 per cent in conventional
banks). Residual assets are constituted by other non-earning activities, which, by
difference, result higher in Islamic banks. Finally, off-balance sheet items are higher in
conventional banks – 29 per cent of total assets – than in Islamic ones (21 per cent)[11].
Another accounting ratio which may provide an idea of a bank’s risk-taking
position is the share of loan loss provisions to gross loans (1 per cent in conventional
banks against 3 per cent in their Islamic counterparts). Finally, liquidity represents
92 per cent of short-term funding in Islamic banks, whereas conventional banks seem
exhibiting a higher degree of risk in these terms, since the ratio is 52 per cent.
Overall, at the aggregate level, conventional and Islamic banks do not show
substantial differences in terms of their balance sheet structure and attitude towards
the risk. On the one hand, Islamic banks retain more ?xed-assets than conventional
intermediaries, thus apparently showing a higher degree of risk. On the other hand,
however, they retain more liquidity in relation to the amount of short-term deposits,
which in some sense counterbalances the two facts above.
Turning to the variables that are used in the empirical analysis to explain risk
attitude measures, we ?rst observe a substantial difference in the dimension of the two
types of intermediaries. Islamic banks record on average 3.71 billion of total assets,
compared to a larger dimension of conventional banks (7.41 billion on average).
Another important difference emerge in terms staff endowment, since the ratio of
employees over total assets is in favour of conventional banks (0.11 per cent against
Variable Label Obs Mean SD Min. Max.
Equity/total assets eq_ta 2,330 16.59 15.66 0 100
Liquid assets/total assets liqass_ta 2,330 29.10 17.97 0.07 98.88
Fixed-assets/total assets ?xass_ta 2,328 1.69 1.95 0 39.35
Gross loans/total assets gloans_ta 2,325 52.23 18.65 0.78 100
Other earning assets/total assets otherea_ta 2,330 36.44 17.84 0.62 100
Deposits/total assets dep_ta 2,311 69.63 16.26 2.91 100
Off-balance sheet assets/total assets offbal_ta 2,085 28.93 28.54 0 314.26
Liquid assets/short-term funding liq_shtmf 2,309 51.75 78.74 0.19 1,000
Loan loss provisions/gross loans llpr_gloans 2,254 1.14 5.02 2100 100
Non-performing loans/gross loans imploans_gloans 1,984 8.41 10.64 0 100
Net income/total assets roa 2,330 1.25 3.01 2100 29.71
Total assets
a
ta 2,330 7.41 14.17 0.02 136.38
Index of ownership dispersion
c
bvd 2,330 4.76 3.54 1 11
Net interest revenues/income netirev_income 2,330 4.91 86.06 2552.03 3,997.09
No. employees/total assets
d
staffnr_ta 2,330 0.11 0.13 0.01 1.29
Staff expenditure/No. employees
b
staffexp 2,330 30.02 65.40 0.43 1,979.25
Interest rate
d
int 2,330 0.07 0.07 0 1
Deprecation rate
d
depr 2,330 0.08 0.14 0 1
Notes:
a
Billion US dollars;
b
thousand US dollars;
c
computed by Bankscope; the index ranges from
1 to 11; higher values indicate higher ownership dispersion; all variables are in percentage except
superscript “a”, “b”, “c”, “d”
Table II.
Descriptive statistics:
conventional banks
Capitalization
191
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
0.05 per cent of Islamic banks), although this ratio does not account for staff members’
quality and skills. Expenditure per employee, which may under some conditions re?ect
the latter aspect, is instead in favour of Islamic banks (57,000 US$ on a yearly basis,
against 30,000 US$ of conventional banks).
Variables referring to other input costs different from personnel expenses are also
summarized in Tables I and II. In particular, ?xed-assets depreciation (computed as
non-interest expenses over ?xed-assets) is 8 per cent in conventional banks, whereas it
is 14 per cent in Islamic banks. Interest rate[12] on deposits is, respectively, 6 and
7 per cent in Islamic and conventional banks.
Finally, in order to verify whether risk may re?ect on bank pro?tability, we account
for the return on assets (ROA) and ef?ciency scores computed trough pro?t ef?ciency
frontier techniques. As for the ROA, descriptive statistics exhibit records of 1.44 and
1.25 in Islamic and conventional banks, respectively.
In the next section we describe the estimated equations and the empirical techniques
used to compute ef?ciency scores.
4. Empirical analysis
4.1 Analysis of risk undertaking positions and measures of pro?tability
We estimate the following regression for each type of bank:
y
it
¼ a
0
þa
1
eq_ta
it
þa
2
X
it
þm
i
þm
t
þ1
it
ð1Þ
where i identi?es the bank, and t refers to the year. y
it
is either one measure of risk
undertaking discussed in the previous section (share of liquidity, loans, other earning
assets, off-balance sheet items, deposits to total assets, loan loss provisions and
non-performing loans over gross loans, and liquidity over short-term funding) or a
measure of pro?tability (the return on assets and its one-year lag).
In particular, a
1
is the estimated parameter of interest which refers to the
effect of the capital ratio (eq_ta) on risk and pro?tability of bank i at time t. X
it
is a vector of covariates that consist in bank size (total assets), staff endowment
relative to the dimension of the bank, a measure of the incidence of traditional lending
activity on the overall activities of the bank (measured by interest revenues over
normalized total income), and an ownership dispersion index computed by
Bankscope. We also include in X
it
the unitary costs of inputs (employees’ annual
salary, ?xed-asset depreciation rate, and interest rates paid on deposits) in order to
control for the convenience of undertaking certain investments or speci?c ?nancing
policies.
In addition, a
0
is a constant term; m
i
are bank ?xed effects[13], while m
t
are year
dummies used to capture the economic cycle and other time features. We omit the
dummy relative to 2005 in order to avoid collinearity[14]. 1
it
is an idiosyncratic error,
such that Eð1
it
jeq_ta
it
; X
it
; m
i
; m
t
Þ ¼ 0. Robust standard errors are used.
Equation (1) is estimated for both the sample of conventional as well as for Islamic
banks through random-effects general least squares, after testing that the error term is
not correlated with the regressors[15].
4.2 Pro?t ef?ciency
Pro?t (in)ef?ciency[16] is interpreted as the diversion between the income achievable
by a ?rm given the faced unitary cost of inputs and its actual observed income.
ARJ
25,3
192
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
It can be measured in several ways. Some diffused techniques, such as data
envelopment models and the free disposable hull analysis, are non-parametric.
Non-parametric methods, however, embody a strong assumption, consisting of the fact
that all deviations from the maximum income are due to inef?ciency. For this reason,
several studies in the banking literature (Maudos et al., 2002; Berger and Mester, 1997;
Berger and Humphrey, 1997; Berger and Bonaccorsi di Patti, 2006) have been conducted
using a stochastic frontier analysis (SFA), which allows for the presence of both a
one-sided inef?ciency component and a random error term. We follow this tradition
estimating the (indirect) relationship betweenbankcapital structure andbankinef?ciency
with SFA.
The most complicated feature to be managed when using stochastic frontier
techniques is to separate the actual inef?ciency component from other purely random
factors affecting bank pro?tability. The availability of panel data, in our case, allows
avoiding any strong hypotheses on the distribution of the inef?ciency component and
its independence from other factors determining bank behaviour, which is instead the
case of cross-section analyses[17].
We assume that the deterministic component of the pro?t ef?ciency frontier for
bank i in year t, depends upon input prices and output quantities, and has a translog
form[18]:
lnðP
ijt
þuÞ ¼ b
0
þ
m
X
a
m
ln y
mijt
þ
n
X
b
n
ln w
nijt
þ
1
2
m
X
h
X
a
mh
ln y
mijt
ln y
hijt
þ
1
2
n
X
k
X
b
nk
ln w
nijt
ln w
kijt
þ
n
X
m
X
g
nm
ln w
nijt
ln y
mijt
þm
i
þ u
ijt
þ v
ijt
ð2Þ
where P
ijt
is each bank’s observed pro?t represented by earnings on variable outputs
minus variable costs expended to produce these outputs, u is a scale-constant added to
every bank’s pro?t so that the natural log is taken of a positive number, w
n
are unitary
input prices, y
m
are output quantities, and m
i
are ?rm-speci?c effects made available by
the panel-structure of the dataset.
As for the inef?ciency component, u
ijt
is a one-sided pro?t inef?ciency term typical
of each bank i in country j at time t, while v
ijt
is a two-sided normal random noise term
with zero mean and a standard deviation s
v
. Following Battese and Coelli (1995)[19] we
assume that the u
ijt
are time-variant, constructed as the interaction between bank
inef?ciency and time.
Furthermore, w
ijt
is made of three input costs (included in X in equation (1)): the
unitary cost of capital to the bank (interest paid on deposits), the unitary cost of
?xed-assets (depreciation) and the unitary cost of labour (average annual salary).
Following the outstanding literature on bank ef?ciency measurement (Berger and
Humphrey, 1997) we include three outputs in vector y
ijt
: net loans, other earning assets,
and off-balance sheet transactions. Non-performing loans are subtracted from loans, in
such a way that a risk-adjusted frontier is obtained from the data.
After imposing the usual linear restrictions characterizing the translog pro?t
function[20], we performed our estimates following the one-step maximum likelihood
introduced by Battese and Coelli (1995)[21].
Capitalization
193
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Such technique also provides an equation in which ef?ciency terms are speci?ed as
a function of a set of explanatory variables. The estimated equation is the following:
j
ijt
¼a
0
þa
1
eq_ta
ijt
*isl
i
þa
2
eq_ta
ijt
*conv
i
þa
3
isl
i
þa
3
X
ijt
þm
i
þm
j
þm
t
þ1
ijt
ð3Þ
where j
ijt
are the ef?ciency components correspondent to the estimated u
ijt
, while m
i
, m
j
and m
t
are, respectively, bank, country, and time speci?c effects. a
0
is a constant term
and 1
ijt
is a random error.
Ratios between equity and total assets are separately accounted for Islamic and
conventional banks as the interaction between the equity-to-assets variable and
individual dummy variables for each type of banking (eq_ta
ijt*
isl
i
and eq_ta
ijt*
conv
i
). As
standard practice when interacting terms, one of the dummies (in this case isl
i
) is also
accounted for as a separate regressor. This is particularly important in order to
distinguish the contribution to ef?ciency provided by an increase of capital ratios in each
individual model of bankingfromother elements of inef?ciencywhichare typical of either
Islamic or conventional intermediaries but are not speci?cally related to capitalization.
Finally, X
ij
is a vector of control variables (logarithm of total assets, number of
employees over total assets, net interest revenues over normalized total income, and the
Bankscope ownership dispersion index) equal to the one included in equation (1), with
the exception of input costs.
5. Results
5.1 Regression output
The output from the estimation of equation (1) is reported in Table III for Islamic banks
and in Table IV for conventional banks. In each speci?cation we consider the same set
of controls while changing the dependent variables.
Results show that differently capitalized Islamic banks adopt different strategies
compared to conventional banks. More in detail, as reported in Table III, a higher capital
ratio corresponds to lower lending activity in Islamic banks (the parameter associated to
eq_ta in the gross loans-to-total assets is negative and strongly signi?cant). This
translates into holding more liquid assets than less capitalized Islamic intermediaries
(the parameter associated to eq_ta in the liquidity-to-total assets equation shows
opposite sign and almost the same magnitude as the one associated to loans).
It is then not surprising that the shift from loans to liquid assets that is observed in
Islamic banks also raises liquidity over short-term funding. Most important,
non-performing loans appear reduced and this is likely to slightly increase returns
afterwards, since the parameter associated to the one-period lagged return on assets
turns positive and, although weakly, signi?cant. In a nutshell, more capitalized Islamic
banks seem to engage in safer policies compared to less capitalized in terms of their
assets structure and this increases pro?tability with some delay.
As reported in Table IV, more capitalized conventional banks also refrain from
engaging in more traditional lending activities (as it occurs in Islamic banks, the
parameter of eq_ta in the regression of gross loans-to-total assets is also signi?cant
although considerably lower than in the case of Islamic banks) at the advantage of
investment in other pro?t-generating activities (the parameter associated to eq_ta turns
positive and signi?cant in the third column of Table IV). Disinvesting in traditional
lending also leads conventional banks to increase liquidity, although to lesser extent
than in their Islamic counterparts (the parameter associated to eq_ta turns signi?cant in
ARJ
25,3
194
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
l
i
q
a
s
s
_
t
a
g
l
o
a
n
s
_
t
a
o
t
h
e
r
e
a
_
t
a
o
f
f
b
a
l
_
t
a
d
e
p
_
t
a
l
i
q
_
s
h
t
m
f
l
l
p
r
_
g
l
o
a
n
s
i
m
p
l
o
a
n
s
_
g
l
o
a
n
s
a
r
o
a
r
o
a
_
1
e
q
_
t
a
0
.
2
3
3
5
9
*
*
*
2
0
.
2
7
1
9
7
*
*
*
0
.
0
4
9
7
4
2
0
.
0
1
1
6
6
2
0
.
1
4
6
4
1
2
.
6
4
2
4
8
*
*
2
0
.
0
3
7
2
3
2
0
.
0
9
7
0
3
*
*
0
.
0
1
8
7
2
0
.
0
4
7
5
1
*
(
0
.
0
7
1
8
0
)
(
0
.
0
7
8
8
3
)
(
0
.
0
8
9
8
8
)
(
0
.
0
3
6
6
7
)
(
0
.
1
0
9
3
1
)
(
1
.
0
9
8
4
6
)
(
0
.
0
3
9
1
5
)
(
0
.
0
4
4
1
9
)
(
0
.
0
2
5
3
0
)
(
0
.
0
2
8
4
6
)
t
a
0
.
1
3
4
4
3
0
.
4
1
9
8
2
*
2
0
.
3
3
5
0
3
0
.
5
2
8
0
9
*
0
.
0
4
6
2
2
2
1
.
4
9
6
2
0
2
0
.
1
5
6
2
7
0
.
0
8
3
9
2
0
.
0
5
2
4
7
*
2
0
.
0
4
5
5
6
(
0
.
1
2
0
3
6
)
(
0
.
2
1
4
5
5
)
(
0
.
2
3
5
7
9
)
(
0
.
2
7
1
3
0
)
(
0
.
2
2
4
2
6
)
(
1
.
1
6
9
6
1
)
(
0
.
1
2
0
8
9
)
(
0
.
1
0
0
2
6
)
(
0
.
0
3
0
8
3
)
(
0
.
0
7
3
9
2
)
b
v
d
0
.
1
9
3
0
1
2
2
.
4
3
6
6
6
*
*
*
0
.
8
1
2
2
2
0
.
5
8
0
4
3
2
0
.
9
0
3
4
4
1
.
3
5
4
9
1
0
.
5
3
8
9
5
2
0
.
7
0
2
7
2
*
0
.
1
6
4
0
6
*
*
0
.
3
7
5
4
4
*
*
(
0
.
6
3
3
2
5
)
(
0
.
8
3
3
4
6
)
(
0
.
7
2
3
9
2
)
(
1
.
5
2
3
6
2
)
(
0
.
9
4
7
6
0
)
(
3
.
0
5
5
8
7
)
(
0
.
4
7
1
0
5
)
(
0
.
4
0
1
6
1
)
(
0
.
0
8
0
1
3
)
(
0
.
1
5
4
6
3
)
n
e
t
i
r
e
v
_
i
n
c
o
m
e
2
0
.
0
0
3
1
4
0
.
0
5
6
2
7
2
0
.
0
1
7
8
0
2
0
.
1
5
4
4
1
*
*
*
0
.
0
3
0
2
6
*
*
0
.
5
0
5
2
0
2
0
.
0
4
2
8
1
2
0
.
0
2
3
3
7
0
.
0
1
0
0
8
0
.
0
3
2
6
6
(
0
.
0
2
9
3
5
)
(
0
.
0
4
0
9
0
)
(
0
.
0
5
4
1
7
)
(
0
.
0
3
0
2
1
)
(
0
.
0
1
5
0
8
)
(
0
.
6
8
5
4
1
)
(
0
.
0
2
9
2
1
)
(
0
.
0
2
4
7
5
)
(
0
.
0
1
1
0
6
)
(
0
.
0
3
6
4
9
)
s
t
a
f
f
n
r
_
t
a
1
,
4
0
9
.
7
6
2
2
,
0
3
6
.
0
0
*
2
2
,
7
5
3
.
6
1
*
*
6
8
1
.
0
1
5
1
,
5
2
7
.
7
7
1
1
,
2
1
1
.
8
1
1
,
2
3
5
.
8
5
5
8
5
8
,
9
1
2
.
0
7
*
*
*
2
8
7
5
.
2
0
2
*
2
4
,
4
6
2
.
0
6
*
*
*
(
1
,
7
1
2
.
4
5
3
4
0
)
(
1
,
1
4
2
.
8
5
6
5
4
)
(
1
,
4
0
1
.
2
6
2
8
5
)
(
2
,
6
1
3
.
9
8
0
8
2
)
(
1
,
7
5
8
.
5
7
7
8
4
)
(
1
3
,
7
0
0
.
1
9
4
1
5
)
(
1
,
2
7
5
.
2
4
7
4
3
)
(
3
,
2
7
9
.
0
8
6
9
3
)
(
4
6
7
.
2
8
5
4
6
)
(
8
1
7
.
4
8
4
6
7
)
s
t
a
f
f
e
x
p
2
0
.
0
3
3
4
7
2
0
.
0
4
2
9
3
0
.
0
7
9
7
3
*
*
2
0
.
0
8
2
8
2
*
2
0
.
0
6
5
2
9
0
.
3
7
8
2
3
0
.
0
1
4
7
9
0
.
0
1
0
2
6
2
0
.
0
0
2
6
6
2
0
.
0
0
5
2
7
(
0
.
0
2
2
0
8
)
(
0
.
0
3
5
8
7
)
(
0
.
0
3
8
1
6
)
(
0
.
0
4
2
9
1
)
(
0
.
0
7
2
7
2
)
(
0
.
3
8
6
2
1
)
(
0
.
0
1
8
7
8
)
(
0
.
0
1
7
6
8
)
(
0
.
0
0
3
2
0
)
(
0
.
0
0
9
1
8
)
d
e
p
r
2
1
8
.
5
8
2
1
1
.
8
6
1
.
6
8
9
2
1
.
8
5
8
1
0
.
9
8
*
9
0
.
5
3
3
.
7
3
2
8
.
9
4
3
2
3
.
8
5
2
2
2
.
8
2
1
(
1
7
.
7
6
)
(
1
2
.
4
2
)
(
1
8
.
4
5
)
(
3
.
2
4
8
)
(
5
.
8
6
4
)
(
7
7
.
7
2
)
(
2
.
8
0
5
)
(
9
.
8
9
3
)
(
2
.
9
8
6
)
(
4
.
7
9
3
)
i
n
t
2
5
.
7
0
2
2
1
5
.
8
0
1
3
.
8
0
2
2
.
9
0
2
2
4
5
.
9
3
*
*
*
2
4
6
.
8
*
*
*
1
.
3
5
7
9
.
5
3
6
.
2
3
4
*
*
*
4
.
7
6
0
*
(
5
.
0
0
3
)
(
1
4
.
7
0
)
(
1
2
.
7
9
1
)
(
7
.
8
2
5
)
(
1
6
.
8
3
)
(
1
1
5
.
9
)
(
4
.
5
6
1
)
(
7
.
0
9
6
)
(
2
.
7
9
2
)
(
2
.
9
8
1
)
2
0
0
1
2
4
.
6
5
0
8
1
0
.
7
9
7
7
5
3
.
2
3
1
3
1
1
.
3
1
4
6
7
2
1
.
5
0
6
8
4
1
9
.
5
3
6
0
2
8
.
9
2
2
4
1
2
2
.
2
4
0
9
5
*
(
3
.
8
6
3
7
1
)
(
5
.
2
6
7
9
5
)
(
6
.
4
3
8
4
2
)
(
3
.
7
5
5
9
7
)
(
4
.
6
0
6
4
7
)
(
4
4
.
5
5
9
9
0
)
(
9
.
1
3
6
9
9
)
(
1
.
2
9
6
3
6
)
2
0
0
2
2
2
.
5
7
3
7
5
2
.
5
1
4
1
5
4
.
4
8
0
0
2
5
.
7
1
5
5
7
2
3
.
4
5
8
3
8
2
5
.
9
9
0
3
9
2
0
.
3
3
3
7
7
2
2
.
2
9
6
1
1
*
*
2
3
.
4
7
6
0
5
(
3
.
5
5
7
3
0
)
(
3
.
7
4
4
4
7
)
(
4
.
9
7
8
1
9
)
(
5
.
7
3
5
2
8
)
(
6
.
5
6
0
4
2
)
(
5
1
.
2
5
5
7
9
)
(
1
.
3
2
7
1
7
)
(
1
.
1
3
1
5
3
)
(
3
.
0
5
1
8
5
)
2
0
0
3
2
2
.
9
9
4
0
5
4
.
3
2
4
1
6
0
.
5
7
4
2
7
7
.
3
2
3
8
3
2
.
0
4
9
6
6
7
.
5
8
1
8
3
2
0
.
5
5
4
6
0
2
2
.
3
9
3
2
7
2
2
.
1
5
7
6
7
*
*
*
2
3
.
8
9
4
3
2
(
2
.
4
9
0
8
6
)
(
2
.
8
8
5
6
7
)
(
4
.
4
2
7
9
7
)
(
7
.
9
2
2
5
2
)
(
5
.
4
7
6
4
3
)
(
3
4
.
2
0
8
8
7
)
(
1
.
9
1
7
0
4
)
(
2
.
3
8
5
3
1
)
(
0
.
7
3
0
4
5
)
(
2
.
9
4
1
4
0
)
2
0
0
4
2
4
.
2
0
5
7
1
*
*
2
.
5
2
3
0
7
2
0
.
7
7
9
6
9
2
.
9
1
3
4
3
3
.
3
3
1
7
6
3
9
.
2
9
5
5
8
0
.
5
7
2
7
5
2
1
.
7
2
5
9
3
2
1
.
7
1
8
0
7
*
*
2
3
.
2
6
4
2
5
(
1
.
9
4
6
8
0
)
(
2
.
3
8
0
5
9
)
(
3
.
3
7
2
6
4
)
(
2
.
2
1
6
8
7
)
(
3
.
4
5
7
9
2
)
(
4
4
.
5
2
6
2
9
)
(
1
.
8
4
2
8
2
)
(
1
.
8
8
8
3
5
)
(
0
.
7
1
5
1
7
)
(
2
.
7
8
2
0
9
)
(
c
o
n
t
i
n
u
e
d
)
Table III.
Effect of capitalization on
risk and pro?tability:
Islamic banks
Capitalization
195
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
l
i
q
a
s
s
_
t
a
g
l
o
a
n
s
_
t
a
o
t
h
e
r
e
a
_
t
a
o
f
f
b
a
l
_
t
a
d
e
p
_
t
a
l
i
q
_
s
h
t
m
f
l
l
p
r
_
g
l
o
a
n
s
i
m
p
l
o
a
n
s
_
g
l
o
a
n
s
a
r
o
a
r
o
a
_
1
2
0
0
6
0
.
8
0
3
5
9
2
3
.
6
2
2
3
1
6
.
2
8
4
8
1
*
*
1
.
4
9
2
1
7
2
5
.
6
0
6
7
8
2
0
.
6
7
8
7
2
1
.
9
7
5
3
7
2
0
.
2
0
4
7
2
2
2
.
2
1
2
7
5
*
*
2
5
.
3
7
2
9
6
*
*
(
2
.
3
1
3
7
4
)
(
2
.
7
8
3
3
7
)
(
3
.
1
0
7
8
3
)
(
2
.
3
0
2
3
2
)
(
3
.
5
0
4
9
7
)
(
3
3
.
9
1
1
3
6
)
(
2
.
4
0
1
3
1
)
(
1
.
2
6
7
2
5
)
(
0
.
9
5
1
5
6
)
(
2
.
4
5
2
9
7
)
2
0
0
7
2
4
.
4
4
7
7
9
*
2
1
.
4
8
2
9
4
1
.
7
2
5
5
8
2
.
7
5
0
6
5
2
0
.
9
8
0
1
2
2
2
5
.
6
4
4
5
0
0
.
3
6
3
9
8
2
0
.
6
5
9
8
1
2
1
.
0
5
3
8
0
2
0
.
8
8
1
4
6
(
2
.
3
4
5
8
6
)
(
2
.
8
6
2
8
5
)
(
2
.
9
5
9
7
8
)
(
3
.
3
1
9
4
4
)
(
3
.
3
7
7
4
1
)
(
1
9
.
3
1
1
2
5
)
(
0
.
8
1
5
5
7
)
(
1
.
1
1
7
9
7
)
(
0
.
7
1
0
7
4
)
(
2
.
4
0
2
2
8
)
2
0
0
8
2
6
.
6
5
9
3
8
*
*
0
.
2
2
5
5
6
2
0
.
7
9
9
0
2
2
.
1
3
0
4
0
2
1
.
7
3
9
0
0
2
3
1
.
9
0
9
7
1
*
3
.
3
8
6
6
6
1
.
3
5
6
7
8
2
2
.
6
3
4
5
0
*
*
2
3
.
2
7
7
8
9
(
2
.
7
2
1
8
1
)
(
3
.
2
3
9
5
6
)
(
3
.
1
1
0
8
1
)
(
4
.
7
6
9
2
4
)
(
3
.
3
7
8
0
3
)
(
1
7
.
9
0
5
3
9
)
(
2
.
2
9
4
2
7
)
(
3
.
0
1
9
1
2
)
(
1
.
1
9
7
9
3
)
(
2
.
3
6
5
7
9
)
2
0
0
9
2
6
.
9
6
0
3
7
*
*
2
1
.
0
8
6
3
1
0
.
0
4
0
7
4
2
0
.
8
4
1
5
0
2
1
.
0
4
5
3
2
2
3
.
0
2
8
7
9
6
.
5
8
3
6
9
*
*
2
0
.
4
9
3
5
2
2
5
.
2
9
1
8
2
*
*
*
2
6
.
8
7
7
8
8
*
*
*
(
2
.
9
0
9
9
4
)
(
3
.
6
8
5
9
9
)
(
3
.
3
7
9
9
1
)
(
4
.
7
0
5
2
8
)
(
3
.
5
9
6
8
2
)
(
2
8
.
5
8
5
0
3
)
(
3
.
3
0
8
4
1
)
(
3
.
4
9
0
2
8
)
(
1
.
7
8
1
7
6
)
(
2
.
3
4
3
7
1
)
2
0
1
0
2
8
.
9
4
8
1
2
*
*
*
2
2
.
8
8
9
3
2
1
.
0
7
3
5
5
2
0
.
7
4
9
5
6
2
0
.
0
7
0
6
8
2
1
3
.
5
4
7
0
5
5
.
4
2
7
4
4
*
*
0
.
7
9
3
7
2
2
6
.
5
9
1
3
9
*
*
*
2
7
.
8
9
3
1
3
*
*
*
(
3
.
2
3
0
0
2
)
(
4
.
1
5
7
8
8
)
(
3
.
8
4
3
4
7
)
(
5
.
3
6
8
0
5
)
(
3
.
4
8
5
0
6
)
(
2
7
.
7
4
4
4
0
)
(
2
.
6
9
0
1
9
)
(
3
.
5
4
2
6
4
)
(
1
.
9
4
0
5
7
)
(
2
.
3
6
3
4
8
)
2
0
1
1
2
9
.
9
1
4
6
0
*
*
*
2
4
.
7
7
1
9
6
1
.
8
1
6
9
5
2
8
.
9
5
7
2
2
*
1
.
8
1
0
5
0
2
2
3
.
8
0
4
0
0
0
.
2
3
7
9
8
0
.
4
5
0
6
3
2
4
.
3
3
1
5
3
*
*
*
2
6
.
2
6
6
4
4
*
*
(
3
.
5
0
6
0
4
)
(
4
.
0
9
8
9
1
)
(
4
.
2
9
7
3
0
)
(
5
.
2
4
1
4
1
)
(
3
.
4
5
4
7
8
)
(
2
3
.
5
1
7
4
8
)
(
2
.
7
1
8
1
4
)
(
3
.
0
1
2
1
3
)
(
1
.
2
9
4
6
5
)
(
2
.
4
8
5
2
2
)
C
o
n
s
t
a
n
t
2
6
.
9
7
2
3
9
*
*
*
6
9
.
7
0
5
0
6
*
*
*
2
6
.
3
3
4
2
7
*
*
*
2
2
.
6
6
6
0
4
*
*
*
6
4
.
9
8
6
0
6
*
*
*
2
7
.
4
7
0
2
5
2
2
.
4
3
3
7
2
5
.
8
9
1
2
6
3
.
6
8
1
2
4
*
*
*
6
.
8
3
1
5
7
*
*
*
(
4
.
4
1
2
6
7
)
(
6
.
0
9
0
3
8
)
(
5
.
7
7
4
6
0
)
(
8
.
0
2
4
3
4
)
(
7
.
5
3
4
4
5
)
(
2
8
.
8
2
5
2
1
)
(
2
.
4
3
5
6
8
)
(
4
.
6
6
0
9
2
)
(
1
.
0
6
8
6
9
)
(
2
.
3
3
3
7
3
)
B
a
n
k
?
x
e
d
e
f
f
e
c
t
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
O
b
s
e
r
v
a
t
i
o
n
s
4
3
4
4
1
4
4
3
3
3
5
2
3
6
4
3
6
2
3
3
3
2
9
6
4
3
4
3
7
7
R
2
0
.
1
4
1
0
0
.
2
3
2
5
0
.
2
6
0
5
0
.
0
8
2
8
0
.
2
3
4
9
0
.
3
4
1
3
0
.
1
7
3
9
0
.
3
5
8
3
0
.
1
2
8
2
0
.
1
6
9
3
N
o
t
e
s
:
S
i
g
n
i
?
c
a
n
t
a
t
:
*
p
,
0
.
1
,
*
*
p
,
0
.
0
5
,
a
n
d
*
*
*
p
,
0
.
0
1
;
O
L
S
e
s
t
i
m
a
t
e
s
,
b
a
n
k
-
s
p
e
c
i
?
c
a
f
f
e
c
t
s
a
r
e
t
r
e
a
t
e
d
a
s
r
a
n
d
o
m
;
r
o
b
u
s
t
s
t
a
n
d
a
r
d
e
r
r
o
r
s
i
n
p
a
r
e
n
t
h
e
s
e
s
;
a
g
r
o
s
s
l
o
a
n
s
r
e
f
e
r
t
o
t
w
o
y
e
a
r
s
p
r
e
v
i
o
u
s
t
o
i
m
p
a
i
r
e
d
l
o
a
n
s
Table III.
ARJ
25,3
196
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
l
i
q
a
s
s
_
t
a
g
l
o
a
n
s
_
t
a
o
t
h
e
r
e
a
_
t
a
o
f
f
b
a
l
_
t
a
d
e
p
_
t
a
l
i
q
_
s
h
t
m
f
l
l
p
r
_
g
l
o
a
n
s
i
m
p
l
o
a
n
s
_
g
l
o
a
n
s
a
r
o
a
r
o
a
_
1
e
q
_
t
a
0
.
1
1
5
3
2
*
*
*
2
0
.
0
8
3
3
5
*
*
0
.
0
8
1
6
7
*
*
0
.
0
0
2
0
9
2
0
.
0
7
0
4
8
*
*
0
.
1
1
2
7
5
2
0
.
0
5
3
3
0
*
*
0
.
0
7
1
5
6
*
*
0
.
0
1
9
3
4
*
*
*
0
.
0
0
7
4
7
*
(
0
.
0
4
1
3
7
)
(
0
.
0
3
7
4
8
)
(
0
.
0
3
6
5
9
)
(
0
.
0
4
5
9
2
)
(
0
.
0
3
1
8
3
)
(
0
.
3
2
0
7
0
)
(
0
.
0
2
3
6
9
)
(
0
.
0
3
5
4
5
)
(
0
.
0
0
6
5
8
)
(
0
.
0
0
4
0
4
)
t
a
2
0
.
0
9
7
7
3
*
*
0
.
0
2
7
9
7
2
0
.
0
3
2
9
2
2
0
.
0
0
0
3
5
2
0
.
0
8
1
2
1
*
*
2
0
.
1
2
7
7
2
2
0
.
0
1
0
5
8
*
2
0
.
0
5
0
8
3
*
*
*
0
.
0
1
0
2
4
*
*
*
0
.
0
0
7
0
9
(
0
.
0
4
8
4
0
)
(
0
.
0
4
5
9
7
)
(
0
.
0
4
7
0
3
)
(
0
.
0
7
9
0
2
)
(
0
.
0
3
6
3
2
)
(
0
.
1
1
4
8
6
)
(
0
.
0
0
5
4
7
)
(
0
.
0
1
6
2
4
)
(
0
.
0
0
3
6
8
)
(
0
.
0
0
5
5
2
)
b
v
d
2
1
.
0
3
5
2
6
*
*
*
0
.
9
3
7
2
5
*
*
*
2
0
.
4
9
2
1
9
*
0
.
4
5
0
8
4
0
.
7
5
3
5
8
*
*
*
2
3
.
2
2
9
4
5
*
*
*
2
0
.
0
4
8
2
3
2
0
.
0
9
3
7
6
0
.
0
8
1
6
0
*
*
*
0
.
0
7
5
6
2
*
*
*
(
0
.
2
2
1
5
7
)
(
0
.
2
8
1
7
7
)
(
0
.
2
5
3
6
9
)
(
0
.
3
3
1
4
7
)
(
0
.
2
3
4
9
7
)
(
1
.
0
4
7
9
6
)
(
0
.
0
3
6
8
1
)
(
0
.
1
5
1
5
2
)
(
0
.
0
2
2
8
4
)
(
0
.
0
2
7
4
0
)
n
e
t
i
r
e
v
_
i
n
c
o
m
e
0
.
0
0
1
3
0
2
0
.
0
0
0
4
3
0
.
0
0
0
2
0
2
0
.
0
0
3
7
2
0
.
0
0
0
9
7
*
*
0
.
0
0
2
0
1
0
.
0
0
0
1
8
0
.
0
0
0
7
4
2
0
.
0
0
0
2
4
2
0
.
0
0
0
2
3
(
0
.
0
0
0
9
3
)
(
0
.
0
0
1
0
2
)
(
0
.
0
0
0
9
9
)
(
0
.
0
0
2
3
9
)
(
0
.
0
0
0
4
1
)
(
0
.
0
0
1
9
2
)
(
0
.
0
0
0
4
3
)
(
0
.
0
0
1
3
9
)
(
0
.
0
0
0
1
8
)
(
0
.
0
0
0
5
0
)
s
t
a
f
f
n
r
_
t
a
5
5
4
.
4
0
2
1
5
2
1
,
2
6
6
.
3
5
5
8
0
*
*
2
1
6
7
.
2
9
5
7
8
2
3
,
1
0
7
.
4
6
4
9
7
*
*
*
6
4
9
.
8
7
1
3
5
2
1
,
1
9
8
.
1
2
8
7
2
0
.
7
9
4
1
1
5
9
3
.
2
0
8
2
0
2
1
9
0
.
7
1
5
*
*
*
2
2
7
4
.
5
7
1
*
*
*
(
6
2
6
.
3
0
7
2
2
)
(
6
3
7
.
6
0
0
2
0
)
(
5
8
7
.
7
5
8
0
9
)
(
5
4
8
.
7
5
5
5
4
)
(
5
5
8
.
2
7
5
6
7
)
(
2
,
2
8
0
.
9
6
9
0
3
)
(
7
1
.
2
9
2
7
9
)
(
4
0
3
.
6
3
6
2
3
)
(
7
0
.
1
7
9
0
7
)
(
6
8
.
5
4
2
5
5
)
s
t
a
f
f
e
x
p
0
.
0
0
5
3
4
2
0
.
0
0
9
4
2
*
*
0
.
0
0
8
0
7
*
*
2
0
.
0
0
0
4
6
2
0
.
0
0
6
4
7
*
0
.
0
4
8
3
1
2
0
.
0
0
2
7
9
2
0
.
0
0
0
2
8
0
.
0
0
2
1
0
*
0
.
0
0
1
6
9
*
(
0
.
0
0
3
7
4
)
(
0
.
0
0
4
4
6
)
(
0
.
0
0
3
9
8
)
(
0
.
0
0
5
8
0
)
(
0
.
0
0
3
3
9
)
(
0
.
0
4
8
6
0
)
(
0
.
0
0
2
5
5
)
(
0
.
0
0
0
9
8
)
(
0
.
0
0
1
1
2
)
(
0
.
0
0
0
8
7
)
d
e
p
r
2
2
.
8
5
4
3
.
8
2
2
6
.
8
6
3
*
2
0
.
0
5
6
2
2
0
.
0
8
5
4
2
2
4
.
2
3
1
.
5
5
8
3
.
7
8
3
*
*
*
0
.
6
4
2
0
.
2
6
8
(
4
.
6
7
9
)
(
3
.
6
8
2
)
(
3
.
5
0
7
)
(
2
.
6
6
8
)
(
2
1
.
6
4
2
)
(
4
5
.
8
1
)
(
2
.
7
8
9
)
(
1
.
0
5
6
)
(
0
.
7
9
8
)
(
0
.
3
7
6
)
i
n
t
9
.
6
7
8
2
4
.
4
8
9
1
.
8
4
2
1
5
.
0
7
2
4
5
.
9
0
*
*
*
5
6
2
.
1
*
*
*
0
.
2
7
5
2
2
.
6
7
9
5
.
4
8
6
*
6
.
9
5
4
*
*
*
(
9
.
5
9
2
)
(
8
.
0
7
4
)
(
7
.
8
9
3
)
(
1
7
.
2
4
)
(
1
1
.
7
7
)
(
1
6
7
.
8
)
(
6
.
8
2
1
)
(
4
.
7
3
0
)
(
3
.
3
5
1
)
(
2
.
7
8
3
)
2
0
0
1
6
.
1
3
6
7
8
*
*
*
2
3
.
5
5
0
0
4
*
*
*
6
.
4
4
1
2
9
*
*
*
2
1
.
4
0
4
9
8
2
0
.
4
5
9
4
3
6
.
5
1
1
7
2
1
.
0
0
8
8
2
*
2
0
.
7
1
2
0
9
*
*
(
1
.
3
0
8
0
7
)
(
1
.
3
1
0
5
8
)
(
1
.
2
4
3
9
5
)
(
1
.
8
8
5
1
6
)
(
0
.
8
9
8
2
5
)
(
3
.
9
6
6
6
1
)
(
0
.
5
7
3
9
0
)
(
0
.
3
0
8
0
2
)
2
0
0
2
6
.
3
1
3
9
4
*
*
*
2
3
.
5
0
3
5
2
*
*
*
7
.
1
7
7
3
7
*
*
*
2
2
.
2
3
0
8
3
0
.
2
0
6
8
2
1
1
.
7
3
0
1
2
*
2
0
.
1
6
5
1
3
2
0
.
3
2
9
0
6
2
0
.
7
4
0
9
1
*
*
*
(
1
.
2
5
2
8
4
)
(
1
.
1
5
5
2
3
)
(
1
.
0
8
0
6
9
)
(
1
.
7
9
2
2
0
)
(
0
.
8
4
2
3
4
)
(
7
.
0
5
3
1
3
)
(
0
.
7
8
5
6
0
)
(
0
.
2
1
3
6
4
)
(
0
.
2
0
2
7
4
)
2
0
0
3
3
.
7
3
7
3
4
*
*
*
2
1
.
3
9
4
2
8
3
.
7
3
8
3
4
*
*
*
2
1
.
5
4
0
8
7
0
.
1
2
8
7
3
1
1
.
9
5
4
7
9
*
0
.
0
1
3
8
1
3
.
1
7
7
6
8
*
*
*
2
0
.
0
0
8
6
4
2
0
.
3
6
8
0
9
*
(
0
.
9
0
8
1
5
)
(
0
.
8
8
8
5
6
)
(
0
.
8
7
8
0
1
)
(
1
.
5
4
4
7
8
)
(
0
.
6
9
2
0
0
)
(
7
.
2
2
2
7
2
)
(
0
.
4
8
7
9
1
)
(
0
.
7
0
8
3
6
)
(
0
.
1
7
8
6
8
)
(
0
.
1
9
9
5
9
)
2
0
0
4
2
.
6
9
2
9
7
*
*
*
2
1
.
6
3
3
5
4
*
*
*
2
.
2
1
0
7
6
*
*
*
2
0
.
1
7
4
6
7
2
0
.
5
4
9
9
8
6
.
0
3
4
8
7
2
0
.
5
1
5
2
9
1
.
5
0
2
6
3
*
0
.
0
4
2
1
9
2
0
.
3
6
0
8
7
*
(
0
.
6
6
8
0
4
)
(
0
.
5
9
9
4
4
)
(
0
.
6
9
8
0
1
)
(
1
.
2
4
4
6
9
)
(
0
.
6
1
3
4
0
)
(
5
.
9
2
9
1
5
)
(
0
.
6
1
6
3
0
)
(
0
.
8
2
6
8
3
)
(
0
.
2
3
9
0
2
)
(
0
.
1
9
6
7
2
)
2
0
0
6
1
.
2
1
4
8
0
*
*
0
.
0
9
4
0
1
0
.
0
5
8
1
7
2
0
.
6
9
2
4
8
2
0
.
4
9
1
1
8
2
0
.
8
7
2
4
4
0
.
0
4
5
9
6
2
0
.
2
7
0
8
7
2
0
.
1
1
0
4
3
2
0
.
2
7
7
5
2
(
0
.
5
6
8
2
2
)
(
0
.
5
4
9
1
9
)
(
0
.
6
2
1
3
9
)
(
2
.
3
2
9
5
4
)
(
0
.
4
8
9
1
1
)
(
3
.
8
4
8
6
0
)
(
0
.
5
2
2
4
7
)
(
0
.
5
6
8
4
9
)
(
0
.
1
6
6
9
8
)
(
0
.
1
9
3
4
5
)
(
c
o
n
t
i
n
u
e
d
)
Table IV.
Effect of capitalization on
risk and pro?tability:
conventional banks
Capitalization
197
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
l
i
q
a
s
s
_
t
a
g
l
o
a
n
s
_
t
a
o
t
h
e
r
e
a
_
t
a
o
f
f
b
a
l
_
t
a
d
e
p
_
t
a
l
i
q
_
s
h
t
m
f
l
l
p
r
_
g
l
o
a
n
s
i
m
p
l
o
a
n
s
_
g
l
o
a
n
s
a
r
o
a
r
o
a
_
1
2
0
0
7
0
.
0
3
0
9
8
0
.
6
2
0
2
7
2
2
.
2
8
5
3
7
*
*
*
1
.
8
5
0
0
3
2
0
.
5
7
1
5
6
2
4
.
2
4
2
8
2
0
.
5
9
3
1
9
2
1
.
2
4
2
0
1
*
2
0
.
2
2
9
8
1
2
0
.
3
6
2
2
2
*
(
0
.
8
5
1
6
7
)
(
0
.
6
8
0
7
5
)
(
0
.
7
8
7
3
0
)
(
2
.
3
4
2
9
3
)
(
0
.
5
6
4
9
8
)
(
3
.
0
8
4
0
8
)
(
0
.
5
4
5
4
8
)
(
0
.
6
5
3
2
4
)
(
0
.
1
6
8
0
1
)
(
0
.
1
9
1
1
8
)
2
0
0
8
2
5
.
5
1
8
0
2
*
*
*
4
.
6
9
7
7
7
*
*
*
2
5
.
4
1
8
3
2
*
*
*
1
.
0
9
8
6
2
0
.
7
1
2
2
9
2
1
5
.
3
4
4
2
7
*
*
*
0
.
7
5
3
1
7
2
0
.
6
9
4
4
0
2
0
.
8
4
5
3
8
*
2
1
.
0
6
2
6
2
*
*
*
(
1
.
0
0
3
1
3
)
(
0
.
8
7
6
6
3
)
(
0
.
9
0
6
2
9
)
(
2
.
3
1
7
7
3
)
(
0
.
5
9
9
8
9
)
(
4
.
7
6
1
4
1
)
(
0
.
6
9
9
0
7
)
(
0
.
7
9
6
2
3
)
(
0
.
4
6
6
2
0
)
(
0
.
1
9
1
4
3
)
2
0
0
9
2
4
.
0
4
3
5
4
*
*
*
2
.
3
3
5
8
4
*
*
2
3
.
0
2
0
7
8
*
*
*
2
3
.
7
7
3
8
8
*
1
.
8
8
4
7
8
*
*
*
2
1
0
.
4
6
1
7
8
*
*
0
.
5
3
4
9
6
2
0
.
6
9
9
7
3
2
0
.
5
0
3
9
9
*
*
*
2
0
.
8
7
8
8
4
*
*
*
(
1
.
0
5
4
9
9
)
(
0
.
9
9
1
1
8
)
(
1
.
0
2
4
0
4
)
(
1
.
9
7
6
7
6
)
(
0
.
6
9
5
8
0
)
(
4
.
9
2
3
9
8
)
(
0
.
5
4
5
5
7
)
(
0
.
7
3
5
8
0
)
(
0
.
1
8
5
2
1
)
(
0
.
1
9
2
2
1
)
2
0
1
0
2
3
.
7
7
1
6
4
*
*
*
3
.
4
8
8
7
6
*
*
*
2
5
.
0
7
4
0
2
*
*
*
2
5
.
2
8
8
5
8
*
*
1
.
6
4
1
1
9
*
*
2
1
0
.
0
2
6
5
7
0
.
0
6
7
2
5
2
1
.
1
6
0
8
8
2
0
.
3
3
8
4
5
*
2
0
.
7
2
8
5
7
*
*
*
(
1
.
0
6
1
9
7
)
(
1
.
0
4
8
7
3
)
(
1
.
1
2
8
8
2
)
(
2
.
2
8
9
0
3
)
(
0
.
7
7
6
0
3
)
(
6
.
2
1
5
8
4
)
(
0
.
5
0
8
6
1
)
(
0
.
7
3
2
5
6
)
(
0
.
1
8
5
2
9
)
(
0
.
1
9
2
5
0
)
2
0
1
1
2
3
.
8
9
4
8
4
*
*
*
3
.
9
7
2
6
3
*
*
*
2
5
.
7
7
2
0
9
*
*
*
2
6
.
8
8
4
6
9
*
*
*
1
.
0
8
2
8
7
2
1
0
.
7
9
6
3
7
*
*
2
0
.
0
8
9
4
0
2
1
.
0
1
2
8
8
2
0
.
4
4
9
4
8
*
*
2
0
.
9
2
2
6
6
*
*
*
(
1
.
0
3
5
6
0
)
(
1
.
0
2
1
2
0
)
(
1
.
1
6
7
5
3
)
(
2
.
0
7
2
2
7
)
(
0
.
8
6
0
3
5
)
(
4
.
5
1
1
6
4
)
(
0
.
4
9
1
7
3
)
(
0
.
7
6
0
4
7
)
(
0
.
1
8
4
8
1
)
(
0
.
2
1
7
6
3
)
C
o
n
s
t
a
n
t
3
2
.
3
0
6
0
0
*
*
*
4
9
.
4
1
1
6
6
*
*
*
3
8
.
3
2
9
3
9
*
*
*
3
1
.
3
5
2
1
5
*
*
*
6
6
.
1
1
6
0
0
*
*
*
7
1
.
8
8
3
9
0
*
*
*
2
.
2
4
4
7
0
*
*
*
7
.
5
2
3
1
7
*
*
*
0
.
9
2
8
0
2
*
*
*
1
.
9
1
3
4
7
*
*
*
(
1
.
9
8
2
4
6
)
(
2
.
1
2
2
0
4
)
(
1
.
9
2
3
8
2
)
(
3
.
1
9
7
1
3
)
(
1
.
9
2
7
8
4
)
(
9
.
9
1
0
0
6
)
(
0
.
7
6
6
5
6
)
(
1
.
2
0
0
0
8
)
(
0
.
2
6
9
5
8
)
(
0
.
2
3
5
3
3
)
B
a
n
k
?
x
e
d
e
f
f
e
c
t
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
Y
e
s
O
b
s
e
r
v
a
t
i
o
n
s
2
,
3
3
0
2
,
3
2
5
2
,
3
3
0
2
,
0
8
5
2
,
3
1
1
2
,
3
0
9
2
,
2
5
4
1
,
7
4
2
2
,
3
3
0
2
,
0
2
9
R
2
0
.
1
5
4
4
0
.
1
4
3
8
0
.
1
6
8
6
0
.
0
5
4
0
0
.
1
4
5
1
0
.
1
0
9
9
0
.
2
3
0
6
0
.
1
1
7
5
0
.
1
2
6
5
0
.
1
1
4
8
N
o
t
e
s
:
S
i
g
n
i
?
c
a
n
t
a
t
:
*
p
,
0
.
1
,
*
*
p
,
0
.
0
5
,
a
n
d
*
*
*
p
,
0
.
0
1
;
O
L
S
e
s
t
i
m
a
t
e
s
,
b
a
n
k
-
s
p
e
c
i
?
c
a
f
f
e
c
t
s
a
r
e
t
r
e
a
t
e
d
a
s
r
a
n
d
o
m
;
r
o
b
u
s
t
s
t
a
n
d
a
r
d
e
r
r
o
r
s
i
n
p
a
r
e
n
t
h
e
s
e
s
;
a
g
r
o
s
s
l
o
a
n
s
r
e
f
e
r
t
o
t
w
o
y
e
a
r
s
p
r
e
v
i
o
u
s
t
o
i
m
p
a
i
r
e
d
l
o
a
n
s
Table IV.
ARJ
25,3
198
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
the ?rst column of Table III but it is considerably smaller that the correspondent one
reported in Table III). More capitalized conventional banks, however, accumulate less
reserves (the parameter associated to eq_ta inthe seventh columnof Table IVis negative
and signi?cant) although impaired loans tend to increase (eighth column, Table IV).
From to the negative parameter associated to deposits, we can also infer that
non-traditional assets of conventional banks are likely to be ?nanced through sources
different from customers’ deposits, plausibly bank securities.
Although perhaps intended for handling risk, the strategy of conventional banks as a
consequence of recapitalization seems rather unsafe, although pro?table, especially in the
very short-run. One possible rationale for this behaviour is that increasing the share of
equity over other funding sources calls for higher stock rewards and hence investing in
more pro?table activities. In fact, the return on asset positively responds to a
contemporaneous increase incapitalization, althoughturningalmost negligible afterwards.
Turning to other determinants of banks’ asset structure, we can observe that larger
Islamic banks (in terms of total assets) tend to operate more through traditional lending
(in terms of a higher share of gross loans-to-assets). Off-balance sheets items are also
higher in large Islamic banks. Interesting evidence is that a positive association
between dimension and pro?tability (roa) holds for Islamic intermediaries, a result
which seems in line, for example, with what Bashir (1999) ?nds on Sudanese Islamic
banks. Larger conventional banks seem instead to re?ect a different situation since
they appear more ef?cient and highly pro?table (lower shares of liquidity and reserve
provision, less non-performing loans, higher return on assets).
An important element that consistently affects returns in both types of banks is
ownership dispersion, which does not seem a questionable issue according to the
literature. Overstaffed banks, regardless whether they are Islamic or conventional,
show signi?cantly lower returns on their assets, which is perhaps due to inef?ciencies.
Also, banks granting better staff compensation seem to be associated to more
intensi?ed non-traditional activities (high other earning assets), which may involve
more tailored products for their customers and therefore require higher skills. This
situation seems somehow similar in both types of banks.
The economic cycle also has a clear impact on pro?tability in both Islamic and
conventional banking systems since year dummies in both the ninth and tenth columns
become dramatically negative since the beginning of the crisis. Such a reduction of
pro?tability seems to have involved Islamic banks more severely than conventional
ones. Reasonably for the same facts, Islamic banks have also increased loan loss
provisions in 2009-2010.
Finally, the effects of the crisis seem having brought back conventional banks to
operate more on traditional activities consisting in collecting deposits and lending
(the parameters associated to the share of loans-to-assets is positive since 2008), while
this phenomenon does not seem to have affected the balance sheet structure of Islamic
banks to a large extent.
5.2 Pro?t ef?ciency
Table V reports the stochastic translog pro?t frontier parameters for equation (2)
estimated through the maximum-likelihood model. Overall, parameters have the
expected sign where signi?cant. The only variable which does not signi?cantly affect
pro?ts is the amount of other earning assets (lnotherea). On the other hand, the cost
Capitalization
199
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Pro?t frontier
Dependent variable: log of income
lnloans 0.57739
* * *
(0.07185)
lnotherea 0.11420 (0.08352)
lnoffbal 0.10422
*
(0.05831)
lndepr 20.71826
* * *
(0.05574)
lnint 20.34760
* * *
(0.07756)
lnstaffexp 0.62933
* * *
(0.07659)
1/2lnloans
*
lnloans 0.06076
* * *
(0.01028)
1/2lnotherea
*
lnotherea 20.06053
* * *
(0.01269)
1/2lnoffbal
*
lnoffbal 20.01530
* * *
(0.00510)
lnloans
*
lnotherea 20.07864
* * *
(0.00933)
lnloans
*
lnoffbal 0.01412
* *
(0.00656)
lnotherea
*
lnoffbal 0.00614 (0.00707)
1/2lndepr
*
lndepr 20.07547
* * *
(0.00380)
1/2lnint
*
lnint 20.09690
* * *
(0.00633)
1/2lnstaffexp
*
lnstaffexp 0.01076
*
(0.00628)
lndepr
*
lnint 0.18314
* * *
(0.00504)
lndepr
*
lnstaffexp 20.03220
* * *
(0.00773)
lnint
*
lnstaffexp 0.01067 (0.01164)
lnloans
*
lndepr 0.05568
* * *
(0.00575)
lnloans
*
lnint 20.08719
* * *
(0.00804)
lnloans
*
lnstaffexp 0.03151
* * *
(0.00755)
lnotherea
*
lndepr 20.07008
* * *
(0.00688)
lnotherea
*
lnint 0.13979
* * *
(0.00939)
lnotherea
*
lnstaffexp 20.06971
* * *
(0.00850)
lnoffbal
*
lndepr 20.00333 (0.00487)
lnoffbal
*
lnint 0.00295 (0.00806)
lnoffbal
*
lnstaffexp 0.00038 (0.00614)
Constant 20.04240
* * *
(5.85727)
lnsigma2 2.18229
* * *
(0.07956)
gamma 0.26994 (0.58538)
mu 0.91347
* * *
(0.11557)
eta 20.00467 (0.01005)
LR test one side error 6,583.43
Log likelihood 2239.94689
Regression of correlates with pro?t ef?ciency scores
Dependent variable: pro?t ef?ciency
eq_ta
*
isl 20.00055 (0.00064)
eq_ta
*
conv 0.00122
* * *
(0.00033)
isl 20.00003 (0.00146)
log ta 0.02375
* * *
(0.00333)
bvd 0.00402
* *
(0.00183)
netirev_income 20.00003
*
(0.00001)
staffnr_ta
a
13.11555 (14.69708)
Constant 0.320692
* * *
(0.005477)
LR test one side error 6583.43
Log likelihood 2239.94689
Note: Signi?cant at:
*
p , 0.1,
* *
p , 0.05, and
* * *
p , 0.01; year, bank, and country ?xed effects
included;
a
parameter and standard error divided by 100
Table V.
Effect of capitalization
on pro?t ef?ciency
ARJ
25,3
200
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
of labour (lnstaffexp) is surprisingly positive in the pro?t function, although this
anomaly is not new to the empirical literature on Islamic bank ef?ciency (Srairi, 2010).
Estimates of equation (3) are also reported in Table V. In this case, parameters allow
comparing the impact of the capital structure on ef?ciency in the two banking systems,
as well as drawing some useful conclusions on the ef?cient use of inputs.
Turning to the estimation of equation (3), the most relevant outcome is the
signi?cance of parameter a
2
, which is associated to the interaction term between
equity over total assets and the dummy variable for conventional banks (eq_ta
*
conv).
Its positive sign indicates that, controlling for other elements of ef?ciency related to the
type of banking, increasing capitalization in conventional banks leads to higher pro?t
ef?ciency. According to the regression output discussed above, this possibly suggests
that equity, since requiring higher rewards than other sources of funds, may induce
more pressure on management to generate income.
Such a result does not hold, however, for Islamic banks. In fact, the parameter a
1
,
associated to the interaction term between equity over total assets and the dummy
variable for Islamic banks (eq_ta
*
isl), is not signi?cant. This can be an indication that
differences between depositors and shareholders provide less remarkable effects
compared to conventional banks.
Interesting evidence also concerns the dummy for Islamic banks taken alone.
Indeed, contrary to what is commonly argued, Islamic banks do not seem to be less
ef?cient than their conventional counterparts since the coef?cient associated to the
Islamic bank dummy variable (isl) is not signi?cant.
Important effects on ef?ciency are also provided by both banks’ dimension and
ownership dispersion. The coef?cients associated to the logarithm of total assets
(log ta) and to the ownership dispersion index (bvd) are indeed positive and highly
signi?cant, suggesting that larger banks having a structure closer to public companies
tend to be more ef?cient, which in line with the literature on bank ef?ciency (Berger and
Bonaccorsi di Patti, 2006).
Finally, producing a higher share of income through traditional lending activity
may reduce ef?ciency. The – although weakly – negative sign of the variable
netirev_income, in particular, seems mostly in line with the strategy of conventional
banks, namely shifting to less traditional activities in order to boost pro?tability. Such
a result, which does not look particularly surprising, is likely to be driven by practices
adopted in the pre-crisis period. The assets related to such activities, however, recently
ceased to yield abnormal returns and have been discarded in some cases.
6. Concluding remarks
Pro?t and loss sharing principles, by reformulating the allocation of risk between
shareholders and depositors, may affect the incentives of both depositors to monitor the
conduct of banking activity and managers to undertake risky business. On the one hand,
under Islamic banking principles a number of depositors supply investment accounts and
participate in the bank investment activities through risk-sharing schemes, thus,
in principle, requiring less protection than conventional bank depositors (Ariss and
Sarieddine, 2007). Onthe other hand, the possibilityof shiftingrisks ondepositors raises the
risk of moral hazard, so that all investors should look for a higher degree of safekeeping.
An appropriate capital structure is normally recommended in order to assure
the optimal level of protection to stakeholders from excessive risk undertaking.
Capitalization
201
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
From the perspective of Islamic banks, however, it is not clear whether several forces
acting in different directions are likely to shape the “ideal” recommendations in terms of
bank capital structure similar, for example, to those set up by the Basel Committee for
European banks.
Given this debate, in this paper we have focused on the impact of capitalization on
both Islamic and conventional banks’ risk-taking ef?ciency in the period 2001-2011 in
countries where both bank types are available.
Results, obtained by regressions investigating bank capital structure with attention
to risk undertaking exposure have shown that more capitalized Islamic banks are
likely to pursue safer policies than their conventional counterparts. Increasing liquid
assets is an example. This has the effect of reducing non-performing loans, with weak
bene?ts on returns.
On the other hand, more capitalized conventional banks tend to shift from
traditional lending activities to investment in other pro?t-generating assets. However,
we have observed that this policy raises impaired loans, although not offsetting the
previous effect. Thus, pro?tability boosts in correspondence to an increase in
capitalization, although becoming almost negligible afterwards.
We have also used stochastic frontier techniques to complete the analysis of
ef?cient risk undertaking. We have found that conventional banks show a positive
relationship between capitalization and pro?t ef?ciency, whereas there are no similar
effects in Islamic banks. This suggests that equity, compared to deposits, is likely to
represent a stinging blow to banks to perform, but the similarities between stocks and
deposits in the Islamic banking context may de?nitely smooth this effect.
Finally, the implications of this analysis are important from a regulatory
perspective. First, it provides an empirical evaluation of an issue which does not have
any theoretical univocal answer, since there are valid arguments to suggest either
increasing or decreasing capital standards in Islamic banks. Second, predictions are in
line with regulatory provisions set up in some European countries in relation to the
application of Basel principles to Islamic banks (or branches of Islamic banks)
operating in these countries. It is not a case that, for example, in the UK the FSA
declared that when capital is not an appropriate mitigating tool for Islamic banks, then
other ways of managing risk are to be set up. These include, for example,
recommendations in favour of sound corporate governance mechanisms and other
appropriate systems of control (Ainley et al., 2007).
Notes
1. During economic slumps, low capitalization standards and inef?cient risk undertaking also
appear to be a good proxy for bank failures (Beim and Calomiris, 2001; Estrella et al., 2000).
2. It has been observed that most of the contrasting opinions in the empirical literature may
stem from the use of various measures of performance, going from basic accounting ratios to
more sophisticated ones. On this point, it is increasingly becoming acknowledged that “it is
necessary to recognize explicitly the concept of ef?ciency as a chief measure of performance
in the empirical models linking bank capital to risk, in such a way that it becomes possible to
distinguish between ef?cient and inef?cient risk undertaking” (Hughes and Moon, 1995;
Kwan and Eisenbeis, 1997; Mathuva, 2009; Weill, 2002).
3. Although Islamic bank are increasingly recently shying away from strictly applying
Mudaraba and Musharaka in their lending activity in favor of more standardized products,
ARJ
25,3
202
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
these principles still characterize the structure of the liability side (see for example Hassan
and Lewis, 2007 for a discussion on the evolution of Islamic products).
4. See also Ariff (1988) for an overview of Islamic banking principles.
5. Islamic deposits can range from current accounts to saving and investment deposits, and
other special investment accounts. As in conventional banks, current accounts do not receive
any remuneration and are exempted from losses. In all other cases, deposits are typically
pooled by the bank under a Mudarabah agreement to provide funds to its customers. See
Lewis and Algaud (2001) for details on these and other types of contacts.
6. In this group we consider only pure commercial Islamic banks, i.e. those providing
Shariah-compatible products. Mixed banks with Islamic services are not included.
7. Although a parallel analysis has been conducted limited to the period previous to the recent
?nancial crisis (results, which are not substantially different from the ones reported in this
paper, are available upon request), here we also include years 2009-2011. In order to allow
this extension we accurately dealt with outliers encountered, in particular, in the last three
years of the overall time span.
8. We follow C
?
iha´k and Hesse (2010) including Bahrain, Bangladesh, Brunei, Egypt, Indonesia,
Iran, Jordan, Kuwait, Lebanon, Malaysia, Pakistan, PalestinianTerritory, Qatar, Saudi Arabia,
Sudan, Tunisia, United Arab Emirates, and Yemen. We are forced to exclude Gambia and
Mauritania since only one Islamic bank in each country provided data to Bankscope in the
time span considered. Like C
?
iha´k and Hesse we keep Iran and Sudan in the sample due to the
weight of their Islamic bank sector, although being conscious of the fact that proper
conventional bankingis not allowedinthese countries. We also include Turkey, Iraq andSyria
because of the relevance of both their Islamic and conventional banking systems, although
estimates which exclude these countries do not provide substantially different results
compared to those presented further in this paper. Weill (2011) uses a similar set of countries
(which includes Sudan and Iran) in a study performed on “all the countries where Islamic
banks and conventional banks coexisted during the period 2000-2007” (Weill, 2011).
9. Note that there is no form of selection bias in the sample, except from information
availability at the time of data collection.
10. Including non-performing loans.
11. Islamic banks have not traditionally held off-balance sheet exposures as is instead the case
conventional banks. Furthermore, the cost of holding off-balance sheet items may be higher
in Islamic banks since a number of these off-balance sheet items require more procedures
which in turn lead to higher costs.
12. Interests are implicitly computed for Islamic banks.
13. We also carried out parallel regressions where bank ?xed effects have been replaced by
country ?xed effects. Results are not substantially different from those reported in the next
section, apart from the higher ef?ciency of parameters associated to some control variables
when using country ?xed effects.
14. Year 2005 has been chosen as a residual category due to reasons related to the economic
cycle, so as to ease the interpretation of all other time dummies by comparison with the
upper-limit for economic expansion in our time span.
15. Hausman (1978) test supports the hypothesis of absence of correlation in all regressions
presented in the next section.
16. We use pro?t ef?ciency rather than cost ef?ciency because, as highlighted by the literature
(see Berger and Bonaccorsi di Patti, 2006 for a discussion), when dealing with agency
Capitalization
203
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
problems it is better suited to account for how well managers raise revenues as well as
control costs, so as to maximize corporate value.
17. Schmidt and Sickles (1984), among others, note that cross-sectional models require that the
inef?ciency component be independent from the regressors, although this hypothesis may
often be violated.
18. See for example Greene (1980) and Carvallo and Kasman (2005) for a discussion on the
advantages of using the translog speci?cation.
19. See also Aigner et al. (1977), Jondrow et al. (1982) and Lee and Schmidt (1993).
20. See Kumbhakar and Lovell (2000) for further details.
21. See also Coelli (1996).
References
Aigner, D.J., Lovell, C.A.K. and Schmidt, P. (1977), “Formulation and estimation of stochastic
frontier production function models”, Journal of Econometrics, Vol. 6 No. 1, pp. 21-37.
Ainley, M., Mashayekhi, A., Hicks, R., Rahman, A. and Ravalia, A. (2007), Islamic Finance in the
UK: Regulation and Challenges, Financial Services Authority, London.
Al-Fayoumi, N.A. and Alkour, E.M. (2008), “Cost and pro?t ef?ciency in the Jordan commercial
banks: parametric and non-parametric approaches for estimating ef?ciency”, Dirasat,
Vol. 35, pp. 21-38.
Al-Jarrah, I. and Molyneux, P. (2003), “Cost ef?ciency, scale elasticity and scale economies in
Arabian banking”, paper presented at 10th Annual Conference of Economic Research
Form for the Arab Countries, Iran and Turkey in Morocco, December 16-18.
Ariff, M. (1988), “Islamic banking”, Asian-Paci?c Economic Literature, Vol. 2 No. 2, pp. 48-64.
Ariss, R.T. and Sarieddine, Y. (2007), “Challenges in implementing capital adequacy guidelines to
Islamic banks”, Journal of Banking Regulation, Vol. 9 No. 1.
Awdeh, A., El-Moussawi, C. and Machrouh, F. (2011), “The effect of capital requirements on
banking risk”, International Research Journal of Finance and Economics, Vol. 66, pp. 133-46.
Bashir, A.H.M. (1999), “Risk and pro?tability measures in Islamic banks: the case of two
Sudanese banks”, Islamic Economic Studies, Vol. 6 No. 2, pp. 1-24.
Battese, G.E. and Coelli, T.J. (1995), “A model for technical inef?ciency effects in stochastic
frontier production function for panel data”, Empirical Economics, Vol. 20, pp. 325-32.
Beim, D.O. and Calomiris, C.W. (2001), Emerging Financial Markets, McGraw-Hill, NewYork, NY.
Berger, A. and Mester, L.J. (1997), “Inside the black box: what explains differences in the
ef?ciencies of ?nancial institutions?”, Journal of Banking and Finance, Vol. 21, pp. 895-947.
Berger, A.N. and Bonaccorsi di Patti, E. (2006), “Capital structure and ?rm performance: a new
approach to testing agency theory and an application to the banking industry”, Journal of
Banking and Finance, Vol. 30 No. 4, pp. 1065-102.
Berger, A.N. and Humphrey, D.B. (1997), “Ef?ciency of ?nancial institutions: international
survey and directions for future research”, Center for Financial Institutions
Working Papers 97-05, Wharton School Center for Financial Institutions, University of
Pennsylvania, Philadelphia, PA.
Carvallo, O. and Kasman, A. (2005), “Cost ef?ciency in the Latin American and Caribbean banking
systems”, International Financial Market Institutions and Money, Vol. 15, pp. 55-72.
C
?
iha´k, M. and Hesse, H. (2010), “Islamic banks and ?nancial stability: an empirical analysis”,
Journal of Financial Services Research, Vol. 38 No. 2, pp. 95-113.
ARJ
25,3
204
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Coelli, T. (1996), “A guide to frontier version 4.1: a computer program for stochastic frontier
production and cost function estimation”, CEPA Working Paper.
Dahl, D. and Shrieves, R.E. (1990), “The impact of regulation on bank equity infusions”, Journal
of Banking and Finance, Vol. 6, pp. 1209-28.
Editz, T., Michael, I. and Perraudin, W. (1998), “The impact of capital requirements on UK bank
behaviour”, Federal Reserve Bank of New York Policy Review, Vol. 4 No. 3, pp. 15-22.
Estrella, A., Park, S. and Peristiani, S. (2000), “Capital ratios as predictors of bank failure”,
Federal Reserve Bank of New York Economic Policy Review, Vol. 6 No. 2, pp. 33-52.
Greene, W.H. (1980), “On the estimation of a ?exible frontier production model”, Journal of
Econometrics, Vol. 13, pp. 101-15.
Hasan, Z. (2004), “Measuring the ef?ciency of Islamic banks: criteria, methods and social
priorities”, Islamic Economic Review, Vol. 8 No. 2, pp. 1-30.
Hassan, K.M. (2005), “The cost, pro?t and X-ef?ciency of Islamic banks”, Proceeding of the
Economic Research Forum 12th Annual Conference, Cairo, Egypt, working paper.
Hassan, K.M. and Lewis, M.K. (2007), Handbook of Islamic Banking, Edward Elgar, Cheltenham.
Hausman, J. (1978), “Speci?cation tests in econometrics”, Econometrica, Vol. 46, pp. 1251-72.
Hesse, H., Jobst, A. and Sole, J. (2008), “Trends and challenges in Islamic ?nance”, World
Economics, Vol. 9 No. 2, pp. 175-93.
Hughes, J. and Moon, C.G. (1995), “Measuring bank ef?ciency when managers trade return for
reduced risk”, working paper, Department of Economics, Rutgers University,
New Brunswick, NJ.
Iqbal, M. and Llewellyn, D.T. (2002), Islamic Banking and Finance: New Perspective on
Pro?t-Sharing and Risk, Edward Elgar, Cheltenham.
Jensen, M.C. (1986), “Agency costs of free cash ?ow, corporate ?nance and takeovers”, American
Economic Review, Vol. 76, pp. 323-9.
Jensen, M.C. and Meckling, W.H. (1976), “Theory of the ?rm: managerial behavior, agency costs
and ownership structure”, Journal of Financial Economics, Vol. 4, pp. 305-60.
Jondrow, J., Lovell, C.A.K., Materov, I.S. and Schmidt, P. (1982), “On estimation of technical
inef?ciency in the stochastic frontier production function model”, Journal of Econometrics,
Vol. 19 Nos 2/3, pp. 233-8.
Kader, J.M., Asaporta, A.J. and Al-Maghaireh, A. (2007), “Comparative ?nancial performance of
Islamic banks vis-a`-vis conventional banks in the UAE”, Proceeding of the Annual Student
Research Symposium and the Chancellor’s Undergraduate Research Award, available at:
http://sra.uaeu.ac.ae/CURA/Proceedings
Kahane, M.C. (1977), “Capital adequacy and the regulation of ?nancial intermediaries”, Journal of
Banking and Finance, Vol. 1, pp. 207-18.
Kamaruddin, B.H., Safa, M.S. and Mohd, R. (2008), “Assessing production ef?ciency of Islamic
banks and conventional bank Islamic windows in Malaysia”, International Journal of
Business and Management, Vol. 1 No. 1, pp. 31-48.
Kim, D. and Santomero, A.M. (1988), “Risk in banking and capital regulation”, Journal of
Finance, Vol. 43, pp. 1219-33.
Koehn, M. and Santomero, A.M. (1980), “Regulation of bank capital and portfolio risk”, Journal of
Finance, Vol. 35, pp. 1235-44.
Kumbhakar, S.C. and Lovell, C.A.K. (2000), Stochastic Frontier Analysis, Cambridge University
Press, Cambridge.
Capitalization
205
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Kwan, S. and Eisenbeis, R.A. (1997), An Analysis of Inef?ciencies in Banking: A Stochastic
Frontier Approach, Federal Reserve Bank of San Francisco, San Francisco, CA.
Lee, Y.H. and Schmidt, P. (1993), “A production frontier model with ?exible temporal variation in
technical inef?ciency”, in Fried, H.O., Lovell, C.A.K. and Schmidt, S.S. (Eds),
The Measurement of Productive Ef?ciency: Techniques and Applications, Oxford
University Press, New York, NY.
Lewis, M.K. and Algaud, L.M. (2001), Islamic Banking, Edward Elgar, Cheltenham.
Mathuva, D.M. (2009), “Capital adequacy, cost income ratio and the performance of commercial
banks: the Kenyan scenario”, International Journal of Applied Economics and Finance,
Vol. 3, pp. 35-47.
Maudos, J., Pastor, J.M., Perez, F. andQuesada, J. (2002), “Cost andpro?t ef?ciencyinEuropeanbanks”,
Journal of International Financial Markets, Institutions and Money, Vol. 12 No. 1, pp. 33-58.
Modigliani, F. and Miller, M.H. (1958), “The cost of capital, corporation ?nance, and the theory of
investment”, American Economic Review, Vol. 48, pp. 261-97.
Moktar, H.S., Abdullah, N. and Al-Habshi, S.M. (2006), “Ef?ciency of Islamic banks in Malaysia:
a stochastic frontier approach”, Journal of Economic Cooperation Among Islam Countries,
Vol. 27 No. 2, pp. 37-70.
Myers, S.C. (1977), “Determinants of corporate borrowing”, Journal of Financial Economics, Vol. 5,
pp. 147-75.
Niazi, G.S.K., Raoof, A., Rehman, R. and Shahid, H. (2010), “Ef?ciencies comparison of Islamic
and conventional banks of Pakistan”, International Research Journal of Finance and
Economics, Vol. 49.
Qayyum, A. (2007), Financial Sector Reforms and the Ef?ciency of Banking in Pakistan,
South Asian Network of Economic Research Institutes (SANEI), Dhaka, available at:
www.saneinetwork.net
Schmidt, P. and Sickles, R.C. (1984), “Production frontiers and panel data”, Journal of Business
and Economic Statistics, Vol. 2 No. 4, pp. 367-74.
Short, H. (1994), “Ownership, control, ?nancial structure, and the performance of ?rms”, Journal
of Economic Surveys, Vol. 8 No. 3, pp. 203-49.
Srairi, S. (2010), “Cost and pro?t ef?ciency of conventional and Islamic banks in GCC countries”,
Journal of Productivity Analysis, Vol. 34 No. 1, pp. 45-62.
Stiglitz, J. (2001), “Principles of ?nancial regulation: a dynamic portfolio approach”, The World
Bank Research Observer, Spring, pp. 1-18.
Su?an, F. (2007), “The ef?ciency of Islamic banking industry in Malaysia: foreign vs domestic
banks”, International Journal of Systems and Ethics, Vol. 23 No. 3, pp. 174-92.
Sundararajan, V. and Errico, L. (2002), “Islamic ?nancial institutions and products in the global
?nancial system: key issues in risk management and challenges ahead”, IMF Working
Paper No. 02/192, International Monetary Fund, Washington, DC.
Weill, L. (2002), “Leverage and corporate performance: a frontier ef?ciency analysis on European
countries”, Universite´ Robert Schuman, Institut d’Etudes Politiques.
Weill, L. (2011), “Do Islamic banks have greater market power?”, Comparative Economic Studies,
Vol. 53 No. 2, pp. 291-306.
Yudistira, D. (2004), “Ef?ciency in Islamic banking: an empirical analysis of eighteen banks”,
Islamic Economic Studies, Vol. 12 No. 1, pp. 1-19.
ARJ
25,3
206
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Further reading
Hussein, K.A. (2004), Operational Ef?ciency in Islamic Banking: The Sudanese Experience,
Islamic Development Bank, Jeddah.
About the author
Lucia Dalla Pellegrina is an Assistant Professor at the University of Milan-Bicocca where she
teaches Macroeconomics and Advanced Macroeconomics. She is Fellow of the Paolo Baf? Center
for Economic and Financial Research, Bocconi University and a faculty member of the PhD in
Economics at the University of Pavia where she teaches Law and Finance. Lucia Dalla Pellegrina
can be contacted at: [email protected]
Capitalization
207
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
6
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_606921860.pdf