Pacific-Basin Finance Journal 34 (2015) 253–272

Contents lists available at ScienceDirect

Pacific-Basin Finance Journal

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

Can Islamic banking ever become Islamic?

Saad Azmat a,1, Michael Skully b,2, Kym Brown b,⁎
a
Suleman Dawood School of Business, Lahore University of Management Sciences, D.H.A., Lahore Cantt, 54792, Pakistan
b
Department of Accounting and Finance, Monash University, P.O. Box 197, Caulfield East, Victoria 3145, Australia

a r t i c l e i n f o a b s t r a c t

Article history: This paper attempts to explain the dominance of asset side debt
Received 1 August 2014 contracts in Islamic banks, even though many consider alternative
Accepted 8 March 2015 Islamic joint venture (IJV) contracts to be the ideal Islamic financing
Available online 14 March 2015
mode. Theoretical models based on asymmetric information are used
to argue that adverse selection and moral hazard alone cannot explain
JEL classification:
this phenomenon. The model is augmented with risk averse depositors
G21
to show that the emergence of asset side IJV could be deterred by Islamic
G24
banks' liability side. This paper suggests that for IJV, affiliated venture
Keywords: capital and private equity might prove more successful institutions
Islamic banks than banking.
Moral hazard Crown Copyright © 2015 Published by Elsevier B.V. All rights reserved.
Islamic joint venture finance

1. Introduction

The concept of risk sharing is a key feature that distinguished Islamic from conventional banking. Islamic
joint venture (IJV) finance3 is perhaps the best example, but Islamic banks (IBs) in practice seem to prefer
more conventional looking debt contracts. Several authors (Khan, 2010; Chong and Liu, 2009) have criticised
this practice and suggested change. Our study adds to this literature by developing a model to help explain
this position. We propose that risk averse depositors drive this Islamic bank preference for an intermediation
model whose risk reward framework causes debt rather than IJV to become the preferred choice when
financing borrowers. We also contribute to the asymmetric information and costly state verification literature
(Leland and Pyle, 1977; Diamond, 1984; Townsend, 1979; Gale and Hellwig, 1985; Ueda, 2004) by suggesting

⁎ Corresponding author. Tel.: +61 3 9903 1053; fax: +61 3 9903 2422.
E-mail addresses: saad.azmat@lums.edu.pk (S. Azmat), Michael.Skully@monash.edu (M. Skully), Kym.Brown@monash.edu (K. Brown).
1
Tel.: +92 42 111 11 5867.
2
Tel.: +61 3 9903 2407; fax: +61 3 9903 2422.
3
IJVs are equity style products based on partnerships, where the creditor is a partner in the venture. This entitles the creditor to a
pre-agreed percentage of the profit, but also to share any loss according to their investment ratio.

http://dx.doi.org/10.1016/j.pacfin.2015.03.001
0927-538X/Crown Copyright © 2015 Published by Elsevier B.V. All rights reserved.
254 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272

that asymmetric information alone cannot explain the lack of joint venture contracts on the asset side of
Islamic banks. The nature of risk averse bank depositors must also be considered when explaining why Islamic
banks do not invest in VC type technology to neutralise asymmetric information.
Islamic finance has grown rapidly and expects to reach $2 trillion in assets by 2015 (EY, 2014). To support
this initial growth, Islamic banks hired conventional bankers as well as applied conventional banking models
across their operations. Islamic banks, currently, use a traditional banking model taking deposits on the
liability side and making predominantly debt contracts on their asset side. As their customers had mainly
dealt with conventional banks, some Islamic banks look all too similar to conventional ones. Supporting
this stance, Beck et al. (2013) concluded that Islamic banks' business models may not be dissimilar to those
of conventional banks.4 Likewise some instrument techniques have been applied with only a modest
consideration of true risk and reward positions that Islamic finance affords (Khan, 2010). Building on the
context developed by Chong and Liu (2009) and Khan (2010) we consider if Islamic banks can ever offer
more profit and loss sharing (PLS) on their asset side, rather than the current predominance of debt styled
contracts, and thus become more ‘Islamic’.
In a Miller and Modigliani world of perfect information, specialised financial institutions, such as banks
and venture capitalists, with their abilities to neutralise problems of asymmetric information would not
exist (Akerlof, 1970; Leland and Pyle, 1977). In such a world the question of whether an Islamic bank should
use Islamic joint venture (IJV)5 or debt financing to fund its borrowers would be unimportant. Unfortunately,
such a world is far from reality. Given the presence of asymmetric information, Islamic banks must select a
mode of finance that will maximise their return while minimising risk. Asymmetric information could be
one reason therefore why debt based Islamic products have come to dominate the Islamic bank assets,
even though Islamic scholars consider Islamic joint ventures as the ideal religious and ethical alternative to
debt based conventional financing (Usmani, 2002; Ayub, 2007).
Asymmetric information results in moral hazard and adverse selection problems. These in turn may affect
the investment preferences of financial institutions. Private equity and joint venture products, which have
much in common with IJV, are more susceptible to moral hazards and adverse selection problems than
traditional debt. Venture capitalists (hereafter VCs), through better screening, contracting and monitoring,
can more effectively neutralise these problems than banks (Kaplan and Stromberg, 2001). This allows them
to offer the joint venture type products which banks tend to avoid (Landier, 2001; Ueda, 2004). The question
that arises, however, is that if profits are sufficiently large, banks should specialise in VC type funding to
increase their profitability. A possible explanation stems from differences on their liability side. VCs raise
their funds for around a 7 year commitment and so can avoid withdrawal risk (Gompers and Lerner, 1999).
In contrast, a banks' reliance on short-term deposits exposes them to considerable liquidity risk (Diamond
and Dyvbig, 1983). This causes them to prefer debt instruments which can be liquidated on the asset side.
Unlike conventional banks, the withdrawal risk for Islamic banks should be less of an issue to the
extent that their liability side is joint venture funded.6 From an idealistic viewpoint, Islamic customers
should be considered like equity fund investors where the Islamic bank has no obligation to redeem or
guarantee the principal (Usmani, 2002). This means that Islamic banks should be able to pass on any
losses to their customers as well as refuse withdrawals. In such a world, withdrawal risk, would not
fully explain why VC type financing has failed to dominate Islamic banking. In practice of course, customer
pressure and competition may force Islamic banks to be reluctant to enforce these powers fully.

4
In Malaysia the Islamic Financial Services Act 2013 has enabled an Islamic bank to run from its inception to its winding up under
Shariah compliance. This should support such banks to use Islamic conventions and not be required to comply with conventional bank
rulings. Part IV of Division 28 of the Act requires Islamic banks to be fully Shariah compliant.
5
IJVs are equity style products based on partnerships, where the creditor is a partner in the venture. This entitles the creditor to a pre-
agreed percentage of the profit, but also to share any loss according to their investment ratio.
6
While some might suggest that Islamic banks have a captive clientele and so a rather limited interest rate or displaced commercial risk
compared to conventional banks, this might not be the case. Firstly, Islamic banks compete with each other and so poor returns in one
Islamic bank could cause clients to shift to a better performing Islamic bank. Secondly, Islamic banks can have non-Muslim clients (once
an important source of Islamic bank funding in Malaysia) that would have no problems in moving funds to a conventional bank. Finally,
while religion might seem to afford strong client retention within the industry, this has not been the case in Indonesia where despite the
predominance of Muslims, Kasri and Kassim (2009) found that depositors changed from Islamic to conventional banks when the latter
offered more attractive returns.
 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272 255

One explanation not considered in the prior literature is the risk aversion amongst bank depositors,
both conventional and Islamic. Islamic bank depositors with risk averse utility functions would have
a high preference for a stable return on their principal. Hence, they would seek Islamic banks
whose asset side is dominated by fixed income instruments. Moreover, before investing with an IJV
dominated Islamic bank, they would demand a sufficiently high risk premium. This in turn, leaves
the bank with little incentive to utilise IJVs. This gap in the literature motivates the following research
question: Can the presence of asymmetric information and risk averse customers jointly explain Islamic
banks' lack of IJV financing?
In the spirit of Leland and Pyle (1977), Diamond (1984) and Ueda (2004), this paper develops a firm
focused model. This implies that the choice of an IJV asset financing rests with the borrowing firm. The
Islamic bank in our model offers both debt and IJV financing as long as they provide a positive net present
value (Bester, 1985). The firm's decision to choose IJV or debt financing, however, depends on which
maximises its own profit. Our model shows that in the presence of perfect information, a borrowing
firm would be indifferent between debt and IJV contracts. Asymmetric information is then introduced
by assuming that the bank can observe signals regarding the firm's quality and so infer its true nature
with a certain probability (Ueda, 2004). Here, the model shows that high risk firms would prefer
IJV whereas low risk firms would prefer debt contracts. Moral hazard concerns, emanating from
misreporting and cheating, are then addressed. Our model shows that moral hazard is not as severe as
adverse selection, since it can be neutralised using long-term relationships or punishment strategies.
To explain why banks do not adopt VC type screening methods to neutralise asymmetric information,
the risk averse bank depositors are added. When the model considers the depositors' risk aversion,
firms prefer debt to IJV.
This paper contributes to the conventional and Islamic finance literature by developing a theoretical
model to explain the behaviour of Islamic bank clients and thereby the Islamic banks' response. We then
use this to confirm that choice of Islamic bank instruments is simply a function of the risk and return
preferences of firstly their borrowers and then secondly those of its depositors. Besides providing
evidence as to what is behind the Islamic banks current practices, it also adds to the asymmetric information
and costly state verification literature.
The rest of the paper is organised as follows. Section 2 discusses the unique features of Islamic joint
venture (IJV) in Islamic banks. Section 3 provides the literature review while Section 4 presents the theoretical
model and its proofs and Section 5 concludes the paper.

2. Islamic joint venture

Islamic finance differs from its conventional counterparts due to the Islamic law's prohibitions relating to
interest (riba), uncertainty (gharar) and gambling (maysir) (Ayub, 2007). Islamic banks, therefore, have
developed alternatives accordingly. These include interest free loans (Qard Hasana), Islamic joint ventures
(Musharakah/Mudarabah) and trade or lease based financing (Murabaha/Ijarah/Salam). Of all these Islamic
scholars prefer Islamic joint venture (Musharakah/Mudarabah) as a non-debt based mode of financing.
Although interest free debt is permissible in Islam, it is still viewed pejoratively7 (Usmani, 2002). In contrast,
joint ventures are considered blessed (Iqbal et al., 1998). Moreover, Islamic joint venture financing is also
believed to produce a more equitable distribution of wealth (Usmani, 2002). Despite the Islamic attractiveness,
Islamic banks have made negligible use of it on the asset side of their balance sheet (Khan, 2010). In contrast
IJV dominate Islamic bank funding liabilities.
IJVs are equity style products based on partnerships, where the creditor instead of lending money and
charging interest becomes a partner in the venture. This entitles them to a pre-agreed percentage of the profit,
but also to share any loss according to their investment ratio (Usmani, 2002). The two modes of Islamic joint
venture financing are Musharakah and Mudarabah.

7
Interest free debt can actually be viewed in a positive manner when it is done to help the poor, such as with benevolent lending. In
practice, it can also be used for other purposes. An Islamic pawn broker, for example, gives an interest free loan to the borrower but then
charges for storing the pawned items and insuring it. These fees are surprisingly not unlike those charged for a conventional bank loan.
256 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272

In Musharakah all parties contribute some part of the total capital. The profit ratio in Musharkah is decided
with mutual consent but the loss must be equal to each party's investment proportion.8 It allows an Islamic
bank to participate in the project's management.9
In a Mudarabah contract one party provides the capital while the other provides the expertise (Mudarib).
As in Musharkah the profit in Mudarabah can be divided in any pre-agreed ratio. The loss, however, must be
borne only by the capital provider and not the party providing the expertise (Mudarib)10 (Usmani, 2002).11
Trade or lease based products (Murabaha/Ijarah/Salam/Istisna),12 however, are the most utilised of Islamic
finance.13 Those based on purchase and resale or leasing are the most common (Ayub, 2007) and are offered
where the customer needs to buy a real asset. Instead of lending the customer money to buy the asset, the
bank buys the asset and resells it to the customer at a higher price. This difference provides the bank its return.
Lease based financing similarly involves the bank first buying the real asset and then leasing it to the custom-
er. The primary difference between Islamic finance and conventional financial products is that Islamic banks
own the underlying assets and, hence, bear their risk of loss (Usmani, 2002).
From a theoretical perspective, on the liability side of an Islamic bank's balance sheet, the prohibition of
interest implies that an Islamic bank cannot guarantee a deposit on which it offers a return.14 This means
their deposit funding is confined to the Islamic equivalent of a conventional bank's current account with no
risk but no reward either (Usmani, 2002). An Islamic bank's main funding is therefore not from conventional
savings or term deposits of a commercial bank but rather via a joint venture style investment account (Ayub,
2007). In a typical IJV relationship, the Islamic bank and its customers share profit according to a pre-agreed
ratio but bear any loss according to their investment proportion (Usmani, 2002). This position has two impli-
cations for Islamic banks. Firstly, since the bank neither guarantees the return nor the principal, it has the right
to refuse withdrawals. Secondly, in bankruptcy the customers will be ranked equally with shareholders and
will have to bear any loss according to their investment.15 It should be noted that the results of our model
would hold true, even in the presence of deposit guarantee and bank's inability to refuse withdrawal pay-
ments as our propositions are driven by the uncertainty in returns and risk averse nature of depositors. In
practice, of course Islamic banks would be quite conscious of the reputational risks of enforcement of their
legal rights, which influence their actual decision to honour depositor funds.

3. Literature review

This section first examines how banks and venture capitalists seek to neutralise or otherwise overcome
these adverse selection problems and then how these arise within the context of a joint venture. The issue
of moral hazards is then addressed by banks in general, by venture capitalists and finally by Islamic banks.
Finally, we consider the implications of these views for risk averse depositors.

8
For example, if two people start an Islamic joint venture, each contributing 50% of the capital, then the loss has to be shared 50:50, but
the profit can be shared in any proportion, such as 80:20.
9
See for example Appendix B of ‘Guidelines on Musharakah and Mudarabah Contracts for Islamic Banking Institutions’, Bank Negara
Malaysia, BNM/RH/GL/007-9 that illustrates how an Islamic bank can appoint a board representative for a Musharakah contract.
10
In the event of loss, the party contributing its expertise will of course suffer since it will not receive any profit or other kind of
remuneration. The reputation as business people would also be damaged.
11
In Malaysia as of 31 December 2014, the level of Mudharabah contracts outstanding was 0.023% and Musharakah contracts some
6.93% of total financing by Islamic banks (Bank Negara Malaysia statistics, Table 1.18.2).
12
These are the most popular and widely acceptable modes of Islamic financing. However, there are modes other than these, such as
Tawarruq, that are also used in some countries (Ayub, 2007).
13
In Malaysia Murabahah contracts accounted for 26.1% of total financing by Islamic banks as at 31 December 2014 (Bank Negara
Malaysia statistics, Table 1.18.2).
14
While the Islamic bank itself cannot guarantee such deposits, scholars in some countries have allowed deposit guarantees by third
parties. Malaysia, for example, offers a dual deposit insurance under which Islamic bank depositors are protected against Islamic bank
failure under the Malaysian Deposit Insurance Scheme. Islamic bank depositors in Sudan are similarly protected under the Sudan Bank
Deposit Security Fund (Islamic Association of Deposit Insurers, 2010). Countries in the Middle East and Pakistan, however, do not approve
this practice.
15
In theory, this holds true for a Musharakah based IJV. Where the Islamic joint venture is Mudarabah based, in case of bankruptcy all
the loss has to be borne by the IJV account holders.
 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272 257

3.1. Adverse selection issues

Adverse selection occurs when one of the contracting parties has better information than the others about
its risk, especially when it is costly for the less informed party to obtain that information (Akerlof, 1970). In the
context of bank loan contracts, the borrowing firm should know more about their probability of success. This
allows them to determine whether the cost of borrowing is truly representative of their underlying risk.

3.1.1. Banking techniques to neutralise adverse selection

Banks discern valuable information from observing attributes such as high collateral, high owner's capital,
good reputation, high net worth and low leverage. Reliable collateral may help to mitigate the banks'
asymmetric information concerns by ensuring repayment (Bester, 1985; Chan and Thakor, 1987). By
investing a significant proportion of the owner's capital, entrepreneurs also send positive signals about the
project's quality (Leland and Pyle, 1977). Reputational effect may also limit adverse selection problems as
the fear of a reputational blemish might cause a firm to choose less risky projects (Diamond, 1989). High
net worth and low leverage are also a signal of a firm having less risk. Lack of these attributes, however,
can impede the transmission of market signals reflecting the firm's true characteristics (Diamond, 1984,
1989). To compensate for adverse selection risks, banks may either charge high payments or refrain from
funding risky firm. Therefore, new firms from industries more prone to asymmetric information problems,
such as technology, have a better chance of venture capitalist than bank funding (Gompers, 1995;
Hellmann, 1998).

3.1.2. Venture capitalists' methods to overcome adverse selection

Venture capitalists (hereafter VCs), by the virtue of their better screening, contracting and monitoring
skills, can more effectively neutralise moral hazards and adverse selection than banks (Barry, 1994; Kaplan
and Stromberg, 2001). They do this by using rigorous selection methods so that only a small proportion
(around 1%) of all applications is funded (Gompers and Lerner, 1999). Moreover, VCs do not just provide
finance but also attempt to add value through their management and technical assistance (Sahlman, 1990;
Barry, 1994). They develop market and technical expertise from their portfolio investments in related
industries and so can neutralise adverse selection effectively.

3.1.3. Adverse selection in joint ventures

Joint venture financing is more susceptible to adverse selection problems than debt as the financier's
return varies with the firm's performance. The financier16 shares the profit if the firm is successful and the
loss with its failure. Therefore, the financier needs to know the different expected profit and loss making states
and the associated probability of each occurring. This uncertainty exposes the contract to asymmetric
information problems (Gompers, 1995; Ueda, 2004). In contrast, the maximum return on debt is independent
of the firm's high profitability while the loss is bounded by the firm's net assets. The literature on costly state
verification also suggests that debt contracts are preferred over equity ones as debt contracts minimise
monitoring cost (Townsend, 1979; Gale and Hellwig, 1985). As the decision to offer debt requires knowledge
of only the firm's ability to return the principal and the accrued interest, a detailed understanding of the firm's
different profit and loss making states is not a pre-requisite. So, compared to joint ventures, debt contracts are
somewhat less prone to asymmetric information (Ueda, 2004).

3.2. Moral hazard issues

Moral hazard may occur when asymmetric information causes a change in the behaviour of the
contracting parties. In the bank–firm relationship, this stems from the firm's incentive to indulge in excessive
risk taking behaviour. Since firms have limited liability, a default would only entitle the banks to the net
proceeds from the assets. If the principal is not recovered, the banks suffer accordingly. If this risk is
foreseeable, the banks can compensate by charging a premium. Once the loan is approved, however,

16
The funder can be either a bank or VC.
258 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272

the firm can indulge in unanticipated excessive risk taking behaviour causing moral hazard concerns
(Akerlof, 1970; Stiglitz and Weiss, 1981).

3.2.1. Moral hazard in conventional banking

Moral hazard in banking results from excessive risk taking firm behaviour. Banks use various means to
neutralise moral hazards including covenants, lending to reputable borrowers, long-term funding relationships
and threat of litigation (Diamond, 1984, 1989).

3.2.2. Moral hazard in venture capital

The moral hazard problem is much more severe amongst VCs than banks. This is because the VC joint
venture returns depend on firm's actual performance. As its profit is shared with the VC, the firm has an
incentive to under-report it. Moreover, the firm has no obligation to return the principal and VC has to
share any losses. This may motivate excessive risk taking behaviour (Kaplan and Stromberg, 2001). VCs
attempt to overcome such concerns through staged financing, active monitoring and controlling of board
and voting rights (Sahlman, 1990; Gompers, 1995). With staged financing, VCs typically start with small
investments and then provide additional funds subject to specific performance targets. The threat of no
subsequent funding mitigates excessive risk taking and misreporting. Board and voting rights also enable
VCs to monitor their firm and take remedial action (Kaplan and Stromberg, 2001). VCs can also use their
board and voting power to replace poorly performing managers (Sahlman, 1990).

3.2.3. Moral hazard in Islamic banks

Islamic bank's IJV funding also suffers from moral hazard problem resulting from misreporting and
excessive risk taking (Iqbal et al., 1998; Khalil et al., 2002; and Ahmed, 2002). As an Islamic bank shares
in the IJV business profit, this might motivate the entrepreneur to under report its profitability. Similarly
as the Islamic bank also bears part of the downward risk; the entrepreneur might indulge in excessive
risk taking. Similar to conventional banks, Islamic banks might use long-term relationships or litigation
threat to neutralise moral hazard.

3.3. Risk averse bank depositors

While asymmetric information can explain the banks' lack of joint venture type products, it is not the sole
explanation. If profits were sufficiently large, banks would simply adopt VC type monitoring and screening to
increase profitability. This, however, does not appear to happen. Another possible explanation stems from
differences on their liability side. The conventional banks reliance on short-term depositors exposes them
to considerable liquidity and withdrawal risk. The threat of bank runs provides considerable aversion towards
joint ventures whose return is uncertain and long-term. In the presence of these risks, debt instruments have
become Islamic bank's preferred investment choice on the asset side (Ahmed, 2001, 2005). In contrast, VCs
raise their funds with around 7 year's commitment and, hence, are less susceptible to withdrawal risk
(Gompers and Lerner, 1999). Moreover, the high return that the VC investors receive compensates for any
liquidity risk for their funding.
Islamic banks' liabilities differ from conventional banks in that Islamic banks are normally under no
obligation to return the account holder's principal amount (apart from current accounts). An Islamic bank's
main funding is, therefore, not from conventional savings or term deposits but rather from joint venture
styled accounts. Such accounts are protected from withdrawal risk as the bank has the authority to refuse
withdrawals (Usmani, 2002). Hence, this risk does not really explain the lack of IJV products in Islamic
banks. Another explanation of the IJV puzzle, ignored in the literature is their risk averse customer base.
While VCs raise their funds from institutional investors, Islamic banks rely on more risk averse customers
(Ahmed, 2001; Gompers and Lerner, 1999). Their high risk aversion implies that a high risk premium
would be required before investing with Islamic banks providing mainly IJV financing. This is because a
large proportion of IJV assets would mean higher uncertainty for account holders' return. This additional
cost would eventually be paid by the IJV contracting firm funded by the bank. The premium demanded by
risk averse bank customers would further dissuade firm from using IJV.
 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272 259

Table 1
List of variables.

Variable Description

F Total cost of the project

φ Net revenuea
f (1 + R)F or firm's project funding requirement
R Rate of return charged on bank loan
r Proportion of the profit received by the bank in an IJV contract
P Probability that the project would make a profit
π Total profit at equilibrium
πh Profits of firms with high success probability
πl Profits of firms with low success probability
Ph High success probability firm
Pl Low success probability firm
s Performance and risk signals of the borrowing firm
Pr Probability
α Bank observes perfect signal re asymmetric information
n Number of periods
θ Lower result of θ implies misreporting
Ps Probability of the firm's misreporting going unnoticed
A Lump sum fine
ρ Degree of risk aversion
σ2 Variance of depositors' earnings
U Utility
a
This is net revenue after deducting all the costs but before the servicing of debt. It is not called profit to
distinguish it from the equilibrium profit π.

4. Firm focused model

4.1. Model for debt financing firm

The model for debt financing firm encapsulates the relationship between two parties: firm and bank.
The variables used in the models are presented in Table 1.

4.1.1. The role of firm and bank

The role of the firm and the banks is modelled assuming that the firm can be of two types: One with a high
success probability (Ph) and another with a low success probability (Pl) (where Ph N Pl). If the firm succeeds
(with probability P), they generate revenue ‘φ’. The firm's total project funding requirement is ‘f ’ and the
banks' are the only available financing source. If the firm choose debt financing, banks charges them a rate
of return ‘R’. This implies that at maturity the firm repays (1 + R)f to the bank.
A typical firm's expected profit when using debt financing can be calculated by:

π ¼ P ðφ − ð1 þ RÞf Þ þ ð1 − P Þð f Þ ð1Þ

In line with Bester (1985) and Ueda (2004), a perfectly competitive banking sector is assumed. This
implies that the bank will finance all projects with a positive net present value (Bester, 1985).17 Following
Bester (1985), the bank would participate as long as the expected profit is non-negative. The bank's
participation constraint, therefore, is given by Eq. (2).

P ðRÞf þ ð1 − P Þð − f Þ ≥ 0 ð2Þ

1−P
R≥ ð3Þ
P

17
In a firm focus model like the one used in this study this assumption only helps to simplify analysis and does not change the eventual
outcome.
260 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272

Since the constraint would be binding in a competitive market and f N 0 therefore:

1−P 1
R¼ ¼ −1 ð4Þ
P P

Eqs. (2) to (4) suggest that in a perfectly competitive banking sector, the rate of return is inversely propor-
tional to the project's success probability. In other words, it is directly proportional to the project's risk. Eq. (4)
can be substituted in the firm's profit function to generate their equilibrium profit for firms with high (Ph) and
low success probabilities (Pl) that employ debt financing:
debt
πh ¼ P h ðφ − f Þ ð5Þ

debt
πl ¼ P l ðφ − f Þ ð6Þ

The next section analyses the firm's equilibrium IJV financing payoff under perfect information. This is
then compared to the firm's debt financing profits given in Eqs. (5) and (6).

4.1.2. Firm's profit outcome under IJV financing

The model developed here captures two distinguishing features of IJV financing: the pre-agreed profit ratio and
the investment dependent loss ratio.18 In an IJV, the firm has to pay a pre-agreed percentage ‘r’ of their net revenue
‘φ’ to the bank. Therefore, if the firm succeeds with probability P, the bank's revenue and the firm's cost would
be ‘φr’. The investment dependent loss ratio, however, implies that in the event of the firm failing with probability
(1 − P), the bank bears any loss according to its investment ratio.19 In this model the banks are assumed to fund
the full project and hence bear all the losses. It is similar to saying that upon failure the firm has the option to sell
the project to the banks at a price of ‘f ’ (Ueda, 2004). The profit outcome for the firm is given in Panel A, Fig. 1.
The profit function of IJV financing firm with both high and low success probabilities is given in Eqs. (7) and (8).

I JV
πh ¼ P h ðφ − φrÞ þ ð1 − P h Þf ð7Þ

I JV
πl ¼ P l ðφ − φr Þ þ ð1 − P l Þf ð8Þ

In an IJV the bank's expected profit depends on the firm's success probability (Iqbal et al., 1998; Usmani,
2002). This provides banks the incentive to gather additional information about the firm before negotiating
the profit sharing ratio ‘r’. If one assumes that the bank possess perfect information about the firm's success
probability, the bank would offer each firm a profit rate ‘r’ that represents their underlying risk. The profit
outcome for the bank in an IJV contract is given by Panel B, Fig. 1.
The bank's participatory constraint for a firm with a high success probability would therefore be as follows:

P h ðφr − f Þ − ð1 − P h Þ f ≥ 0 ð9Þ

where φr can be reduced to:

f
φr ≥ ð10Þ
Ph

Given that the banking industry is competitive, Eq. (9) has to be binding; otherwise the payoff to the firm
can be increased without violating the constraint. Constraint 4 can be plugged back in to the firm's profit
outcome given by Eq. (7) to yield the following equilibrium results:

I JV
π h ≤ P h ðφ − f Þ ð11Þ

I JV
πh ¼ minðφ; P h ðφ − f ÞÞ ð12Þ

18
In an IJV the profit ratio can be pre-agreed but the loss has to be born according to the investment proportion (Usmani, 2002).
19
For simplicity a Mudarabah contract is considered here, where the entire cost of the project is funded by the Islamic banks. The im-
plications of the results, however, can be extended to a Musharakah contract.
 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272 261

A) Firms Profit
P

Firm

B) Banks Profit
P

Bank

1–P

Fig. 1. Profits in IJV under perfect information.

Since

P h ðφ − f Þ b φ ð13Þ

I JV
πh ¼ P h ðφ − f Þ ð14Þ

Eq. (14) above gives the equilibrium profit for a firm funding with IJV that has a high success probability.
The results can be easily extended to show that the equilibrium profit for low success probability firm
undertaking IJV is given by Eq. (15).

I JV
πl ¼ P l ðφ − f Þ ð15Þ

The profit functions of the IJV financing firm under perfect information given by Eqs. (14) and (15) can be
compared with their debt financing profit to show that the payoffs are equal. Since there is no difference in IJV
and debt financing profits for both firm types, they should be indifferent between choosing either of the
financing modes. This proves Proposition 1:

Proposition 1. In the absence of asymmetric information, a firm should be indifferent between IJV and
debt financing.

4.2. Asymmetric information model

In practice, the initial perfect information assumption is unrealistic as the real world is fraught with
asymmetric information (Akerlof, 1970). Our model in this section is now updated to incorporate asymmetric
information and is then used to compare the firm's equilibrium IJV and debt financing returns.

4.2.1. The impact of asymmetric information on the firm and the bank
Asymmetric information is incorporated in the model by assuming that the bank lacks the knowledge
regarding the firm's success probability (Ueda, 2004). This would not directly impact the firm, however,
as they would already be aware of their own success probabilities.20 Therefore, the firm's profit function
will be similar to the perfect information case as given in Eqs. (13) and (14).

20
Note that there will be an indirect impact on the firm's equilibrium profit because of a change in the banks' participatory constraint.
262 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272

Banks use different screening methods to neutralise asymmetric information (Bester, 1985; Chan and
Thakor, 1987; Diamond, 1984, 1989). These methods are more effective for debt financing than IJVs
(Gompers, 1995; Ueda, 2004). For simplicity it is assumed that the bank can perfectly neutralise asymmetric
information for debt financing.21 This implies that even with asymmetric information, the debt financing
equilibrium profit for the bank and the firm would not differ from that of the perfect information case
shown in Eqs. (5) and (6). For IJV financing, however, the bank's inability to neutralise asymmetric information
would affect their participatory constraint and, hence, the firm's equilibrium IJV financing payoff.

4.2.2. Asymmetric information in IJV financing

Asymmetric information is captured in the IJV financing model by assuming that the bank infers the success
probabilities Ph and Pl of the approaching firm by observing performance and risk signals ‘s’ during firm screening
and negotiation (Ueda, 2004). The bank can perceive two types of signals: a signal sh implying that the firm has a
high success probability and a signal sl implying a low success probability. The bank's ability to neutralise
asymmetric information is captured by the probability of bank observing a ‘perfect signal’ in line with the firm's
underlying success probability (Ueda, 2004). This implies that the bank observes a signal sh for a high success
probability firm and sl for a low success probability firm. In contrast, the degree of asymmetric information in
the model is reflected by the probability of bank observing a signal sh (or sl) when in reality the approaching
firm has low success probability (or high success probability). These probabilities are shown in Eqs. (16) and (17).



h l
Pr s j P h ¼ Pr s j P l ¼ α ð16Þ



h l
Pr s j P l ¼ Pr s j P h ¼ 1 − α where α ≥ 0:5 ð17Þ

‘α’ reflects the extent to which the bank perceives the ‘perfect signal’ for the associated firm, while 1 − α
represents the probability of the bank observing the ‘imperfect signal’. In the presence of perfect information,
the value of α is 1. With complete information asymmetry, the value of α is 0.5. It is also assumed that prior
to negotiations and screening, the bank do not possess any information about the firm and, hence, puts equal
probabilities (0.5) of a high and low success probability (Ueda, 2004). The bank realises its inability to
neutralise asymmetric information, and so calculates the conditional probabilities after observing signal ‘s’
as shown by Eq. (18).


h

 Pr s jP h PrðP h Þ 0:5α
h
Pr P j sh ¼     ¼ ¼α ð18Þ
Pr sh j P h PrðP h Þ þ Pr sh j P l PrðP l Þ 0:5α þ 0:5ð1 − α Þ




l h l
Pr P j sl ¼ α; Pr P j sl ¼ Pr P j sh ¼ 1 − α ð19Þ

The information asymmetry captured by Eqs. (18) and (19) are incorporated then in to the bank's
participatory constraint. This changes in the presence of the asymmetric information as follows:

α ðP h ðφr − f Þ−ð1 − P h Þf Þ þ ð1 − α Þð P l ðφr − f Þ − ð1 − P l Þf Þ ≥ 0 ð20Þ

which reduces to:

f
φr ≥ ð21Þ
αP h þ ð1 − α Þ P l

In a competitive market Eq. (21) has to be binding.22 So a binding constraint back into the firm's payoff
function yields the following:

I JV Ph f
πh ¼ P h ðφ − f Þ þ f − ð22Þ
αP h þ ð1 − α Þ P l

21
This assumption is for simplicity only and does not change the outcome of the model.
22
Otherwise the payoff to the firm can be increased without violating the constraint.
 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272 263

Since,

P l ≤ αP h þ ð1 − α ÞP l ≤ P h ð23Þ
Hence,
Ph
≥1 ð24Þ
αP h þ ð1 − α ÞP l

Ph f
f− b0 ð25Þ
αP h þ ð1 − α Þ P l

Therefore,

Ph f
P h ðφ − f Þ þ f − b P h ðφ − f Þ ð26Þ
αP h þ ð1 − α ÞP l

The right hand side in Eq. (26) captures the debt financing profit for a high success probability firm while
the left hand side shows the firm's IJV financing profit (given by Eq. (22)). Eq. (26), therefore, implies that for
the firm with a high success probability, their debt financing profit will always be higher than the IJV payoff
(given by Eq. (22)). Similarly, the payoff for low success profitability firm can be shown to be as follows:

I JV Pl f
πl ¼ P l ðφ− f Þ þ f − ð27Þ
αP h þ ð1 − α ÞP l

Pl f
f− N0 ð28Þ
αP h þ ð1 − α Þ P l

Pl f
P l ðφ − f Þ þ f − N P l ðφ − f Þ ð29Þ
αP h þ ð1 − α ÞP l

The right hand side in Eq. (29) captures the debt financing profit for low success probability firms while its
left hand side shows the firm's IJV financing profit (Eq. (27)). Eq. (29), therefore, implies that the IJV financing
profit for a firm with a low success probability will always be greater than their debt financing payoff. Since,
the firm will always prefer the financing mode that maximises their payoff, this leads to Proposition 2.

Proposition 2. High success probability firms will always prefer debt over IJV, while low success profitability
firm will prefer IJV over debt.

4.3. Moral hazards in IJV

The previous section investigated the impact of adverse selection and IJV financing in Islamic banking. This
section focuses on the impact of moral hazard on the firm's choice to use IJV financing. A bank can use
long-term relationships (Diamond, 1984, 1989; Boot, 2000; Bharath et al., 2007) and threat of punishment
(Merges, 1992; Landier, 2001; Ueda, 2004) to neutralise moral hazards. These two strategies are now
added to the model to show that where long-term relationships or threat of legal punishment exist,
IJV banking can operate even in the presence of moral hazard.
Moral hazard in IJV financing stems from ineffective monitoring resulting in excessive risk taking
and misreporting profit (Iqbal et al., 1998). Excessive risk taking results from the bank bearing part of the
downward IJV risk. Profit misreporting, however, is an outcome of the bank's performance dependent return.
If these risks are foreseen, the bank can incorporate them in the negotiated profit ratio ‘r’. These moral hazard
concerns, however, occur only after the financing contract has been negotiated. The following section
incorporates firm's incentive to indulge in excessive risk taking behaviour in IJV and shows how Islamic bank's
use of long-term relationships can neutralise it.
Our model captures excessive risk taking behaviour by assuming that bank does not monitor firm once
their IJV financing is approved (Gompers, 1995; Iqbal et al., 1998; Hellmann, 1998; Ahmed, 2001). This
264 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272

implies that a firm with a high success probability (Ph), after negotiating a lower profit rate ‘r’ with the Islamic
bank, can deviate from their strategy and implement risky projects with low success probabilities (Pl).
For simplicity the bank is assumed to have perfect information during the screening stage.23 This means
that the bank can differentiate between firms with high and low success probabilities during negotiations
for profit rate ‘r’. If the firm with a high success probability (Ph), negotiates an IJV contract and implements
it truthfully, then its payoff would be given by Eq. (30).

I JV
πh ¼ P h ðφ− f Þ ð30Þ

If the firm instead decides to shirk after negotiating with the bank and undertakes a risky project with low
success probability (Pl), then the payoff function would change to Eq. (31). The bank's participatory constraint
given by Eq. (38), however, would not change.

I JV
πl ¼ P l ðφ − φr Þ þ ð1 − P l Þf ð31Þ

The bank's participatory constraint:

P h ðφr − f Þ − ð1 − P h Þf ≥ 0 ð32Þ

f
φr ≥ ð33Þ
Ph

Putting the binding constraint given in Eq. (33) into Eq. (31) gives the firm's equilibrium payoff given by
Eq. (34).

I JV Pl f
πl ¼ P l ðφ − f Þ þ f − ð34Þ
Ph

The incentive to indulge in risk taking behaviour only occurs if the payoff from shirking (given by Eq. (34))
is higher than the payoff from the original less risky strategy as shown back in Eq. (14) (see Eq. (35)).

Pl f
P l ðφ − f Þ þ f − ≥ P h ðφ − f Þ ð35Þ
Ph

P l ðP h ðφ − f Þ − f Þ ≥ P h ðP h ðφ − f Þ − f ð36Þ

Since Pl b Ph Eq. (36) can only be true if Eq. (37) holds.

P h ðφ − f Þ − f ≤ 0 ð37Þ

Eq. (37) above can be simplified to Eqs. (38) and (39).

f
Ph ≤ ð38Þ
ðφ − f Þ
 
1
φ ≤ f 1þ ð39Þ
Ph

Eqs. (38) and (39) illustrate the conditions under which the firm with a high success probability might
implement the risky strategy. Eq. (38) provides the success probability threshold while Eq. (39) gives the
revenue (φ) limit, below which the firm will have an incentive to deviate and to take excessive risk. Next
we incorporate long-term relationships in the model to see if this feature can neutralise excessive risk taking
behaviour. Long-term IJV relationships may allow the bank to address excessive risk taking behaviour in a

23
This assumption is for simplicity only and does not change the outcome of the model.
 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272 265

particular period by terminating the contract in subsequent periods (Diamond, 1984, 1989; Iqbal et al., 1998;
Boot, 2000; Bharath et al., 2007).
The perfect information model discussed above is a static one period model. To incorporate long-term
relationships, the model must first be modified into multi-periods. This is achieved by assuming the bank
spreads the funding over multiple periods. Furthermore, the bank terminates subsequent funding if
the firm shirks from their negotiated strategy or performs poorly in the previous period. Therefore, should
the firm implement the low success probability (Pl) strategy rather than the high success probability (Ph)
negotiated strategy, the contract will be terminated. Moreover, if the firm performs poorly and fails with
probability (1 − Ph) in the previous period, the bank will refuse subsequent funding.
From the firm's perspective, the decision to deviate from the original strategy in the multi-period model
(other than depending on Eqs. (44) and (45)) depends on their loss from having no bank funding.24 If the
firm's profit incentive from shirking is lower than their expected multi-period profit then firm would lack
the incentive to shirk. This implies that in the presence of long-term relationships excessive risk taking
behaviour can be neutralised.
If high success probability firm implements the negotiated strategy in each period, it earns the expected
profit in Eq. (40).

I JV
πh ¼ P h ðφ − f Þ ð40Þ

If the firm follows the negotiated high success probability strategy, their expected profit in ‘n’ periods is
represented by Eq. (41)25 (for simplicity the discount rate R is assumed to be zero).

I JV 2 3 n
πh ¼ P h ðφ − f Þ þ P h ðφ − f Þ þ P h ðφ − f Þ þ :::::::::::: þ P h ðφ − f Þ ð41Þ

Assuming an infinite horizon, Eq. (41) can be reduced to Eq. (42).

I JV P h ðφ − f Þ
lim π h ¼ ð42Þ
n→∞ 1 − Ph

In each period, the firm faces this choice of earning an expected perpetual profit26 given by Eq. (42) against
the one time profit from deviating to the risky strategy given in Eq. (34). For the firm to cheat, Eq. (43) has to
hold true.

P l f P h ðφ − f Þ
P l ðφ − f Þ þ f − ≥ ð43Þ
Ph 1 − Ph

The above equation can be reduced to the following:

Ph Ph 2 f
Pl b þ ð44Þ
1 − P h ð1 − P h ÞðP h ðφ − f Þ − f Þ

Eq. (44) above can be reduced to Eq. (45).

P h 2 ðφ − f Þ − P h f P 2f
P l ðP h ðφ − f Þ − f Þ N þ h ð45Þ
1 − Ph 1 − Ph

24
For simplicity it is assumed that banks are the only available source of financing for the firm. This is not an unrealistic assump-
tion for if the firm cheats their reputational loss might deter other financial institutions from lending to them. This might also force
them to shut down.
25
The expected profit also incorporates the assumption that the banks would terminate funding if the firm fails with probability (1 − P)
in the previous period.
26
The results can also be reduced to a finite period.
266 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272

Given

P h ðφ − f Þ − f b 0 ð46Þ

P h ðP h ðφ − f Þ − f Þ Ph 2 f
Pl b þ ð47Þ
ð1 − P h ÞðP h ðφ − f Þ − f Þ ð1 − P h Þ

Solving the above equation yields the following:


Ph Ph f
Pl b 1− ð48Þ
1 − Ph ð f − P h ðφ − f ÞÞ

If it can be shown that Phf N f − Ph(φ − f), then there will be no value of Pl N 0 for which Eq. (48) holds. The
proof given is:

P h f N f − P h ðφ − f Þ ð49Þ

f − Ph φ b 0 ð50Þ

Ph φ N f ð51Þ

Eq. (51) contradicts the conditions given in Eqs. (38) and (39), under which the firm will have an incentive
to implement the risky strategy. This proves that there will be no values of Pl N 0 for which the firm will have
an incentive to cheat and so leads to Proposition 3.

Proposition 3. In the presence of long-term relationships the firm has no incentive to indulge in excessive risk
taking in IJV contracts.

4.4. Legal punishment and misreporting

The previous section showed that the presence of long-term relationships enables the bank to neutralise
their moral hazard concerns arising from excessive risk taking behaviour. This section analyses the impact
of legal punishment in neutralising other moral hazard concerns such as misreporting (Merges, 1992; Iqbal
et al., 1998; Landier, 2001).
The model assumes that after the initial profit rate has been negotiated, the firm can choose to report profit
to the bank by factor θ. Misreporting (lower θ) would reduce the IJV profit that is shared with the bank.
For simplicity it is assumed that the adverse selection problem does not exist and the bank has perfect
information about the nature of firm's success probability during initial negotiations. In such a setting the
firm's profit function would change, while the bank's participatory constraint would not change.27 Eq. (52)
provides the firm's expected profit equation in the presence of misreporting factor θ. The bank's participatory
constraint is given by Eq. (53).

I JV
π h ¼ P h ðφ − θφr Þ þ ð1 − P h Þf ð52Þ

Bank's participation constraint:

P h ðφr − f Þ − ð1 − P h Þ f ≥ 0 ð53Þ

Eqs. (52) and (53) above can be solved to yield the following:

I JV
πh ¼ P h ðφ − f Þ þ f − θf ð54Þ

27
The participatory constraint would have changed if this risk was foreseen by the banks.
 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272 267

Eq. (54) suggests that the firm's expected equilibrium profit increases with the increase in the
misreporting (a decrease in θ). Therefore, the firm will always have an incentive to misreport profit. Next
we show how the threat of punishment can neutralise misreporting concerns.
The model assumes that the probability of a firm's misreporting going unnoticed is Ps. Moreover, if the
firms are caught with probability (1 − Ps), they will be fined a lump sum amount ‘A’. In such a setting, the
firm will have an incentive to cheat only if the expected profit from cheating is greater than their profit
otherwise. This is captured by Eq. (55).

P s ðP h ðφ − f Þ þ f − θ f Þ − ð1 − P s ÞðAÞ N P h ðφ − f Þ ð55Þ

P s ððP h ðφ − f Þ þ f − θ f Þ þ AÞ N A þ P h ðφ − f Þ ð56Þ

A þ P h ðφ − f Þ
Ps N ð57Þ
ðP h ðφ − f Þ þ f − θ f Þ þ AÞ

The inequality in Eq. (57) shows that the firm's incentive to misreport profit is an outcome of their
probability of not getting caught. Therefore, in countries where the legal system is proficient in catching
defrauding firms and punishing them severely, Eq. (57) above would not hold true. This would neutralise
moral hazard concern arising from misreporting. The results prove Proposition 4.

Proposition 4. With a strong legal system, IJV banking can operate even in the presence of moral hazard.

4.5. Model with risk averse utility

Propositions 1 to 4 suggest that adverse selection rather than moral hazard is the biggest impediment in
the implementation of Islamic bank's IJV financing. If profits are sufficiently large, however, Islamic banks
can seek to specialise in VC type screening to offer IJV financing. This, however, does not appear to have hap-
pened in practice. The presence of risk averse bank customers on the Islamic bank's liability side (depositors)
could explain the bank's disincentive for using IJV screening methods. This section augments the asymmetric
information model with the risk averse utility function of Islamic bank customers to show that in the presence
of risk averse Islamic bank customers, firms would prefer debt over IJV.
The proof presented in this section explores the impact of bank customers' risk averse utility function on
the firm's IJV and debt financing equilibrium profits. On the Islamic bank's liability side if the bank customers
(depositors) invest in IJV financing accounts, their return will be uncertain with high risk of losing their principal
(Iqbal et al., 1998). Hence, they will have to be paid a premium to be induced into investing in these accounts.
In line with Leland and Pyle (1977), the degree of risk aversion can be captured by assigning bank
customers a von-Neumann type utility function with the following property:
 
1 2
EU ðφr Þ ¼ U Eðφr Þ− ρ σ ð58Þ
2

where ρ is the degree of risk aversion σ 2 is the variance in the depositor's earnings.
The model borrows all the assumptions from the asymmetric information IJV model. In such a setting, if the
bank customers invest in an IJV financing account, their expected utility is given by Eq. (59). In contrast, if the
account holders invest in a debt financing account with certain returns their utility is given by Eq. (60).
1
2 2

EU ðφr Þ ¼ Uðα P h φr þ ð1 − α Þ P l φr − ρ α ð1 − α ÞðP h − P l Þ φr ð59Þ
2
U ðð1 þ RÞf Þ ð60Þ

For the depositors to prefer the IJV financing account, the IJV financing utility given in Eq. (59) should be
greater than their debt financing utility given in Eq. (60). This is captured by Eq. (61).

1
2 2

Uðα P h φr þ ð1 − α Þ P l φr − ρ α ð1 − α ÞðP h − P l Þ φr ð61Þ
2
≥ U ðð1 þ RÞf Þ
268 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272

Assuming that the standard utility function assumptions such as, “more is preferred to less” and transitivity
hold, the above can be reduced to the following:

1 2 2
α P h φr þ ð1 − α Þ P l φr − ρ ðα ð1 − α Þ ðP h − P l Þ φr ≥ ð1 þ RÞf ð62Þ
2

Eq. (62) above can be reduced to Eq. (63) below (see Appendix A for the proof)
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
α P h þ ð1 − α ÞP l − ðα P h þ ð1 − α ÞP l Þ2 − 2ðρð1 þ RÞf ðα ð1 − α ÞðP h − P l Þ2
φr ¼ ð63Þ
ρðα ð1 − α ÞðP h − P l Þ2

Eq. (63) suggests that the in the presence of Eq. (62), the bank's participatory constraint should automatically
hold true. Substituting Eq. (63) in the firm's profit function gives the following equilibrium payoffs.
2 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 3
α P h þ ð1 − α ÞP l − ðα P h þ ð1 − α ÞP l Þ2 − 2ðρf ðα ð1 − α ÞðP h − P l Þ2
π h ¼ P h ðφ−f Þ þ f −P h 4 5 ð64Þ
I JV
ρðα ð1 − α Þ ðP h − P l Þ2

A high success probability firm prefers IJV to debt if Eq. (65) holds.
2 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 3
α P h þ ð1 − α ÞP l − ðαP h þ ð1 − α ÞP l Þ2 − 2ðρð1 þ RÞf ðα ð1 − α ÞðP h − P l Þ2
P h ðφ − f Þ þ f − P h 4 5 ≥ P h ðφ − f Þ
ρðα ð1 − α ÞðP h − P l Þ2

ð65Þ

Eq. (65) above can be reduced to Eq. (66) below (see proof in Appendix B).

2P h ððαP h þ ð1 − α ÞP l Þ − P h ð1 þ RÞÞ
ρ≤ ð66Þ
f ðα ð1 − α ÞðP h − P l Þ2

Since ((a Ph + (1 − a)Pl) − Ph) b 0, Eq. (66) implies that ρ b 0. This suggests that a high success probability
firm will only use IJV if the Islamic bank depositors are risk seeking.
For a firm with low success probability, it can easily be shown that (for a small enough R), they would
prefer IJV financing only if Eq. (67) holds.

2 P l ððαP h þ ð1 − α ÞP l Þ − P l ð1 þ RÞÞ
ρN ð67Þ
f ðα ð1 − α ÞðP h − P l ÞÞ2

Since ((αPh + (1 − α)Pl) − Pl) N 0, for R = 0 (or low enough ‘R’), Eq. (67) implies that ρ N 0. This suggests
that a firm with a low success probability might be interested in IJV financing, even if the depositors are risk
averse. It can be easily shown, however, that if ‘R’ increases beyond a certain optimal then even a low success
probability firm would prefer IJV only if depositors are risk loving with ρ b 0.
Eq. (66), therefore, suggests that the equilibrium payoff for a high success probability IJV financing
firm in the presence of risk averse Islamic bank customers will always be less than that of their debt
financing payoff. This implies that a high success probability firm should always prefer debt to IJV
financing. Eq. (67) suggests that the profit incentive for low success probability firms in the presence
of highly risk averse bank customers would also erode away with a high enough ‘R’. This leads to the
following proposition.

Proposition 5. In the presence of risk averse Islamic bank depositors, firms would prefer debt over IJV.

4.6. Summary of the propositions and their results

The impact of adverse selection on a firm's choice of financing was reflected in Propositions 1 and 2.
Proposition 1 was proved by developing a firm focused model in a perfectly competitive banking sector and
assumed perfect information. The presented proof suggested that under perfect information a borrowing
 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272 269

firm should be indifferent between IJV and debt financing. Asymmetric information was then introduced. The
proof showed that joint venture contracts, which are more prone to asymmetric information than debt,
reduce the borrowing firm's profitability. This lead to Proposition 2, implying that the presence of asymmetric
information makes debt rather than IJV the borrowing firm's preferred financing choice.
Propositions 3 and 4 imply that long-term relationships and legal punishment can neutralise IJV moral
hazard concerns. To prove these propositions, the model incorporated the impact of long-term relationships
and legal punishment to neutralise moral hazards such as excessive risk taking behaviour and misreporting.
For long-term relationships, we extended the model from a single to multiple periods. It also assumed that
if the firm deviated from their contractual strategy and indulged in excessive risk taking behaviour in any
particular period, the bank would terminate the firm's contract in the subsequent period. The proof showed
that the firm's loss from contract termination was much higher than its gains from excessive risk taking.
Legal punishment was then introduced. The proof showed that in countries where property rights are well
protected, moral hazard concerns (such as misreporting) are not severe.
Together Propositions 1, 2, 3 and 4 show that adverse selection deters IJV financing more than moral
hazards. The reason why Islamic banks' failure to implement VC type screening and so neutralise adverse
selection is reflected in Proposition 5.
For Proposition 5, the model incorporated the impact of risk averse bank customers. The proof showed
that in the presence of liability side funding from risk averse bank customers (depositors), any gains from
an Islamic bank specialising in VC type screening are lost by the firm and the bank. This is because any
gains are absorbed by the additional risk premium passed to bank customers to persuade them to invest
with an Islamic bank implementing IJV financing.

5. Conclusion

This paper questioned whether Islamic banks would be likely to offer more profit and loss sharing
(Musharakah/Mudarabah) type contracts and therefore to be considered more ‘Islamic’, rather than
continuing with their predominantly trade or debt styled contracts (Murabaha). We explained the dominance
of debt products of the asset side of Islamic banks through developing theoretical models that captured the
effects of asymmetric information, risk averse bank depositors, long-term contracts and legal punishments.
This shows that in the presence of asymmetric information and Islamic banks funded by risk averse customers
(depositors), a borrowing firm could maximise their profit through debt contracts rather than IJV. This
minimises a firm's demand for IJV and hence explains the lack of IJV client financing by Islamic banks.
The results also show that the presence of asymmetric information alone cannot fully explain the absence
of IJV financing. It is only when the models are augmented to reflect risk averse Islamic bank customers
(depositors), that it can offer a comprehensive explanation. So in answer to our initial question we do not
believe Islamic banks are likely to offer more IJV contracts in the future.
Our findings have several implications for both regulators and financial institutions seeking to facilitate IJV
financing. The first is that the Islamic bank current funding structure is unlikely to result in a significant level of
IJV financing without first changing the attitudes of depositors. This could be addressed partly through better
financial education in general and in Islamic finance more specifically. It might also help to introduce mutual
fund styled investment products designed to attract different risk preference funds. It is not clear that the
Islamic bank itself would be the best home for this product, particularly given global regulatory pressures
on risk taking. The second is that emphasis might instead be placed on offering them through quite separate
venture capital and private equity (PE) corporate structures. This would offer investors a clear choice of risk
and return. Like their conventional VC-PE counterparts, these products would be designed to attract long
terms investors, particularly institutional ones, and have some limitations on withdrawals. The third is
that countries lacking sufficient IJV investment could encourage further development of Takaful insurance
offices and Shariah compliant pension funds as the Malaysian government has done. These changes would
encourage more IJV investment and generally facilitate the use of more stringent VC screening and monitoring
techniques and so hopefully overcome the asymmetric information problems that joint venture funding
entails. The fourth would be to encourage Islamic banks to hire conventional VC and PE staff to run these
businesses under an Islamic framework rather than attempting to do so using their existing banking staff.
Malaysia's International Centre for Education in Islamic Finance (INCEIF) has already had a significant impact
on the quality of Islamic banking and finance education and these programmes could perhaps be broaden to
270 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272

address the needs of a Shariah focused VC and PE industry. In conclusion, this paper shows that it is the Islamic
bank's Shariah conscious ethical depositors that effectively drive the current choice of products rather the
Islamic banks or their regulators.

Acknowledgements

We would sincerely like to thank Sudipto Dasgupta, Munawar Iqbal and an anonymous reviewer for their
suggestions.

Appendix A

1 2 2
α P h φr þ ð1 − α Þ P l φr − ρðα ð1 − α ÞðP h − P l Þ ðφr Þ − ð1 þ RÞf ¼ 0 A:1
2

The above quadratic equation of ‘φr’ can be reduced to:

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
þ
α P h þ ð1 − α ÞP l ð αP h þ ð1 − α ÞP l Þ2 − 2ðρð1 þ RÞf ðα ð1 − α ÞðP h − P l Þ2
−
φr ¼ A:2
ρðα ð1 − α ÞðP h − P l Þ2

To check which of the two solutions are in line with assumption of ∂∂ððφr
ρÞ
Þ
N0 we take the first derivative.

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
þ
−½α P h þ ð1 − α Þ P l ð α P h þ ð1 − α ÞP l Þ2 − 2ðρð1 þ RÞf ðα ð1 − α ÞðP h − P l Þ2 Þ
∂ðφr Þ −
¼
∂ðρÞ ρ2 ðα ð1 − α ÞðP h − P l Þ2
þ A:3
ð − 2ðð1 þ RÞ f ðα ð1 − α ÞðP h − P l Þ2
−
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 ρðα ð1 − α ÞðP h − P l Þ2 ð αP h þ ð1 − α ÞP l Þ2 − 2ðρð1 þ RÞf ðα ð1 − α ÞðP h − P l Þ2

It can be easily shown that:

∂ðφr Þ
b 0 when the sign is positive A:4
∂ðρÞ

∂ðφr Þ
N 0 when the sign is negative A:5
∂ðρÞ

Therefore we reject the solution with the positive sign.

Appendix B
2 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 3
α P h þ ð1 − α Þ P l − ðα P h þ ð1 − α ÞP l Þ2 − 2ðρð1 þ RÞf ðα ð1 − α ÞðP h − P l Þ2
P h ðφ − f Þ þ f − P h 4 5 N P h ðφ − f Þ
ρðα ð1 − α ÞðP h − P l Þ2

ðB:1Þ

2 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 3
α P h þ ð1 − α Þ P l − ðαP h þ ð1 − α ÞP l Þ2 − 2ðρð1 þ RÞf ðα ð1 − α ÞðP h − P l Þ2
f − Ph 4 5N0 ðB:2Þ
ρðα ð1 − α ÞðP h − P l Þ2
2 qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi3
α P h þ ð1 − α Þ P l − ðαP h þ ð1 − α ÞP l Þ2 − 2ðρð1 þ RÞ f ðα ð1 − α ÞðP h − P l Þ2
f N Ph 4 5 ðB:3Þ
ρðα ð1 − α ÞðP h − P l Þ2
 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272 271

h i2
2
f ρðα ð1 − α ÞðP h − P l Þ − P h ðα P h þ ð1 − α ÞP l Þ
ðB:4Þ
2 2
N ð α P h þ ð1 − α ÞP l Þ − 2ðρð1 þ RÞ f ðα ð1 − α ÞðP h − P l Þ

Squaring both sides:

h i

2 2 2
f ρðα ð1 − α ÞðP h − P l Þ − 2ðf ρ α ð1 − α ÞðP h − P l Þ ðP h ðα P h þ ð1 − α ÞP l Þþ
h i ðB:5Þ
2 2 2 2
½P h ðα P h þ ð1 − α Þ P l  N P h ðαP h þ ð1 − α ÞP l Þ − 2ðð1 þ RÞ f ðα ð1 − α ÞðP h − P l Þ

h i

 i
2 2 2
f ρðα ð1 − α ÞðP h − P l Þ − 2 f ρ α ð1 − α ÞðP h − P l Þ ðP h ðαP h þ ð1 − α ÞP l ÞÞ
ðB:6Þ
2 2
N − 2P h ðρð1 þ RÞf ðα ð1 − α ÞðP h − P l Þ

The above inequality can be reduced to:

−2P h 2 ð1 þ RÞ þ 2ðP h ðαP h þ ð1 − α ÞP l ÞÞ

ρb ðB:7Þ
f ðα ð1 − α ÞðP h − P l Þ2

−2P h 2 ð1 þ RÞ þ 2ðP h ðα P h þ ð1 − α ÞP l ÞÞ
ρb ðB:8Þ
f ðα ð1 − α ÞðP h − P l Þ2

2P h ððαP h þ ð1 − α ÞP l Þ − P h ð1 þ RÞÞ
ρb ðB:9Þ
f ðα ð1 − α ÞðP h − P l Þ2

For firm with low success probability with a low enough ‘R’:

2P l ððαP h þ ð1 − α ÞP l Þ − P l ðð1 þ RÞÞ

ρN ðB:10Þ
f ðα ð1 − α ÞðP h − P l Þ2

If ‘R’ is above a certain optimal then the following equation holds:

2P l ððαP h þ ð1 − α ÞP l Þ − P l ðð1 þ RÞÞ

ρb ðB:11Þ
f ðα ð1 − α ÞðP h − P l Þ2

References

Ahmed, H., 2001. A microeconomic model of an Islamic bank. Islamic Development Bank Research Paper 59.
Ahmed, H., 2002. Incentive compatible profit sharing contracts, in Islamic banking and finance. In: Iqbal, Munawar, David, T., Llewellyn,
Edward, Cheltenham, U.K. (Eds.), New Perspectives on Profit Sharing and Risk, pp. 41–57.
Ahmed, H., 2005. Operation structure for Islamic equity finance: lessons from venture capital. Islamic Development Bank Research Paper
69.
Akerlof, G.A., 1970. The market for “lemons”: quality uncertainty and the market mechanism. Q. J. Econ. 84, 488–500. http://dx.doi.org/
10.2307/1879431.
Ayub, M., 2007. Understanding Islamic Finance. John Wiley and Sons, Chichester, UK.
Barry, C.B., 1994. New directions in research on venture capital finance. Financ. Manag. 23, 3–15.
Beck, T., Demirguc-Kunt, A., Merrouche, O., 2013. Islamic vs. conventional banking: business model efficiency, and stability. J. Bank.
Financ. 37, 433–447. http://dx.doi.org/10.1016/j.jbankfin.2012.09.016.
Bester, H., 1985. Screening vs. rationing in credit markets with imperfect information. Am. Econ. Rev. 75, 850–855.
Bharath, S., Dahiya, S., Saunders, A., Srinivasan, A., 2007. So what do I get? The bank's view of lending relationships. J. Financ. Econ. 85,
368–419.
Boot, A.W., 2000. Relationship banking: what do we know? J. Financ. Intermed. 9, 7–25. http://dx.doi.org/10.1006/jfin.2000.0282.
Chan, Y.S., Thakor, A.V., 1987. Collateral and competitive equilibria with moral hazard and private information. J. Financ. 42, 345–363.
http://dx.doi.org/10.1111/j.1540-6261.1987.tb02571.x.
Chong, B.S., Liu, M.H., 2009. Islamic banking: interest free or interest based? Pac. Basin Financ. J. 17, 125–144. http://dx.doi.org/10.1016/j.
pacfin.2007.12.003.
Diamond, D.W., 1984. Financial intermediation and delegated monitoring. Rev. Econ. Stud. 51, 393–414. http://dx.doi.org/10.2307/
2297430.
Diamond, D.W., 1989. Reputation acquisition in debt markets. J. Polit. Econ. 97, 828–862. http://dx.doi.org/10.1086/261630.
Diamond, D., Dybvig, P., 1983. Bank runs deposit insurance, and liquidity. The Journal of Political Economy 91, 401–419.
272 S. Azmat et al. / Pacific-Basin Finance Journal 34 (2015) 253–272

(EY) Ernst & Young, 2014. Global Takaful insights: Market updates.
Gale, D., Hellwig, M., 1985. Incentive compatible debt contracts. Rev. Econ. Stud. 52, 647–663. http://dx.doi.org/10.2307/2297737.
Gompers, P.A., 1995. Optimal investment, monitoring, and the staging of venture capital. J. Financ. 50, 1461–1489. http://dx.doi.org/10.
1111/j.1540-6261.1995.tb05185.x.
Gompers, P., Lerner, J., 1999. The Venture Capital Cycle. MIT Press, London, UK.
Hellmann, T., 1998. The allocation of control rights in venture capital contracts. RAND J. Econ. 29, 57–76.
Iqbal, M., Ahmed, A., Khan, T., 1998. Challenges facing Islamic banking. Islamic Development Bank Paper No. 1.
Islamic Association of Deposit Insurers, 2010. Deposit insurance from the Shariah perspective. Discussion paper. Bank for International
Settlements (February).
Kaplan, S.N., Stromberg, P., 2001. Venture capitals as principals: contracting, screening, and monitoring. Am. Econ. Rev. 91, 426–430.
http://dx.doi.org/10.1257/aer.91.2.426.
Kasri, R.A., Kassim, S.H., 2009. Empirical determinants of saving in Islamic banks: evidence from Indonesia. Islam. Econ. 22 (2), 181–201.
Khalil, A.F., Rickwood, C., Murinde, V., 2002. Evidence on agency-contractual problems in mudarabah financing operations by Islamic
banks. In: Iqbal, Munawar, Llewellyn, David T., Elgar, Edward, Cheltenham, U.K. (Eds.), In Islamic Banking and Finance: New Perspec-
tives on Profit Sharing and Risk, pp. 57–94.
Khan, F., 2010. How ‘Islamic’ is Islamic banking? J. Econ. Behav. Organ. 76, 805–820. http://dx.doi.org/10.1016/j.jebo.2010.09.015.
Landier, A., 2001. Start-up financing: banks vs. venture capital. MIT Working Paper.
Leland, H., Pyle, D., 1977. Informational asymmetries, financial structure and financial intermediation. J. Financ. 32, 371–387. http://dx.
doi.org/10.1111/j.1540-6261.1977.tb03277.x.
Merges, R.P., 1992. Patent Law and Policy: Cases and Materials. Michie Co., Charlottesville, Va.
Sahlman, W.A., 1990. The structure and governance of venture-capital organizations. J. Financ. Econ. 27, 473–521. http://dx.doi.org/10.
1016/0304-405X(90)90065-8.
Stiglitz, J.E., Weiss, A., 1981. Credit rationing in markets with imperfect information. Am. Econ. Rev. 71, 393–410.
Townsend, R.M., 1979. Optimal contracts and comparative markets with costly state verification. J. Econ. Theory 21, 265–293.
Ueda, M., 2004. Banks versus venture capital: project evaluation, screening, and expropriation. J. Financ. 59, 601–621. http://dx.doi.org/
10.1111/j.1540-6261.2004.00643.x.
Usmani, T.M., 2002. An Introduction to Islamic Finance. Kluwer Law International, Hague, Netherlands.