ISLAMIC DEVELOPMENT BANK

ISLAMIC RESEARCH AND TRAINING INSTITUTE

Habib Ahmed

Research Paper
No.59
 ISLAMIC DEVELOPMENT BANK
ISLAMIC RESEARCH AND TRAINING INSTITUTE

Habib Ahmed

Research Paper
No.59

1
 ISLAMIC DEVELOPMENT BANK
ISLAMIC RESEARCH AND TRAINING INSTITUTE

King Fahd National Library Cataloging-in-Publication Data

Ahmed, Habib
A Microeconomic Model of an Islamic Bank.
Jeddah.
40 pages;17 X 24 cm. ( Research paper ; 59 )

ISBN: 9960-32-124-x
1- Islamic economy 2- Microeconomy
I- Title II- Series
330.121 dc 2111/23

Legal Deposit No. 2111/23

ISBN: 9960-32-124-X

The views expressed in this book are not necessarily those of the Islamic Research
and Training Institute or of the Islamic Development Bank.
References and citations are allowed but must be properly acknowledged.

First Edition 1423H (2002)

3
4
 CONTENTS

Foreword . 7
1. Introduction... 9
2. Nature of Banks: Conventional and Islamic.. 10

2.1. Nature of Conventional Banks 10

2.2. Nature of Islamic Banks... 12

2.2.1. Islamic Bank: Commercial Bank or Investment

Intermediary?... 13
2.2.2. Asset-side versus Liability-side Mudarabah... 13

2.2.3. Equity versus Deposits on the Liability Side 14

2.2.4. Primary and Secondary PS Assets 14

2.2.5. Nature of Risks. 15

3. Banking Theories and Islamic Banking.. 16

4. Model... 18

4.1. Islamic Bank as a Provider of Liquidity 18

4.1.1. Liability Side 19

4.1.2. Asset Side . 22

4.1.3. Risk-Return Profiles of Assets and Vulnerability of

Banks.. 25
4.2. Islamic Bank as Investment Intermediary. 27

5. Some Performance and Operational Matters of Islamic

Banks ... 29

5
 5.1. Stability of Islamic Banks. 29

5.2. Liquidity Management... 30

5.3. Risk Management Strategies 31

5.3.1. Stabilization of Rate of Return on Deposits. 31

5.3.2. Asset Management: Combining PS with FI mode.. 32

5.3.3. Bank Capital Allocation. 32

5.4. Quasi-Equity Deposits and Securitization. 33

6. Conclusion.. 34

References .. 36

6
 FOREWORD

The Islamic Research and Training Institute (IRTI) of the Islamic

Development Bank (IDB) was established in 1401H (1981G) to undertake
research for enabling the economic, financial and banking activities in Muslim
countries to conform to Shariah. In order to discharge its responsibilities,
IRTI pays special attention to basic and applied research in the areas of Islamic
economics, banking, and finance relying on its in-house research capabilities
and mobilizing external scholars.
From an urge to stay away from interest, which is prohibited by Shariah,
Islamic banking was revived some three decades ago. When the idea of Islamic
banking was mooted, it was proposed that Islamic banking would adopt profit-
sharing modes of financing (mudarabah and musharakah). Experience,
however, shows that Islamic banks predominantly use fixed-income instruments
(like murabahah and ijarah) on the asset side. While some explanations have
been given to explain this trend, this paper examines the issue by modeling an
Islamic bank. The model, using concepts from contemporary banking theories,
identifies the fundamental nature of Islamic bank as an intermediary. The
research paper explains why Islamic banks choose relatively less risky fixed-
income instruments and argues that the share of profit sharing instruments can
be increased if Islamic financial institutions operate as investment
intermediaries instead of liquidity providers.
As the Islamic financial industry operates in highly competitive markets
it must be able to offer attractive risk-return opportunities to both investors and
user of funds. The future growth and sustainability of Islamic financial industry
will depend on the ability of the researchers and practitioners to understand the
nature of Islamic instruments and institutions and develop these in line with the
principles of Shariah. To do so, the different aspects of Islamic banking and
finance need to be studied analytically. This paper is one of the first
microeconomic models of an Islamic bank. I hope that the publication of this
research paper will stimulate further research along these lines.

Mabid Ali Al-Jarhi

Director, IRTI

7
8
A MICROECONOMIC MODEL OF AN ISLAMIC BANK

1. Introduction
With the inception of Islamic banking in the early 1970s, it was believed
that it would take the form of two-tier mudarabah model. Proponents of Islamic
banking pointed out the advantages of profit-sharing modes of financing over
the conventional interest based financing. These include increase in investment,
allocative efficiency, stability, equity and reduction of poverty. 1 Experiences,
however, show that fixed-income short-term instruments (like murabahah,
mark-up financing) are the dominant modes of financing used by Islamic banks.
Various reasons are given to explain the lack of use of profit-sharing modes by
Islamic banks, the most significant being the moral hazard problem. 2 This is
disheartening to the proponents of Islamic banking as some of them fear that
using fixed-income instruments may not be in the true spirit of Islam as it
represents the status quo (Chapra 1985, p.171 and Siddiqi 1983, p.139). Though
an Islamic economy cannot be a "pure profit sharing" economy, there is an
aspiration to have a balanced mix between fixed-income and profit-sharing
(hereafter PS) modes of financing (Al-Jarhi 1999). The success of the use of PS
modes of financing, however, will depend on understanding the problems
associated with their use. This paper addresses this issue by studying the nature
of Islamic banks in light of contemporary banking theory. By modeling an
Islamic bank, the paper seeks to identify the causes of the modest use of PS
instruments and then discusses some related operational issues.
The paper is organized as follows. In section 2 we discuss different
aspects of conventional financial intermediation and features of Islamic banks.
Section 3 reviews literature on contemporary banking theory, both conventional
and Islamic. Section 4 first outlines a model of an Islamic bank as liquidity
provider. After discussing the nature of liability of the bank, we show how the

1
The advantages of Islamic banking are discussed in Chapra (1985) and Khan
(1995a), and Siddiqi (1981 and 1983).
2
For a discussion on the lack of profit sharing modes of financing by Islamic banks
see Khan (1995b).
9
composition of the asset is determined. The section then discusses the role an
Islamic bank as an investment intermediary. Section 5 elaborates some
operational issues related to Islamic banks. The paper ends with a conclusion.

2. Nature of Banks: Conventional and Islamic

The objective of financial institutions is to maximize profit and
shareholder value-added by providing different financial services. This is
achieved mainly by managing risks (Heffernan 1996, p. 163, Scholtens and
Wensveen 2000, p. 1247). Risks can be classified in different ways. One broad
classification is to distinguish between business risk and financial risk. While
business risk arises from the uncertainty of income flows due to the nature of a
firm's business, financial risk exists in leveraged firms (Reilly and Brown 1997,
p. 20). Some of the other specific risks that financial institutions face are credit
risk, liquidity risk, interest rate risk, operational risk, and capital or solvency
risk. 3
From a management point of view, Oldfield and Santomero (1997)
divide risks into three types: risks that can be eliminated, those that can be
transferred to others, and risks that can be managed by institutions. The practice
of conventional financial intermediaries is to take up activities in which risks
can be efficiently managed, shift risks that are transferable, and avoid activities
that impose additional risks upon them (Santomero 1997). Thus, to understand
the nature of any financial institution one has to examine the kind of risks they
face and investigate how they deal with it. To discuss Islamic banks in same
perspective, we examine the characteristics of conventional financial
institutions. This is done next.

2.1. Nature of Conventional Banks

An effective financial system "facilitates the efficient life-cycle
allocation of household consumption". 4 To enable this allocation, different
financial institutions have evolved to meet the various needs of households.
Financial intermediaries are broadly classified as depositary institutions,

3
See Koch (1995) and Heffernan (1996) for a discussion on different types of risks
that financial institutions face.
4
A quote by R.C. Merton cited in Friexas and Rochet (1999, p. 1)
10
contractual intermediaries, and investment intermediaries. 5 Commercial banks,
forming the bulk of depositary institutions specialize in intermediation,
obtaining most of their loanable funds from deposits of the public. As banks
deal with interest-based deposits, they face financial risks. They deal with risks
by using interest-based contracts on the asset side backed by collateral.
Commercial banks are stock intermediaries in the sense that they are owned by
the stockholders of the bank who receive a share of the profit derived after
deducting all costs including interest payment to depositors from income.
Investment intermediaries offer liquid securities to the public for long-
term investment. Investment intermediaries are mutuals, with customers being
the owners who receive income in the form of dividends and capital gains.
Being stockowners, customers can vote in matters that affect the institution
(Rose and Fraser 1988, p. 15). Investment intermediaries typically invest in
secondary markets and, as such, avail investors opportunities to hold securities
of private and public institutions. Investment intermediaries being mutuals,
eliminate financial risks but take on business risks.
Note that depending on the regulatory framework of a specific country,
financial institutions may perform different functions. For example, universal
banks are consolidated institutions providing different financial services that
may include intermediation, investment management, insurance, brokerage, and
holding equity of non-financial firms (Heffernan 1996). A simple case of
universal bank is one in which the liability is the same as that of the commercial
banks, but the asset side differs. While the assets of a typical commercial bank
are in form of loans only, universal banks can hold equity along with loans. By
holding equity positions, universal banks can essentially get involved in the
decision making and management of the firm. 6
It is important to point out the role of financial intermediaries from the
depositors' perspectives. An efficient financial system should be able to fulfil

5
Contractual intermediaries constitute insurance firms and pension funds. These
institutions are not discussed as they are not related to this paper.
6
A distinction is made between narrow banking and fractional banking. Wallace
(1996), however, shows that narrow banking requiring banks to back demand
deposits by safe short-term liquid assets (implying 100 percent reserve ratio)
eliminates the role of banking. The solution under narrow banking is similar to that
of individuals acting alone (autarky) so that there is no need for financial
intermediation by banks. The implication of this conclusion on Islamic banks does
not directly concern this paper.
11
the current and future financial needs of the people. Specifically, individuals
hold a part of their savings in form of liquid assets for current and unforeseen
needs and invest the rest in assets with different risk/return/maturity profiles.
While investment intermediaries provide the investment opportunities to the
public, a unique role of commercial banks is the provision of liquidity and
access to a safe and efficient payment system through deposits (Friexas and
Rochet 1999, p. 2). Liquidity risk is the uncertainty with regards to an asset's
convertibility into cash and price at which it is done (Reilly and Brown 1997, p.
21). In conventional banking, deposit contracts ensure liquidity to depositors,
both in terms of cash and price (i.e., interest paid in deposits). 7

2.2. Nature of Islamic Banks

Iqbal et.al. (1998) distinguish two models of Islamic banks based on the
structure of the assets. 8 The first is the two-tier mudarabah model that replaces
interest by profit-sharing (PS) modes on both liability and asset sides of the
bank. In particular, in this model all assets are financed by PS modes of
financing (mudarabah). This model of Islamic banking will also take up the role
of an investment intermediary, rather than being a commercial bank only
(Chapra 1985, p. 154). The second model of Islamic banking is the one-tier
mudarabah with multiple investment tools. This model evolved because Islamic
banks faced practical and operational problems in using profit-sharing modes of
financing on the asset side. As a result, they opted for fixed-income (FI) modes
of financing. The FI instruments include murabahah (cost-plus or mark-up
sale), installment sale (medium/long term murabahah), bai-muajjal (price-
deferred sale), istisna/salaam (object deferred sale or pre-paid sale) and ijarah
(leasing). 9
On the liability side of Islamic banks, saving and investment deposits
take the form of profit-sharing investment accounts. Investment accounts can be
further classified as restricted and unrestricted, the former having restrictions on
withdrawals before maturity date. Demand deposits or checking/current

7
8 The exception being the case of a bank run with no deposit insurance.
Iqbal, et.al. (1998) mention three models, the third one being the case where Islamic
banks work as agent (wakeel), managing funds on behalf of clients on basis of fixed
commission.
9
For a discussion on these modes of financing see Ahmad (1993), Kahf and Khan
(1992), and Khan (1991).
12
accounts in Islamic banks take the nature of qard hasan that are returned fully
on demand. Using PS principle to reward depositors is a unique feature of
Islamic banks. This feature, however, changes the nature of risks that Islamic
banks face. Some important issues related to Islamic banks are discussed below.

2.2.1. Islamic Bank: Commercial Bank or Investment Intermediary?

The Islamic bank described above appears to have characteristics of both
investment intermediary and commercial bank. The ownership pattern of the
Islamic bank resembles that of a commercial bank as the depositors do not own
the bank and do not have voting rights. In Islamic finance parlance, this means
that while musharakah contract characterizes the equity owners, deposits take
the form of mudarabah contracts. 10 An Islamic bank, however, has similarities
with an investment intermediary as it shares the profit generated from its
operations with those who hold savings/investment accounts. After paying the
depositors a share of the profit, the residual net-income is given out to the
shareholders as dividends.

2.2.2. Asset-side versus Liability-side Mudarabah

The PS contract on the asset side and liability side of an Islamic bank
are different. Historically, mudarabah and musharakah were partnerships for
specific projects. The fund provided by the fund-owner (rab-ul-maal) to the
fund-user (mudarib) is not debt, as the ownership of capital remains with its
provider during the life of the project. Upon the completion of the project,
capital and share of profit is returned to its owner. An appropriate contemporary
financing technique using PS modes would be diminishing musharakah
principle in which the bank sells off its shares to the entrepreneur in
installments. 11 Though the nature of mudarahah contracts on the asset side of
an Islamic bank will be different from the traditional model, the contract may
still attach funds to individual projects for a relatively longer time. This,
however, is not the case with the mudarabah contract on the liability side (i.e.,
deposit). The deposits can be withdrawn by their holders at relatively short

10
One difference between musharakah and mudarabah is that while in the former the
financier has a role in management of the project, it does not in the latter case.
11
For a discussion on diminishing musharakah see Bendjilali and Khan (1995).
13
notice. 12 In other words, whereas on the liability side the contract is usually
unrestricted (in terms of liquidation), on the asset side it is restricted (funds flow
after the contract matures).

2.2.3. Equity versus Deposits on the Liability Side

Other than ownership feature pointed above, equity holders and
depositors are fundamentally different in some other respects. These differences
can be viewed from two perspectives. One way to look at the difference is the
attitude of the equity-holders and depositors towards risk and return. Depositors
will be more risk averse than equity holders and opt for a low risk/return deposit
contracts (Gangopadhyay and Singh 2000). The implication is that the equity
holders are willing to take up more risks that are compensated by a higher
expected rate of return. 13
Another important difference between equity and deposits relates to
liquidity. When a depositor needs cash, he can withdraw funds from his
deposits reducing the liability of the bank. Note that with restricted deposits this
can be done at a cost. An equity-holder in need of cash, however, sells securities
in the secondary market without affecting the equity-capital held by the bank.
As will be seen, this difference has important implications on the balance sheet
and liquidity management of the bank (discussed in section 5).

2.2.4. Primary and Secondary PS Assets

Islamic banks may differ from investment intermediaries in one
important way. This difference relates to the role of the institutions in the
financial system. Whereas most investment intermediaries are asset managers,
banks are involved in financing. The former institutions deal in the secondary
markets where only the ownership of existing securities (secondary PS assets)
change hand. Other than initial public offerings (IPOs), investment
intermediaries activities do not directly affect the capital of firms. While

12
Restricted investment accounts may be relatively less liquid. Nevertheless, these
deposits can be withdrawn at a cost in a relatively shorter period of time. For a
discussion on the nature of mudarabah deposits in Islamic banks, see Ahmad
(1997).
13
Empirically it has been found that rate of return on equity is much higher than can
be explained by the risk-premium. This is termed as the 'equity premium puzzle' in
the literature. See Kocherlakota (1996) and Mehra and Prescott (1985) for a
discussion on equity-premium puzzle.
14
Islamic banks can invest in securities (as mutual funds do), PS modes on the
asset side of a bank are designed to be financing instruments. In other words,
Islamic banks can directly finance firms (primary PS assets), thereby increasing
their capital and productive capacities. Financing primary PS assets can be done
on the principle of diminishing musharakah. As the shares of primary PS assets
cannot be traded in secondary markets, they are not as liquid as secondary PS
assets.

2.2.5. Nature of Risks

Using PS modes changes the nature of risks that banks face. As demand
deposits are considered as qard-hasan or debt and its repayment is guaranteed,
the bank faces financial risks in using these deposits. By sharing the returns
with the saving/investment depositors the bank eliminates financial risks but
introduces some other risks. As the depositors are rewarded on a profit-loss
sharing (PLS) method, they share the business risks of the banking operations of
the bank. Rewarding depositors with banks variable profit introduces some
additional risks (AAOIFI 1999, p.7). First is 'fiduciary risk' that arises from
breaching of investment contract or mismanagement of funds by the bank.
Second, 'displaced commercial risk' is the transfer of the risk associated with
deposits to equity holders. This arises when under commercial pressure banks
forgo a part of profit to pay the depositors to prevent withdrawals due to a lower
return. 14

The withdrawal possibility due to lower rate of return is a unique feature

of an Islamic bank as returns on their deposits can vary. Depositors want to
protect the real value of their assets. In an Islamic environment the real value of
deposits will decline not only due to inflation, but also due to zakat dues. A
variable rate of return on saving/investment deposits further introduces an
additional source that may affect real value. PLS returns on deposits add to the
uncertainty regarding the real value of deposits. Asset preservation in terms of
minimizing the risk of loss due to a lower rate of return may be an important
factor in depositors' withdrawal decisions. From the banks perspective, this

14
Though there is no formal Shariah approval that permits transferring profit from the
owners of the bank to depositors, there appears to be no Fiqhi objections to do the
same. For example, Fahmy (2001, p.22) is of the opinion that as long as there is
positive profit (no loss), the owners of the bank can opt to transfer a part of their
profit to the depositors.
15
introduces a withdrawal risk that is linked to the lower rate of return. A recent
survey of 17 Islamic financial institutions indicates that these institutions
consider withdrawal risk more serious than some of the traditional risks (like
credit risk, market risk, liquidity risk, etc.) that financial institutions face. 15 In
our model of an Islamic bank, we use withdrawal risk and point out how the
displaced commercial risk arises when discussing operational aspects in
Section 5.

3. Banking Theories and Islamic Banking

In this section, we review literature on contemporary conventional and
Islamic banking theories. Conventional microeconomic models of banks can be
broadly classified into three types. 16 The first kind focuses on the asset side of
the banks, the second type examines the liability side, while the third combines
elements of both assets and liabilities. A brief description of these models is
given below.
Models focusing on asset side of the balance sheet of the bank view
banks as institutions acting as specialized in monitoring to minimize
asymmetric information problems involved in investment (Diamond 1984 and
Ramakrishan and Thakor 1984). Two types of information problems can arise
in financing projects. First, adverse selection problem arises before a contract is
signed as the borrowers (firm owners/managers) have more information on the
project than the investors (or bank). The investors, as such, are in
disadvantageous position to sort out the good projects for investment. Second,
moral hazard problem may arise after the investment has taken place. In this
case, the investors cannot ascertain if the borrowers behavior is in accordance
with what is stipulated in the contract. By acting as intermediaries, banks
minimize the asymmetric information problems and lower the monitoring costs
that would have encountered if each investor would have invested individually.
By pooling funds from individual investors (depositors) for different investment
projects, they can also diversify their assets to reduce the risks involved.
Theoretical work of Diamond and Dybvig (1983) and Gangopadhyay
and Singh (2000) represent the liability side approach to banking theory. These

15
See Khan and Ahmed (2001, Chapter 3) for details of the survey.
16
For surveys of banking theories, see Bhattacharya, et.al. (1998) and Bhattacharya
and Thakor (1993).
16
models focus on the banks' role of liquidity transformation. They assume that
individuals experience random shocks in future consumption and deposit
contracts serve as insurance to such shocks. Without an intermediary, individual
investors will be locked in illiquid investments. In case of a shock, investors are
able to liquidate these long-term investments at a cost. By providing deposit
contracts to individuals, banks transform liquid liability into illiquid assets.
Diamond and Dybvig (1983) model of deposit financing yields multiple
equilibria, one of which can cause bank runs. Gangopadhyay and Singh (2000)
show how by converting deposits into equity a bank-run can be avoided even
without deposit insurance.
Bhattacharya et.al. (1998) provide an integrated model in which banks
provide both asset and liability intermediation. In this model, depositors and
entrepreneurs are distinct and the bank borrows from the depositors and lends to
entrepreneurs through independent debt contracts. A coordination failure,
however, can occur when a large number of investors (depositors) withdraw
funds, causing a bank-run. Bank runs adversely affects the economy as they
force liquidation of entrepreneurs' projects before maturity.
There is a large volume of literature on different aspects of Islamic
banking. The earlier work on the subject focused on the PS modes of financing
and its implications on the economy (for example see, Chapra 1985, Khan 1987,
Khan 1995a, and Siddiqi 1981, 1983 and 1991). Most papers, however, assume
Islamic banks would use PS modes of financing without addressing why Islamic
banks have failed to adopt these instruments in practice. Some more recent
studies explain the causes of Islamic banks' lack of use of PS modes of
financing.
Khan (1995b) discusses different reasons that hamper the use of PS
modes of financing. He focuses on the asset side of the bank and discusses the
demand (fund user) and supply (bank or the investor) sides. He points out,
among other things, that firms prefer FI modes of financing as they can reinvest
their surpluses to enhance growth. Aggarwal and Yousef (2000) model an
Islamic bank and explain the lack of use of sharing modes due to agency
problems. Their paper also concentrates on the asset side of the bank's balance
sheet. They outline the conditions under which an equity contract is optimal in
financing. Profit sharing equity contracts work well when the cost of the project
is small and rate of diversion (due to moral hazard) is low. They conclude that
PS (equity) contracts are not being used due to severity of moral hazard
17
problem faced by the banks. Mirakhor (1987) uses a micro-model to explain the
dominance of short-term instruments used by Islamic banks. He examines the
balance sheet and maintains that banks are worried about bankruptcy and would
keep bank losses lower than total capital. To do so the banks invest in short-
term mark-up instruments.
This paper differs from earlier work in significant ways. We use
elements from contemporary banking theory to model an Islamic bank. We
examine both sides of the balance sheet and analyze the effect of liability side
on the composition of the assets of a bank. In particular, we consider the risks
associated with PS modes to understand the nature of Islamic bank and its
operations. The details of the model are given below.

4. Model
An effective financial system fulfils the current and future financial
needs of the people. While households hold a part of their savings in form of
liquid assets for current and unforeseen needs, they invest the rest in assets with
different risk/return/maturity profiles. As mentioned above, a unique role of
commercial banks is the provision of liquidity and access to a safe and efficient
payment system through deposits. An Islamic bank, however, would also take
the role of an investment intermediary. Given these features, we proceed in two
steps to develop a theoretical model of an Islamic bank. First, we examine an
Islamic bank providing liquidity services to depositors. As such, the bank offers
demand deposits and savings (unrestricted investment) deposits. The depositor
can withdraw funds from these accounts without any cost. In the second step,
we include the investment intermediary nature of Islamic banking by
incorporating (restricted) investment deposits. Withdrawing funds from these
accounts entails a cost to the depositors. Note that while the bank guarantees
payment of funds on demand deposits, it shares profit/loss with
saving/investment account holders. 17

4.1. Islamic Bank as a Provider of Liquidity

17
In reality, a whole spectrum of different accounts can be offered by banks. We
include savings account not only because Islamic banks provide these accounts, but
also because it enables us to study an Islamic bank offering liquidity services with
returns like a conventional interest-bearing bank.
18
 Our partial model of an Islamic bank uses an integrated approach by
studying both the asset and liability sides of the bank. In modeling a bank as
liquidity provider, we follow Diamond and Dybvig (1983) and Gangopadhyay
and Singh (2000). They classify depositors randomly as those having short and
long-term needs for funds in the future. As a result, their papers focus on an
exogenous risk that banks face due to random withdrawals of deposits. As
depositors get a share of profit of the bank in an Islamic banking setting,
withdrawals due to asset preservation arise endogenously. The return/risk on
deposits will be closely linked to the corresponding figures of the assets of the
bank. As mentioned above, this feature introduces fiduciary, displaced
commercial, and withdrawal risks in Islamic banking. We introduce the latter
risk in our model.
Assume that no deposit insurance exists. Assume also that the bank has
equity, E, and offers demand deposits (Dd) and savings account (DS) only. Thus,
the total deposits in the bank is D= Dd+DS held by m number of clients. To
keep our analysis manageable we consider only two kinds of assets other than
reserves; viz. fixed-return assets (e.g., murabahah and ijarah) and profit-sharing
assets (modarabah). The balance sheet of the bank indicates that total assets (A)
must equal deposits and equity (i.e., A=D+E).
Conventional banking theory maintains that fixed interest-based loan
(backed by collateral) is an optimal contract in minimizing the moral hazard
problem arising in financing. In the conventional interest based contract, a firm
taking a loan of I at an interest rate r to finance a project pays an amount of
D0=(1+r)I in the next period. Note that here the bank shifts risk to the firm. To
have a discussion in some perspective, we take this optimal case as a benchmark
for analyzing the Islamic cases.

4.1.1. Liability Side

As mentioned above, people have different financial needs and place
their funds in different assets accordingly. In forming their portfolios, a part of
savings is put aside in liquid assets so that current and unexpected expenses can
be met. The rest will be put in assets/investments with different risk-return
profiles. People will have different attitudes towards risk and return for various
kinds of assets. While agents will be relatively more risk-averse with assets
used for liquidity (e.g., demand/saving deposits), they will accept more risk
with other assets with higher expected returns.
19
 We assume that a share of the funds set aside for liquidity purposes is
deposited partly in demand deposits and partly in savings account. People use
savings account mainly to protect the real value deposits from inflation and
zakat dues. Let the expected rate of return on saving deposits be re and standard
deviation of r. As depositors of Islamic banks get a share of profit, the return
and risk of deposits will be closely linked to that of the banks assets. That is,

re = f(r), f'>0; and

r =g(), g'>0;

where r and are the rate of return and standard deviation of assets of
the bank.
While a depositor knows that there are risks involved on returns on
savings account, he may react if the rate of return of the bank is too low. 18 If the
minimum rate of return that is acceptable to depositor m is rm, then a tolerance
level of this depositor (m) is defined as the difference between the expected and
this minimum return on deposits (i.e., re-rm=m). Note that while attitude
towards risk will be the main determinant of this index, for a Muslim it may
also be influenced by the level of his/her religious belief and feelings that
induces him/her to use an Islamic bank in the first place. Ceteris paribus, a fall
in real value of deposits by more than the tolerance level will lead to a
withdrawal of funds by the depositors to other alternatives. Thus, m represents
the minimum decrease in the rate of return acceptable to depositor m in order to
keep his money with the bank. Specifically, if the actual rate of return (ra) is
low enough to give (re- ra)>m, then the depositor will withdraw the funds from
the bank. We call the resulting withdrawals as asset-preservation withdrawals
and this points to the withdrawal risks faced by Islamic banks.
Thus, total withdrawals W from the banks during a period of time (say a
week) equals,

W= D+V(q), 0<<1; (1)

18
Note that being a partial equilibrium model, we are assuming other things constant.
The reaction of depositors to a lower rate of return will depend on whether bank
under-performs in isolation or due to a general downturn of the economy. This is
discussed in Section 5.
20
 where D represents the average amount of normal withdrawals during
the period for their current transactional needs and V represents the asset-
preservation withdrawals described above. Asset-preservation withdrawals
increase with q, the probability that the minimum rate of return that depositors
expect is not guaranteed (i.e., when re- ra >m). Note that in a conventional
banking setting with deposit insurance, the probability q is close to zero, as the
deposit along with the interest payment is guaranteed. In Islamic banking,
however, as the rate of return on deposits can vary as it depends on the overall
yield of the bank's portfolio, the probability will be positive. We use
Chebychev's inequality to determine this probability for individual m. 19 For a
random variable r with expected value of re and variance r, the Chebychev's
inequality is given by,
q=Prob (|r-re|> m) 2r/m2 (2)
The probability of asset-preservation withdrawals, q, may increase with
the risk of the returns on deposits r, and may decrease with the individual
tolerance factor m. As the rate of return of deposits is directly linked with rate
of return on bank's portfolio, the asset-preservation withdrawals V will be
affected by the overall risk of the bank's portfolio (). Thus, the asset-
preservation withdrawals can be written as,
V=V( , m)= V( , re-rm) (3)
Note that asset-preservation withdrawals varies directly with the risk of
the assets ( ) and inversely with the tolerance factor (m= re-rm). In other words,
withdrawal risk of an Islamic bank increases with overall risk of the banks
portfolio.
Tolerance level for asset preservation m will differ from one depositor to
another as people have different attitudes towards it. At one extreme, a
depositor may expect a positive rate of return after factoring for inflation and
zakat while at the other end a depositor may not worry about these at all.
Between the two extremes, there will be an array of depositors who will have a
varied tolerance levels of how much a fall in the value of real balances they
would sustain. If w=W/A=W/(D+E) is the withdrawals to asset ratio, then this
variation of tolerance among depositors can be represented by an upward
sloping withdrawal curve, w(), as shown in Figure 1. This curve indicates that

19
See Greene (1997, p. 116) for Chebychev's inequality.
21
 a rise in the overall risk of a banks assets () increases the probability of
withdrawal of deposits from the bank (w). Note that the predictable component
of withdrawals due to current transactional (D) needs is shown by w0 in the
figure.

4.1.2. Asset Side

We have noted that the variance of the returns on deposits is directly
linked to the variance of the returns on assets of Islamic banks and the
probability of withdrawals may increase as the bank holds more risky assets.
This leads us to examine the risk-return characteristics of different assets held
by an Islamic bank. We assume that the bank invests in primary PS assets so
that it directly invests and increases the capital of the firm. 20 Specifically, the
bank can invest an amount of I in a representative firm that yields a probability
distribution of profit (net-income) of yi as shown in Table 1. 21 Let the mean and
variance of the outcome of the project be y and y respectively. Note that some
of these outcomes (h<n) will be negative (i.e., y1<y2<< yh=0). Given this
probability distribution, we outline the risk-return features of FI and PS modes
of financing next.
Case 1 (Fixed-Income Assets): The bank finances the entrepreneur by
using an Islamic fixed-income (FI) mode of financing involving payment-
deferred sale (e.g., murabahah). The firm pays the bank D0=(P+)Q, where P
is the price of assets (Q) sold to the firm by the bank and is the mark-up. Note
that PQ=I. Though Islamic FI mode of financing is different from the
conventional interest based financing, the financial flows from these contracts
appear similar. 22 The expected return and risk of the bank in this transaction are
Rfb=f=D0, and fb(0) respectively. 23 The corresponding figures for the firm are

20
The implication of investing in secondary PS assets is discussed in Section 5.
21
The implicit assumption of yi being independent, identically distributed (i.i.d.) is
made to keep the analysis simple. Other than avoiding complications, the i.i.d.
assumption is appropriate for analysis as the focus of the paper is to compare
individual modes of financing and not issues related to a set of assets (like benefits
of diversification).
22
For a discussion on the differences between conventional interest-based loan and
Islamic sale-based debt (murabahah), see Chapra and Khan (2000, p.10) and Al-
Jarhi (2000).
23
In a murabahah contract, risk borne by the bank is limited up to the point when the
good is handed over to the entrepreneur. Once the good is handed to the
22
 ase 2 (Profit-Sharing Assets): We consider a mudarabah type profit-
sharing (PS) contract in which the bank provides funds and the entrepreneur
acts as an agent (mudarib) to implement the project. Let the profit-sharing ratio
be so that the bank gets a share of percent of the profit (and 1- percent is
retained by the entrepreneur). 24 Note, however, that under a mudarabah
contract all loss from the project is borne by the bank. 25 As shown in Table 1
below, the probability distribution of the outcomes in a PS contract for the bank
and the entrepreneur is different from that in the FI case. In the PS case, the net-
income of the entrepreneur in states j=1h becomes zero (i.e., ye1
=ye2==yeh=0) and equals (1-)yk in states k=h+1, n. The outcome for the
bank is yj<0 for states j=1,2,h and yk for states k=h+1, n.

Table 1
Probability Distribution of Profit of the Project,
the Bank, and the Firm in a Profit-Sharing Contract

Probability Project Profit Bank Firm

(pi) (yi) (PS-case) (PS-case)
p1 y1<0 y1<0 0
p2 y2<0 y2<0 0
. . . .
. . . .
. . . .
ph yh=0 yh=0 0
ph+1 yh+1>0 yh+1>0 (1-)yh+1>0
. . . .
. . . .
. . . .

entrepreneur, all risk is borne by him, making the risk of the bank very small in a
murabahah contract (Khan 1991, p. 14.)
24
We assume is exogenously given in the model determined by mutual agreement of
the bank and the firm. For a discussion on determining the profit-sharing ratio in
project financing, see Ahmed (2001).
25
In the mudarabah contract, the profit is shared, but loss is borne by the owner of the
capital. The entrepreneur (mudarib) loses his efforts.
23
 pn yn>0 yn>0 (1-)yn>0
The assumption of asymmetry in outcomes in loss and profit situations
alters the expected income and risk (variance) of both the bank and the firm.
The expected income and variance of the entrepreneur in the PS case are Rpe=
(1-) kpkyk and Vpe= (1-)2k pkyk2-(Rpe)2 respectively. The corresponding
values of the bank are Rpb and Vpb respectively, as shown below.
Rpb = p =jpjyj +kpkyk, (4a)
Vpb =p2= j pjyj2 + 2k pkyk2- Rpb2. (4b)

The asymmetry in the rewards in the mudarahah contract (i.e., sharing

profit without sharing losses) has important implications on the expected return
and risk of the bank. Accepting all financial losses biases the return and risk
against the bank. This is apparent if we compare the profit sharing (with no loss
sharing) (PS) case with the case where both profit and loss are shared. We call
the latter as PLS case. The expected return and variance of the bank in case of
PLS case is given below,
Rplb = jpjyj +kpkyk (5a)
Vplb =(bpl )2= 2j pjyj2 + 2k pkyk2- (Rplb) 2. (5b)

Noting that yj is non-positive, a comparison of the PS case with the PLS

case gives the following results
Rplb/Rpb >1; (6a)
Vplb /Vpb <1. (6b)

The above results indicate that due to asymmetry in profit sharing (PS),
the banks expected return is lower and risk higher than the symmetric PLS
case. In other words, due to the asymmetry in the weights in loss and profit, the
bank share of expected income is less than the profit-sharing ratio. Similarly,
the bank has a larger share in the risk than the ratio at which profit is shared. To
compensate for this bias, the profit sharing ratio should be higher to factor in the
additional risk and lower profit. For the purposes of this paper, however, we
need to discuss the implication of the risk-return elements and its effects on the
banks operations. To do so, we summarize the features of assets held by the
bank in Table 2 below.

24
 Table 2
Distribution and Financial Attributes of Assets of the Bank
Cash Fixed- Profit- Portfolio
income sharing
Assets Assets
Share c f p c+f+p=1
Expected c(=0) f (>0) p (> f >0) =ff+p0
Rate of
Return ()
Risk ( ) c(=0) f (0) p (> f 0) =ff+pp

Table 2 shows that assets of the bank kept as cash (reserves), invested in
fixed income (FI) assets and profit sharing (PS) assets respectively as indicated
by their corresponding shares i (i= c, f, and p). The risk/return feature of the
portfolio of the bank will depend on the relative shares of FI assets and PS
assets. For example holding more PS assets will increase the risk of the bank's
portfolio of assets. The relationship between the share of FI asset and the risk of
the bank's portfolio is given in Figure 1 by the f() line. This line indicates that
an increase in the share of the FI assets or a decrease in the share of PS assets
(a larger f) lowers the overall risk () of the portfolio. Note that risk does not
equal zero when f=1 as there are some risks (f) associated with FI assets. We
discuss the implications of risk on bank operations next.

4.1.3. Risk-Return Profiles of Assets and Vulnerability of Banks

Total withdrawals have a predictable component, D, and an
unpredictable part,V(q). As Equation (3) shows, the unpredictable component
depends on the risks on returns on deposits. If the volatility (or risk) of returns
on assets increase, the probability of withdrawals to preserve their asset value
increases. In other words, withdrawal risk of the bank increases. If a bank is
unable to satisfy large amount of unexpected withdrawals, then the
consequences can be disastrous. When depositors learn about the inability of a
bank to satisfy withdrawal requests, a large number of them rush to withdraw
their money causing a bank-failure. The following conditions show the
vulnerability of the bank:
25
 i) w c => Bank is not vulnerable
ii) w > c => Bank vulnerable to a bank-failure.
where w has been earlier defined as withdrawals as a percentage of
liabilities and equity. A bank is vulnerable to a failure if the withdrawals exceed
the reserves held by the bank. Given the above framework, in order to avoid
bank-run a bank should keep the unpredictable component of withdrawals, V
(i.e., asset-preservation withdrawals) as low as possible. As V is directly related
to risks of assets ( ), the underlying principle in managing assets is to keep the
risks low.
The results of the model are summarized in Figure 1 above. The
interaction between the asset and liability side explains why Islamic banks are
not using PS mode of financing to finance projects. As mentioned earlier, the
risk of bank's portfolio associated with choosing FI assets and PS assets is
shown by the f() line. As the share of FI assets (f) increases the risk of the
portfolio ( ) decreases. The effect of risk on the liability side is shown by the w
line. As has been pointed out, the withdrawal to asset share (w) increases as the
risks on bank's portfolio () increases. This is shown by the upward sloping w
curve in the figure. The vulnerability condition of the banks can be ascertained
by comparing the share of reserves (cash) that the bank holds relative to the
total withdrawals. While cash ratio (c) may be exogenously given by the
regulatory authorities, the share of FI and PS assets (f and p respectively) are
determined endogenously in the model.
The cash reserve share is given by c in Figure 1. The strategy of the
bank is to keep the withdrawal share (w) lower than the cash reserve share (c).
As the figure shows, for a given c, the maximum risk that the bank can
undertake (and avoid a run) is w. The c curve determines the composition of
FI and PS assets that would yield a risk of w. The bank has to choose a
minimum of f* as share of FI assets to be on the safe side. This large share of
FI assets leaves a small share of PS assets, as p (=1-c-f). Though this result
appears discouraging for the use PS modes of financing, it gives one
explanation as to why Islamic banks are not using these instruments in practice.
Note that the above results are produced on the assumption that Islamic banks
provided liquidity services only. We discuss the role of an Islamic bank as an
investment intermediary next.

26
 Figure 1: Islamic Bank as a Liquidity Provider

w()

f()
w a b

w0 c f* 1 w, i

4.2. Islamic Bank as an Investment Intermediary

It has been argued that given the profit-sharing features, Islamic banks
also act as investment intermediaries. This role of Islamic banks is reflected by
their offering (restricted) investment accounts that offer higher expected rate of
return with higher risks. Relatively less risk-averse behavior implies that the
tolerance factor (m) for these deposits would be higher. That will reduce the
withdrawal risks or asset-preservation withdrawals from investment deposits.
Furthermore, these deposits are less liquid as there is a cost of withdrawal of
funds. This, along with the higher tolerance factor would mean that the w()
curve will be steeper for funds in investment accounts. This is shown in Figure
2 below by the rotation of the w() curve from w0() to w1( ), increasing the
acceptable risk that bank can undertake on the asset side from 0w to 1w. As the
figure shows, we move from point b to d on the f() line. This movement
indicates that the bank can lower its share of FI asset from *f1 to *f2 and
increase the share of PS assets in their portfolio. Note that the model shows that
bank will not invest all of its funds in PS assets, but will have a combination of
PS and FI assets in the portfolio.
27
 The above analysis shows that when an Islamic bank acts as an
investment intermediary attracting depositors who are willing to take more
risks, the bank is in a position to increase the share of assets financed by PS
modes. Given this result, a question arises as to why Islamic banks are not
investing in PS assets, particularly with funds from investment deposits. One
answer to this puzzle may lie in the other explanations given in the literature, in
particular the asymmetric information problems that occur more prominently in
PS modes of financing. 26 These problems increase the risks on PS instruments,
inhibiting their use. Another possibility of the dominance of FI instruments in
the assets of Islamic banks, however, may be that they have been operating
more like liquidity providers and less like investment intermediaries. This may
stem from the hangover from conventional banking practices that most of the
managers of Islamic banks are trained in and come from.
Figure 2: Islamic Bank as an Investment Intermediary

w1()
w0()

0f
1w c d
0w a b

w0 c *f2 *f1 1 w, i

5. Some Performance and Operational Matters of Islamic Banks

26
Boyd, et.al (1998) identify two endogenous sources of moral hazard in financial
contracts. One arises when the entrepreneur undertakes an inferior project after
funds are provided and the other is costly state verification problem arising from the
inability of the financier to observe the actual outcome. The former endogenous
moral hazard can also be interpreted as an entrepreneur choosing an effort level that
is lower than what is expected ex-ante (Santos 1999).
28
 The distinguishing PS features of Islamic banks point out to their
investment intermediary nature. This not only has operational and management
ramifications, but also affects the performance of Islamic banks compared to
their conventional counterparts. We discuss below some implications for
Islamic banks in the light of the model given above. 27

5.1. Stability of Islamic Banks

It has been shown above that Islamic banks can use both FI and PS
modes on the assets side. The use of profit-sharing principle to reward
depositors on the liability side, however, has important implications regarding
the stability and management of banks. To understand this, we discuss how the
depositors will react to a lower rate of return relative to other financial
institutions in the economy. When we compare the rate of return with other
financial institutions, we observe that sharing profit with the depositors not only
brings more stability to the bank, but also acts as a disciplinary tool that
punishes inefficiencies.
If the return of the bank is low due to an overall downturn of the
economy, then this will be reflected in lower returns for all banks. Whereas
lower returns on the asset side will be transferred to the depositors in Islamic
banks, this will not be the case in case of conventional banks. Linking the return
on liabilities to return on assets, brings stability and can avoid bank runs
(Gangopadhyay and Singh 2000 and Jacklin 1987). This has been shown by
Khan (1987) for Islamic banks. A downturn causing a decline in the value of
assets is cushioned by the corresponding decrease in the liabilities, protecting
the net-worth of the bank. When depositors see that the overall rate of return in
the economy has gone down, the probability of withdrawals will be very small.
If the lower rate of return, however, is restricted to a particular bank then
the reaction of depositors will be different. An isolated lower rate of return of
the bank will be taken as a signal of mismanagement and cause asset-
preservation withdrawals. Thus, the use of PS modes has disciplinary effect on
the banks as inefficiency and mismanagement are punished by depositors.

5.2. Liquidity Management

27
While there are many issues that can be discussed, the points discussed here are
those that relate to the model.
29
 The promise of banks to provide liquidity to the depositors makes
liquidity management an integral part of banking practice. Evolution of liquidity
management strategies in conventional banking illustrates the changing
characteristics and development of instruments over time. 28 Before 1930s, the
commercial loan theory maintained that banks focus on short-term self-
liquidating loans that match the maturity of the deposits. A bank was considered
liquid if the loan portfolio was entirely short-term. The shiftability theory
asserted that any liquid asset could be used to meet deposit withdrawals. The
growth of treasury bonds constituted the bulk of the liquid assets that could be
sold in the secondary markets prior to maturity. During the early 1950s, the
anticipated income theory tied loan payments to a borrowers expected income.
To meet liquidity requirements, loans were structured in a way so that the
timing of payments of the principal and interest matched borrowers' cash flow
and ability to pay. More recently, the liability side of the bank is also used under
the liability management theory under which banks borrow from the money
market and capital markets to satisfy liquidity needs. Thus, both sides of the
balance sheet are now used to manage liquidity of banks.
The liquidity management strategies outlined above indicate that Islamic
banks face constraints in using some of them. The shiftability theory does not
apply to Islamic banks, as there are limited Islamic financial assets that can be
bought and sold. Similarly, Islamic banks are constrained in using liability
management, as there is no Islamic money market from where they can borrow
funds. 29 They can, however, use the anticipated income theory approach by
carefully matching the withdrawals to expected income. The commercial loan
theory to manage probably explains liquidity management of Islamic banks the
best. This is reflected in Islamic banks concentration of short-term fixed-income
assets. As pointed out, primary PS assets are less liquid. This may be another
reason why Islamic banks have not used these instruments in their portfolios.
One way to resolve the liquidity problem is to invest a part of the funds in
secondary PS assets that are more liquid.

5.3. Risk Management Strategies

28
The historical development discussed here is taken from Koch (1995, p. 481).
29
Due to lack of financial instruments, most Islamic banks have problems with excess
liquidity. For a discussion on managing excess liquidity, see Ahmed (1997).
30
 It has been shown above that Islamic banks face financial risks in using
demand deposits, and business risks in using deposits that share profit/loss. The
variability of returns also introduces withdrawal risks. This calls for different
risk management strategies. Some specific schemes that can be taken up by
banks are outlined below.

5.3.1. Stabilization of Rate of Return on Deposits

It has been noted that the withdrawal risk increases in Islamic banks if
they are not able to give depositors a rate of return comparable to the market.
To stabilize the returns of Islamic banks with that of their conventional
counterpart, the former institutions can opt for following two options. First,
create reserve funds from good performing periods to compensate depositors on
bad performing years. The reserves may be used to cover either losses or
equalize rate of return with the market. An example of the first kind is Jordan
Islamic Banks scheme of loss prevention reserves. A reserve is built by
deducting a percentage of the profit of the bank to cover losses in non-
performing periods. Similarly, Khan and Ahmed (2001) report that around forty
percent of 17 Islamic financial institutions surveyed maintain rate of return
equalizing reserves.
The second way to stabilize the rate of return is that the banks can use
profit-sharing ratio in a staggered manner to reduce the variability of returns on
deposits. When the net-income of the bank is smaller than expected, the profit
share of depositors can be higher to keep the rate of return on deposits relatively
stable. Note that this cannot be done ex-post, as an Islamic contract requires
fixation of profit-sharing ratio ahead of the transaction. The banks can,
however, work out a profit-sharing ratio ex-ante that depends on the profit of
the bank. As the profitability of the bank increases, the share going towards the
bank increases. Such a formula would not only make the return on deposits
relatively more stable, but also has an incentive structure for the bank to manage
its operations efficiently to maximize the profit. Note that this scheme of
weakening the link between the variability of assets and deposits decreases the
withdrawal risk, but increases the displaced commercial risk.

5.3.2. Asset Management: Combining PS with FI Mode

Instead of using either FI or PS mode to finance individual projects,
Islamic banks can combine the two. Specifically, while the FI mode can be used

31
to finance fixed capital and real assets, funds can be made available on PS basis
for working capital. To be acceptable from an Islamic point of view, however,
FI and PS contracts cannot be conditional and must be independent. By
combining FI and PS instruments in different ratios, a variety of risk-return
profiles can be achieved for different projects. The share of FI and PS modes
can vary with the project's risks. As the risk of a project increases, the share of
FI mode of financing (f) will become larger. In projects where uncertainty
and risk are relatively small, Islamic banks can opt for a larger share of PS
mode (p).
Combining the FI and PS modes to finance a project brings Islamic
banks closer to universal banking. Other than reducing the risks, universal
banking may have other advantages. By holding shares, banks will have a say in
the management of the firm. In the Islamic banking framework, this implies that
the mode of financing will take the form of musharakah rather than mudarabah.
Santos (1999) maintains that a mix of debt and equity holdings in a firm can
reduce the moral hazard problem. 30 This strategy would reduce the overall
business risks associated with PS instruments.

5.3.3. Bank Capital Allocation

Regulators require banks to hold capital to protect them from, among
others, credit risk that banks face. New proposals from Basel Committee for
Banking Supervision highlight capital requirement that is linked to the risks
associated with assets. 31 Bank capital acts as a buffer for any negative net-
worth in banking. While loss of an Islamic bank is shared by depositors, too
much variability in the returns may increase the withdrawal risk. To reduce the
variability of returns on the liability side, risky PS assets can be financed partly
by bank capital and partly by investment deposits. One way to do this is to use

30
Boyd, et.al. (1998) and Boyd (1999) argue that when deposit insurance exists,
universal banking can increase the moral hazard problem faced by banks. This
conclusion rests on the assumption that by holding equity position in firms (to whom
loans are also advanced) the incentives of banks and firms align and the 'perks'
gained by moral hazard behavior is also shared by the bank. Diamond (1998),
however, opines that the conclusions of the model are not logical. For a discussion
of the benefits and costs of commercial and universal banking see Santos (1998).
31
For a discussion of regulatory capital for banks, see Khan and Ahmed (2001, Chapter
4).
32
the principle of risk-adjusted rate of return on capital (RAROC). 32 Under this
method, first the risk associated with different assets needs to be estimated. The
standard deviation of the return on specific assets serves as proxies for risk. To
reduce the risk on deposits, equity capital is allocated to cover the expected loss
of investment. Higher the risk and expected loss, higher is the equity allocated
to the project. By doing this, a part of the risk and loss (if it occurs) is absorbed
by bank capital. This will reduce the variance of the return on deposits.
Using bank capital in risky PS investments has another advantage. As
the bank capital is used in the project, the moral hazard problem that exists in a
principal-agent problem in banking relationships is reduced. With its own
capital involved, the bank will be more careful in choosing the projects to invest
(reducing the adverse selection problem) and monitor the activities of the firm
(to minimize the moral hazard problem).

5.4. Quasi-Equity Deposits and Securitization

As mentioned earlier, people have different financial needs. They keep a
part of savings in liquid form for current and unexpected expenditures and
invest the remaining part in assets with different risk-return and maturity
profiles. People use banks mainly for their liquidity needs. It has been shown
that PS modes of financing are riskier with potentials of yielding higher returns.
While investment deposits may serve only a part of investment need, there is a
need to offer other forms of investment facilities to customers to enable people
to park funds for longer time periods in riskier assets with higher expected rates
of return. Attracting such funds will change the nature of liability and make
withdrawals less sensitive to risk.
Islamic banks can offer a variety of investment services to tap funds
from the publics preferences of different risk-return features. One possible
option is to have mutual funds, where the bank invests funds in securities. It
should be noted that some Islamic banks and conventional banks are already
offering these products. 33 Mutual funds, however, will change the nature of

32
For a discussion on RAROC, see Crouhy, et. al (2000, pp. 543-48).
33
Examples of Islamic banks offering mutual funds are Abu Dhabi Islamic Bank, UAE;
Al Rajhi Banking and Investment Bank, Saudi Arabia; Bank Islam Malaysia,
Malaysia; and Al Baraka Investment Bank, Bahrain. Conventional banks that offer
Islamic mutual funds include National Commercial Bank, Saudi Arabia; National
33
operations of banks to asset-management. Another possibility for banks is to
raise funds by issuing securities for specific projects and fixed period of time. A
way to distinguish between equity and such securities is to identify them as
primary and secondary capital respectively. 34 As these quasi-equity securities
represent shares in the projects, they can be sold in the secondary markets.
While dividends are paid periodically, the capital can be redeemed at the end of
the maturity of the security representing the end of the project life. Note that
while funds from mutuals can be withdrawn, the funds from secondary capital
remain intact with banks until the maturity date of the security.

6. Conclusion
Financial intermediaries specialize in managing risks. Different financial
institutions can be identified by the nature of risks they deal with and manage.
Conventional banks intermediate funds from depositors to entrepreneurs on
interest. The risks associated in financing are managed, among other things,
with collateral against which loans are sanctioned. Using profit sharing
principle on the liability and asset sides is a unique feature of Islamic banking.
The use of this mode introduces certain risks not found in conventional banks.
The paper incorporates this feature in a microeconomic model to explain the
composition of assets of an Islamic bank. Specifically, when depositors'
preferences towards liquidity is considered, the paper shows that adopting PS
modes on the asset side introduces risks that are not compatible with the
liability side. This may explain the extensive use of relatively low-risk fixed-
income instruments by Islamic banks in practice. The paper shows that when
the bank acts as an investment intermediary it can increase the share of PS
assets in its portfolio.
The paper points out some implications relating to performance and
management of Islamic banks. While using PS modes on the liability side to
reward depositors makes Islamic banks relatively more stable during economic
downturns, it acts as a disciplinary tool for inefficient banks. Islamic banks,
however, have limitations in using liquidity management strategies mainly due

Bank of Kuwait, Kuwait; Arab National Bank, Saudi Arabia; and HSBC Bank,
USA.
34
These securities appear similar to limited-life (preferred) stock that is considered
secondary capital in conventional banking literature. For a discussion on primary
and secondary capital see Rose and Fraser (1988, pp. 396-97) .
34
to lack of Islamic money market and instruments. The risks arising from PS
modes make the nature of Islamic banks different from conventional banks and
calls for different strategies to manage these institutions. The paper provides
some schemes that Islamic banks can adopt to reduce these risks. Creating
reserve funds for use in times of low performance and staggered PS ratio can be
used to de-link the returns on deposits from those of assets and make Islamic
banks relatively more stable. On the asset side, risks associated with PS modes
of financing can be lowered by blending FI and PS instruments. On the liability
side combining equity and deposits to finance assets using PS modes will also
reduce risks on returns on deposits. Furthermore, the liability side of the bank
can be made more diversified to satisfy different risk/return/maturity features of
the public. Possible products that can be offered include different types of
quasi-equities and mutuals along with investment deposits. The bank will be
able to invest these funds in PS assets without increasing withdrawal risk.
The paper shows that the nature of Islamic banks is different from that of
conventional banks. Using PS modes not only implies that Islamic banks should
have a balance sheet structure different from the conventional banks, but it also
indicates that Islamic banks need to develop different strategies to run the
institutions. In particular, PS modes of financing introduce unique risks that call
for a new approach to manage them. Given this difference, it will be difficult to
enhance PS modes of financing in Islamic banks if they operate within the
framework of conventional banks (as liquidity providers) and follow their
operations. While this paper provides some insights into the problems faced by
Islamic banks, there is a need to conduct further research to understand the
nature of risks these institutions face and to provide appropriate strategies for
dealing with them.

Acknowledgements
The paper has benefited from comments and suggestions of many
people. I gratefully appreciate comments of Ausaf Ahmad, Mabid A. Al-Jarhi,
Usman Babikir, Boualem Bendjilali, Hussain Fahmy, Munawar Iqbal, and M.
Fahim Khan on the first draft of the paper. I would like to thank M. Umer
Chapra and Tariqullah Khan for their thoughtful suggestions that helped in
revising the paper. Finally, I am grateful to two anonymous referees for their
valuable comments. The usual caveat, however, applies.

35
 References

AAOIFI-Accounting and Auditing Organization for Islamic Financial

Institutions (1999), Statement on the Purpose and Calculation of the
Capital Adequacy Ratio for Islamic Banks, AAOIFI,Bahrain.
Aggarwal, Rajesh K. and Tarik Yousef (2000), "Islamic Banks and Investment
Financing," Journal of Money, Credit, and Banking, 32, 93-120.
Ahmad, Ausaf (1993), Contemporary Practices of Islamic Financing
Techniques, Research Paper No. 20, Islamic Research and Training
Institute, Islamic Development Bank, Jeddah.
Ahmad, Ausaf (1997), Structure of Deposits in Selected Islamic Banks:
Implications for Deposit Mobilization, Research Paper No. 48, Islamic
Research and Training Institute, Islamic Development Bank, Jeddah.
Ahmed, Habib (2001), Determinants of Profit-Sharing Ratio in Project
Financing: A Note, Islamic Economic Studies, 9,42-45.
Ahmed, Osman Babikir (1997), Islamic Financial Instruments to Manage
Short-term Excess Liquidity, Research Paper No. 41, Islamic Research
and Training Institute, Islamic Development Bank, Jeddah.
Al-Jarhi, Mabid Ali (1999), "Islamic Economics from a Vantage Point,"
Keynote Lecture, International Conference on Islamic Economics
towards the 21st Century, Kuala Lumpur.
Al-Jarhi, Mabid Ali (2000), "Transactions in Conventional and Islamic
Economies: A Comparison," paper presented at the Fourth Harvard
University Forum on Islamic Finance, Oct. 31-Nov.1, 2000.
Bendjilali, Boualem and Tariqullah Khan (1995), Economics of Diminishing
Musharakah, Research Paper No. 31, Islamic Research and Training
Institute, Islamic Development Bank, Jeddah.
Bhattacharya, S and A. Thakoor (1993), "Contemporary Banking Theory,"
Journal of Financial Intermediation, 3, 2-50.
Bhattacharya, Sudipto, Arnoud W.W. Boot, and Anjan V. Thakor (1998), "The
Economics of Bank Regulation," Journal of Money, Credit, and Banking,
30, 745-770.
Boyd, John H. (1999), "Expansion on Commercial Banking Powersor,
Universal Banking is the Cart, not the Horse," Journal of Banking and
Finance, 23, 655-662.

36
Boyd, John H., Chun Chang, and Bruce D. Smith (1998), "Moral Hazard under
Commercial and Universal Banking," Journal of Money, Credit and
Banking, 30, 426-68.
Chapra, M. Umer (1985), Towards a Just Monetary System, The Islamic
Foundation, Leicester.
Chapra, M. Umer and Tariqullah Khan (2000), Regulation and Supervision of
Islamic Banks, Occasional Paper No. 3, Islamic Research and Training
Institute, Islamic Development Bank, Jeddah.
Crouhy, Michel, Dan Galai, and Robert Mark (2000), Risk Management,
McGraw Hill, New York.
Diamond, Douglas W. (1984), "Financial Intermediation and Delegated
Monitoring," Review of Economic Studies, 51, 393-414.
Diamond, Douglas W. (1998), "Comment on Moral Hazard under Commercial
and Universal Banking," Journal of Money, Credit and Banking, 30,
469-71.
Diamond, Douglas W. and Philip H. Dybvig (1983), "Bank Runs, Deposit
Insurance, and Liquidity," Journal of Political Economy, 91, 401-19.
Fahmy, Hussein Kamal (2001), Joint Murabahah in Islamic Financial
Institutions paper presented at the OIC Fiqh Academy Conference, 22
December, 2001, Kuwait.
Freixas, Xavier and Jean-Charles Rochet (1998), Microeconomics of Banking,
MIT Press, Cambridge.
Gangopadhyay, Shubhashis and Gurbachan Singh (2000), "Avoiding Bank
Runs in Transition Economies: The Role of Risk Neutral Capital,"
Journal of Banking and Finance, 24, 625-42.
Greene, William H. (1997), Econometric Analysis, Prentice Hall International
Inc., Upper Saddle River, NJ.
Heffernan, Shelagh (1996), Modern Banking in Theory and Practice, John
Wiley and Sons, Chichester.
Iqbal, Munawar, Ausaf Ahmad, and Tariqullah Khan (1998), Challenges facing
Islamic Banking, Occasional Paper No. 1, Islamic Research and Training
Institute, Islamic Development Bank, Jeddah.
Jacklin, C.J. (1987), "Demand Deposits, Trading Restrictions, and Risk
Sharing," in E. Prescott and N. Wallace (Eds.), Contractual

37
 Arrangements for Intertemporal Trade, University of Minnesota Press,
Minneapolis.
Kahf, Monzer and Tariqullah Khan (1992), Principles of Islamic Financing,
Research Paper No. 16, Islamic Research and Training Institute, Islamic
Development Bank, Jeddah.
Khan, M. Fahim (1991), Comparative Economics of Some Islamic Financing
Techniques, Research Paper No. 12, Islamic Research and Training
Institute, Islamic Development Bank, Jeddah.
Khan, M. Fahim (1995a), Essays in Islamic Economics, The Islamic
Foundation, Leicester.
Khan, Mohsin S. (1987), "Islamic Interest-free Banking: A Theoretical
Analysis," in Mohsin S. Khan and Abbas Mirakhor (Eds.), Theoretical
Studies in Islamic Banking and Finance, The Institute for Research and
Islamic Studies, Houston.
Khan, Tariqullah (1995b), "Demand for and Supply of Mark-up and PLS Funds
in Islamic Banking: Some Alternative Explanations," Islamic Economic
Studies, 3, 39-78.
Khan, Tariqullah and Habib Ahmed (2001), Risk Management: An Analysis of
Issues in Islamic Financial Industry, Occasional Paper No. 5, Islamic
Research and Training Institute, Islamic Development Bank, Jeddah.
Koch, Timothy (1995), Bank Management, The Dryden Press, Orlando.
Kocherlakota, Narayana R. (1996), "The Equity Premium: It's Still a Puzzle,"
Journal of Economic Literature, 34, 42-71.
Mehra, R. and E. Prescott (1985), "The Equity Premium: A Puzzle," Journal of
Monetary Economics, Vol. 15, pp. 145-61.
Mirakhor, Abbas (1987), Short-term Asset Concentration and Islamic
Banking, in Mohsin Khan and Abbas Mirakhor (Editors), Theoretical
Studies in Islamic Banking and Finance, The Institute for Research and
Islamic Studies, Houston.
Oldfield, George S. and Anthony M. Santomero (1997), "Risk Management in
Financial Institutions," Sloan Management Review, 39.
Ramakrishnan, R.T.S. and A.V. Thakor (1984), "Information Reliability and a
Theory of Financial Intermediation," Review of Economic Studies, 51,
415-32.

38
Reilly, Frank K. and Keith C. Brown (1997), Investment Analysis and Portfolio
Management, The Dryden Press, Forth Worth, TX.
Rose, Peter S. and Donald R. Fraser (1988), Financial Institutions:
Understanding and Managing Financial Services, Third Edition,
Business Publications, Plano, Texas.
Santomero, Anthony M. (1997), "Commercial Bank Risk Management: An
Analysis of the Process," Journal of Financial Services Research, 12,
83-112.
Santos, Joao A.C. (1998), "Banking and Commerce: How does United States
Compare to Other Countries, Federal Reserve Bank of Cleveland
Economic Review, 34, (Quarter 4).
Santos, Joao A.C. (1999), "Bank Capital and Equity Investment Regulations,"
Journal of Banking and Finance, 23, 1095-1120.
Scholtens, Bert and Dick van Wensveen (2000), "A Critique on the Theory of
Financial Intermediation," Journal of Banking and Finance, 24, 1243-51.
Siddiqi, M. N. (1981), Banking without Interest, Islamic Publications Ltd.,
Lahore.
Siddiqi, M. N. (1983), Issues in Islamic Banking, The Islamic Foundation,
Leicester.
Siddiqi, M. N. (1991), "Some Economic Aspects of Mudarabah," Review of
Islamic Economics, 1 (2), 21-33.
Wallace, Neil (1996), "Narrow Banking Meets the Diamond-Dybvig Model,"
Federal Reserve Bank of Minneapolis Quarterly Review, 20, 3-13.

39