�������� ����� ��

Issuer’s Choice of Islamic Bond Type

Saad Azmat, Michael Skully, Kym Brown

PII: S0927-538X(13)00051-6DOI: doi: 10.1016/j.pacfin.2013.08.008Reference: PACFIN 643

To appear in: Pacific-Basin Finance Journal

Received date: 12 July 2013Accepted date: 17 August 2013

Please cite this article as: Azmat, Saad, Skully, Michael, Brown, Kym, Is-suer’s Choice of Islamic Bond Type, Pacific-Basin Finance Journal (2013), doi:10.1016/j.pacfin.2013.08.008

This is a PDF file of an unedited manuscript that has been accepted for publication.As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

1

Issuer’s Choice of Islamic Bond Type

Abstract: This paper analyses factors affecting an issuer’s choice of Islamic bond structure

as compared with conventional financial instruments. This choice is considered in the context

of issuer firm variables, the 2008 Accounting and Auditing Organization for Islamic

Financial Institutions (AAOIFI) Islamic bond recommendations, and Shariah advisor effect in

relation to Islamic instruments. A sample of Malaysian Islamic bonds is analysed using

ordered probit model techniques. The results suggest that there are some significant

differences between Islamic and conventional bond issuer’s choice determinants which can

be attributed to characteristics specific to Islamic bonds. For instance unlike conventional

bonds, the stock valuation of the issuer did not impact debt-equity targets with Islamic Joint

Venture (IJV) bond issuance. Other results from issuer firm characteristics were mixed and

suggest that IJV bonds have little in common with equity and issuers should concentrate on a

bond’s security and seniority as with conventional bonds, rather than their Islamic structure.

Secured Against Real Asset (SARA) bonds were found not to always represent ownership of

the underlying asset. AAOIFIs reported concerns in 2008 over Shariah quality of IJV bonds

appears to have led to an aversion of IJV bond issuance. Finally, Shariah committees as

opposed to individual Shariah advisors demonstrated an aversion to IJV bond issuance.

JEL: G15, G24

Keywords: bond structure, Islamic bonds *

* Saad Azmat

Assistant Professor

Suleman Dawood School of Business

Lahore University of Management Sciences

D.H.A, Lahore Cantt, 54792

Pakistan

Ph: 92 42 111 11 5867; Email: saad.azmat@lums.edu.pk

Professor Michael Skully

Department of Accounting and Finance

Monash University

P.O. Box 197

Caulfield East Victoria 3145

Australia

Ph: 61 3 9903 2407; Fax: 61 3 9903 2422; Email: Michael.Skully@monash.edu

Dr Kym Brown (corresponding author)

Department of Accounting and Finance

Monash University

P.O. Box 197

Caulfield East Victoria 3145 Australia

Ph: 61 3 9903 1053; Fax: 61 3 9903 2422; Email: Kym.Brown@monash.edu

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

2

1. Introduction

The factors that affect an issuer’s choice of different financial instruments have been widely

researched. Decisions like issuing debt as opposed to equity or secured rather than unsecured

bonds have been attributed to different combinations of firm specific variables such as debt

and profitability ratios (Julio et al., 2008). In addition, those characteristics unique to each

instrument such as security, seniority, and tangibility of the assets may also have a significant

impact on this decision as well as industry specific and macroeconomic factors.

The unique characteristics of Islamic bonds may further affect the issuer’s choice. One

difference, for example, is that secured against real asset (SARA) Islamic bonds have

entitlement to actual ownership of the underlying asset rather than just a first charge.1 This

means that SARA bonds are more secure than their conventional counterparts. In contrast,

Islamic joint venture bonds (IJV) have more similarities with equity than debt. The prior

literature has not investigated the impact that this wider range of Islamic bond characteristics

may have on an issuer’s debt selection. This gap motivates us to question whether the factors

that affect an issuer's choice of Islamic bond are the same as for conventional bonds.

This paper analyses the determinants of the issuers’ bond preference for 456 Malaysian

Islamic bonds. The impact of firm specific variables, specific events (such as the 2008

Accounting and Auditing Organization for Islamic Financial Institutions (AAOIFI)

recommendations) and specific Islamic instrument characteristics (such as Shariah advisor

effect) are analysed. The results show that some firm specific variables affect issuer’s choice

in a conventional bond like manner but others impact it differently. The findings suggest that

1 A holder with a first charge over a specific asset will be entitled to be the first in line to be repaid first from the

proceeds of its sale in case of default.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

3

both specific events and Islamic instrument characteristics are significant determinants of the

issuer’s decision. For example, the results show that in the aftermath of 2008 AAOIFI

recommendations, issuer’s aversion for IJV bonds increased. Also it is found that IJV bonds

are less likely to be approved by Shariah committees.

The findings have implications for regulators and Islamic bond issuers. For regulators, it is

suggested that they further support Shariah harmonization for the industry to prosper. For

issuers, the results imply that firm specific variables should affect their choice of Islamic

bond type not unlike that of conventional debt. The rest of the paper is organized as follows.

Section 2 considers the unique features of Islamic bond structures. Section 3 provides the

literature review, with data described in Section 4. Section 5 presents the methodology,

Section 6 the results and Section 7 concludes the paper.

2. Islamic Capital Market Features

This section considers the characteristics of three main types of Islamic bonds: Islamic joint

venture (IJV), secured against real asset (SARA) bonds and debt bonds2, as well as the role of

Shariah advisors and the AAOIFI (2008) updated Shariah standards.

2.1 IJV Bonds

IJV bonds (Musharakah/Mudarabah Sukuk) have both debt and equity characteristics. Their

holders become part owners of the issuer or more commonly of a specific project. If

successful, IJV bond holders share in the profits and receive periodic payments depending on

a pre-agreed percentage. They must also bear any loss according to their investment

2 AAOIFI has standards for more than 14 different types of Islamic bonds but the focus here is on the most

popular.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

4

proportion on maturity; the issuer can then buy back the IJV bonds but at their market price

rather than face value (Usmani, 2006). Furthermore on bankruptcy, IJV bond holders rank

equally with shareholders rather than normal creditors, but unlike shareholders have no

voting rights.

2.2 SARA Bonds

Secured against real asset (SARA) bonds (Ijarah Sukuk) are typically structured using a

special purpose vehicle (SPV). While it may be created by the issuer, the SPV is legally

separate and considered a bankruptcy remote entity just as in conventional structured finance.

The originator sells the SPV some specific assets which are agreed to and then leased back at

a periodic rental rate and then repurchased at a certain price and a certain date. The SPV uses

this set of agreements to support a SARA bond issue and uses the issue’s proceeds to pay the

issuer for the assets. The SPV collects the rental payments and pays them to the bond holders

as well as on maturity, the face value. SARA bonds differ from conventional ones in that

their holders have the legal and economic ownership of the underlying assets rather than just

a charge against them (Usmani, 2006, Ali, 2008). If the initial issuer defaults or goes

bankrupt, the bond holder can claim the possession of the underlying asset.

2.3 Debt Bonds

Debt bonds (Murabaha Sukuk) are based on buy and resale arrangement. This is where the

Islamic bank or an issuer SPV first buys a real asset and then immediately sells it back to the

client at a higher price creating a debt which their client periodically repays. These bonds can

be either secured or unsecured depending on their link with the underlying asset.3 Debt

3 Some Islamic debt bonds have a first charge on the underlying asset and hence can be considered similar to

secured conventional bonds but the rest are unsecured (Dusuki and Mokhtar, 2010).

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

5

bonds, however, do not represent the ownership of the underlying asset and so are less secure

than SARA bonds (Ali, 2005, 2008; Howladar, 2006; Dusuki and Mokhtar, 2010). Unlike

IJV and SARA bonds, debt bonds are considered non-Shariah compliant in most countries

other than Malaysia (Ayub, 2007).

2.4 Shariah Advisors

As Shariah compliance is a key requirement for any Islamic financial instrument, Islamic law

experts known as Shariah advisors are needed (Karim, 1990; Karim and Archer, 2007). They

can be an individual, a firm or several advisors in the form of a Shariah advisory committee

(Grais and Pellegrini, 2006). Since Islamic law (particularly with regards to contracts) can be

subject to different interpretations, designing new financial structures can be challenging,

especially across countries. So an Islamic instrument approved in one country may be

considered non-Shariah compliant in others. Islamic debt bonds for example are considered

Shariah-compliant in Malaysia but not in the Middle East and Pakistan (Usmani, 2002).

These Shariah differences stem from their adherence to particular Islamic school of thought.4

2.5 Shariah Standards

This lack of Shariah harmonization is costly for the Islamic finance industry. Regulatory

bodies such as AAOIFI have tried to address these differences by devising uniform Shariah

standards for most existing instruments. Many countries, like Bahrain, Sudan, Jordan, Qatar,

Saudi Arabia, Dubai, Syria, Lebanon, Pakistan, Singapore, and South Africa, are trying to

adopt these standards.

4 They are four major Madahibs (Islamic school of thought), Hanafi, Shafee, Hanbali and Maliki, named after

their founders.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

6

In February 2008, AAOIFI’s issued stringent guidelines for Islamic bonds (Usmani, 2007) as

many were being issued in a form similar to conventional debt. It argued against IJV bonds

where the issuer guaranteed the return and the principal irrespective of the firm’s

performance. Instead it required the IJV bonds be structured like equity where neither the

principal nor return guaranteed. AAOIFI (2008) also criticised those SARA bonds for being

structured like conventional secured debt. It required instead that SARA bonds represent real

ownership of the underlying asset rather than just a first charge5 and declared those Islamic

debt bonds with no link to the underlying asset non-Shariah compliant (Usmani, 2007). These

2008 recommendations had severe repercussions and billions of dollars of losses (Salah,

2010). This happened irrespective of whether a country complied with AAOIFI or not. For

example, Malaysia although an AAOIFI member country follows its own standards

developed by Bank Negara Malaysia but also suffered after these announcements (Dusuki

and Mokhtar, 2010).

3. Literature Review

This review begins with an overview of the determinants of an issuer’s choice of financial

instruments and then investigates elements that are particularly relevant to Islamic bonds. The

first focus is on those factors affecting corporate issuance of equity as opposed to debt. The

idea is to infer an issuer’s choice of IJV bond with equity like structure. Then the

determinants of a conventional secured bond versus unsecured bonds choice are examined

and then applied to examine the issuer’s choice of the more secured SARA bonds to Islamic

debt bonds. Finally, the impact of changing Shariah standards on the issuer’s choice of

5 SARA bond holders with first charge would only receive payments after undergoing lengthy bankruptcy

proceedings. Moreover, a first charge does not result in the full proceeds from the sale of the asset as some cost

and expenses might need to be paid (Dusuki and Mokhtar, 2010).

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

7

Islamic bonds is investigated as are the impact of Islamic instrument specific variables such

as Shariah advisor effect.

3.1 Overview of Issuer’s Choice of Financial Instruments

Conventional bonds differ in terms of many characteristics including maturity and security.

This diversity is caused by a range of factors. Bond maturity, for example, may be affected

by the firm’s risk. A low quality firm with high risk may avoid short-term debt as their poor

cash position could result in frequent rollovers (Flannery, 1986). The decision on secured as

opposed to unsecured bonds may be affected by other factors such as firm performance and

project risk. Companies that perform poorly or start risky projects may use secured bonds to

attract reluctant investors (Berger and Udell, 1990). Since Islamic bonds have certain

similarities with conventional bonds, implications can be drawn for them by analysing the

determinants of conventional bonds characteristics.

3.2 Determinants of Issuer’s Choice of IJV Bonds

As IJV bonds have a number of features similar to equity, the trade-off theory of capital

structure can be used to analyse issuer’s choice of IJV bond type. This tries to balance the tax

advantage of debt against its cost of financial distress (Modigliani and Miller, 1958).6 While

irrelevant in an ideal Miller-Modigliani world, market imperfections like asymmetric

information, taxes and financial distress costs make capital structure choice important. Given

debt and equity have differing costs and benefits, an optimal capital structure will balance

them. Trade-off theory suggests that taxes and financial distress costs would be important in

choice of IJV over SARA or debt bonds. In most jurisdictions however, IJV bonds, have a

6 There are also agency costs associated with equity, which negatively affect its favourability over debt (Jensen

and Meckling, 1976; Myers, 1977; Stulz, 1990) but some of them are not present in IJV bonds.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

8

similar tax treatment as SARA and debt bonds (Ali, 2005, 2008). Distress cost may also

affect the issuer’s preference for IJV bonds as given their equity like nature IJV bonds are not

susceptible to default.7

Pecking order theory suggests that in a world with asymmetric information, capital structure

offers an important signal about the nature of the firm (Ross, 1977). It assumes that managers

are better informed than investors, (Myers and Majluf, 1984) and so view equity issues as

negative signals about an overvalued share price. This cost makes firms prefer internal as

opposed to external financing. If external financing is inevitable, then firms prefer debt over

equity as debt issues avoids the negative signals of equity financing. So in Islamic terms, all

things being equal, a firm should prefer SARA and debt bonds, which are more similar with

conventional debt, over equity type IJV bonds.

Another strand of the literature uses theoretical arguments from both the trade-off and

pecking order theories and empirically examines the impact of different factors on the firm’s

debt-equity ratio (Bayless and Chaplinsky, 1993; Graham, 1996). It argues that firms target a

particular debt-equity ratio. If the actual ratio differs from the target, the firm would change

its debt or equity to adjust it (Marsh, 1982; Hovakimian et al., 2001). This implies that the

debt-equity ratio is affected by two types of factors: those that shift the target ratio and those

that cause deviations from the target. The former include the cost of financial distress,

bankruptcy risk, size, and asset composition which may affect the target level of debt (Marsh,

1982). In contrast, the primary catalyst for deviations is over or under-valued stocks

(Hovakimian et al., 2001). If stocks are over-valued, issuing equity would be less costly. This

7 We are referring to AAOIFI compliant IJV bonds which like ordinary shares have no default and hence, would

endure no financial distress costs. In the real world though, IJV bonds are being issued in a very debt like

manner and could face a default and hence would have to bear financial distress costs.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

9

in turn might induce firms to issue equity irrespective of their target ratio alternatively if

stocks are under-valued, firms may restrain from issuing equity (Ross, 1977; Myers and

Majluf, 1984).

Given the above arguments, if the firm’s current debt is higher than its targeted level, firms

may prefer an IJV bond over SARA and debt bonds to readjust to their target. In contrast, if

the current debt level is lower, then SARA or debt bonds will be preferred over IJV bonds.

Similarly undervalued firms should prefer SARA or debt bonds over IJV. Overvalued firms,

however, should prefer IJV. This motivates the following hypotheses:

H1: Firms exceeding their targeted debt levels will prefer IJV bonds over SARA bonds and

debt bonds (and vice versa).

H2: Holding other firm specific variables constant, firms with higher market to book values

should prefer IJV bonds over SARA and debt bonds (and vice versa).

3.3 Choice of SARA and Debt Bonds

This section examines the security and seniority preferences of a conventional bond issuer

and then extends them to Islamic bonds (SARA and debt). The security and seniority choices

with conventional bonds are explained by opposing views in the literature. One argument

states that secured debt is suitable for a highly levered firm with potential growth

opportunities as the financing of profitable projects that would otherwise been rejected if

financed through unsecured debt (Stulz and Johnson, 1985). The opposing view states that

only secured debt allows poorly performing firms to raise finance (Berger and Udell, 1990).

According to this view, high risk firms with limited growth opportunities would tend to issue

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

10

secured debt. Julio et al. (2008) for example find that firms issuing secured debt tend to be

small, highly levered, have low cash flows, low stock returns and limited growth options.

SARA and debt Islamic bonds differ in their seniority and security. SARA bonds are secured

by their ownership of the underlying asset. So on default their holders are entitled to the

underlying asset’s sale proceeds. In contrast, Islamic debt bonds are less secured but rank

above IJV bonds. If the views of Berger and Udell (1990) and Julio et al. (2008) can be

extended to Islamic bonds, then highly levered, poor performing and high risk firms would

prefer SARA bonds. In contrast, firms with lower leverage, high profitability and low risk

may prefer debt bonds. This leads to the following hypotheses:

H3: Firms with high leverage (total and long-term debt) prefer SARA over debt bonds and

vice versa.

H4: Firms having high profitability (operating margin and interest coverage ratio) prefer debt

over SARA bonds and vice versa.

H5: Firms having higher risk (market beta and standard error) prefer SARA over debt bonds

and vice versa.

3.4 Impact of 2008 AAOIFI Standards

The Accounting and Auditing Organization for Islamic Financial Institutions (AAOIFI)

stringent guidelines for Islamic bonds were designed to improve their Shariah quality

(Usmani, 2007) and suggested more stringent IJV structures. Issuer’s aversion for IJV bond

should therefore increase in 2008.8 This motivates the following hypothesis.

8 These standards also required SARA bonds to represent real ownership of the underlying asset (AAOIFI,

2008) while Islamic debt bonds were also criticised that it was expected they should have some link with the

underlying asset (Usmani, 2007).

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

11

H6: Issuer’s aversion for IJV bonds increased in 2008, ceteris paribus.9

3.5 Shariah Advisor Effect

Shariah scholars have different opinions over the Shariah compliance of particular Islamic

bond types. Some scholars consider Islamic debt bonds to be non-Shariah compliant while

others dislike particular IJV bond structures. In contrast, SARA bonds are the least

controversial (Usmani, 2002). The effect of Shariah advisor preferences may be captured by

comparing the bonds approved by Shariah committees against those approved by individual

Shariah advisors. The rationale is that a Shariah committee with multiple members is less

likely to be affected by individual preferences and, so would approve only bonds on which

they all agree. Shariah committees, therefore, should have a higher probability of approving

the less controversial SARA bonds than debt and IJV bonds. The issuer’s choice of Islamic

bond type, hence, should be affected by whether the issuer uses a Shariah committee or

individual advisor. This leads to the following hypothesis:

H7: SARA bonds, in contrast to debt and IJV bonds are more likely to be approved by

Shariah committees than individual advisors.

4. Data

The sample consists of 456 Malaysian corporate Islamic bond issues and covers from 2002 to

2010 while data for banks and other financial institutions is excluded (Julio et al., 2008;

9 In the multivariate probit model issuer’s aversion is measured as the probability of IJV issuance.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

12

Blume et al., 1998).10

The issuing firm’s name, issue date, bond type and Shariah advisor

information were obtained from IFIS (Islamic Finance Information Service). Datastream was

used to extract daily data for stock prices, market to book value ratio of the issuer and Dow

Jones Total Market Index. Compustat was employed for other firm specific financial

variables.

An initial sample of 1166 Islamic bonds from a cross section of countries was obtained from

the IFIS database. Given the focus is on corporate Islamic bonds, 206 sovereign Islamic

bonds were excluded as more than 90% of the remaining sample were Malaysian Islamic

bonds, the study was restricted to Malaysia. Due to missing data, some issues had to be

eliminated leaving 456 Islamic bonds from 83 issuers (Table 1). The sample included 391

debt bonds, 48 IJV bonds and 17 SARA bonds.

<Insert Table 1 about here>

5. Methodology

Multinomial probit model is a generalization of the probit model for comparing multiple

choices (more than two) for unordered data (Greene, 2003; Kumar et al., 2010). It uses a

latent variable to establish a link between the discrete variable and its determinants. The data

described is composed of three types of Islamic bonds IJV, SARA and debt bonds. Equation

1 shows the discrete variable, Y, that represents each of three types of Islamic bonds.

10

The availability of data before 2002 was more limited. For removal of outliers to avoid sample bias refer to

Wooldridge (2009) and Greene (2003).

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

13

(1)

The firm’s choice is determined by firm specific, event specific and Islamic instrument

specific variables. These are captured by the latent utility shown in Equation 2.

(2)

The variables are defined in Table 2 with their expected results in IJV bonds and SARA

bonds given in Tables 3 and 4 respectively. When estimating the model, Islamic debt bond

(with 3) is used as the base case against which the IJV bonds (with 1) and SARA

bonds (with 2) choices are compared. The multinomial probit model in this case

generates two panels (Panel A and Panel B) of results (Table 5). Panel A compares the IJV

and Islamic debt bonds. A positive coefficient for the IJV and Islamic debt bond variables

comparison suggest that higher values should make the issuer prefer debt over IJV bonds. A

negative coefficient, however, implies a preference of IJV over debt bonds. Panel B compares

SARA and Islamic debt bonds. A positive coefficient in Panel B again reflects the fact that

variables’ higher values would cause the issuer prefer Islamic debt over SARA bonds. In

contrast a negative value makes SARA preferred over debt bonds.

<Insert Table 2 about here>

<Insert Table 3 about here>

<Insert Table 4 about here>

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

14

5.1 Measurement of the Variables

This section focuses on how firm specific, event specific and Islamic instrument specific

effects are measured for Equation 2 and the determinants of Islamic bond type. The issuer’s

choice between the three Islamic bond types (IJV, SARA and debt bond) can be analysed at

two stages. The first is when the issuer decides between issuing equity against debt. If certain

issuer’s characteristics favour equity, then IJV bonds are preferred. If the issuer chooses debt

instead, a second stage decision is required on whether to issue SARA bonds or Islamic debt

bonds (Dusuki and Mokhtar, 2010).

The decision to issue IJV bonds might be affected by the issuer’s current position relative to

its target debt to equity ratio (H1) and the causes of any deviation from this target (H2). The

firm’s current debt position is captured by long term debt to assets (LTDA) and total debt to

assets (DTA). A firm with high LTDA and DTA compared to the target would issue less debt

and more of equity type IJV bonds (Marsh, 1982; Hovakimian et al., 2001). The coefficient

therefore, should be negative. The target debt to equity ratio is determined by bankruptcy

risk, firm performance and firm size (Marsh, 1982; Hovakimian et al., 2001). Bankruptcy risk

is captured by the standard error (SE) and market beta (BETA). Higher SE and BETA

increases bankruptcy risk and so reduces the firm’s debt target level. These both should,

therefore, have a positive impact on issuing IJV bonds and hence a negative coefficient in the

model. Firm performance is captured by interest coverage (IC) and operating income to sales

ratios (OIS). These two ratios reflect the firm’s ability to service its existing debt and its

propensity to issue more debt (Marsh, 1982; Hovakimian et al., 2001). Higher IC and OIS

ratios, therefore, should discourage issuing the equity like IJV bonds. Hence, the coefficient

should be positive. Firm size is captured by LA (log of assets). Since larger firms issue more

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

15

debt, a higher LA should result in greater debt and lower IJV bond issuance. The coefficients

should be positive.

Over or undervalued stock, captured by the market to book value ratio (MtB), is an important

determinant in the issuer deviating from the target ratio (Hovakimian et al., 2001). Holding

other variables constant, a low MtB suggests undervalued stock and so would discourage

equity raising, including IJV bonds (Ross, 1977; Myers and Majluf, 1984). MtB, therefore,

should have a negative coefficient in the model.

Highly levered (H3), poor performing firms (H4) with high bankruptcy risk (H5) are

expected to issue secured bonds (Berger and Udell, 1990; Julio et al., 2008). Therefore, a

SARA bond issuer should have higher long term debt (LTDA) and total debt (DTA) ratios.

These variables in the model should be negative (H3). Firm performance is captured by

operating margin (OIS) and interest coverage (IC) ratio. Firms issuing SARA bonds should

have a low level of operating margin (OIS) and interest coverage (IC). These ratios should

have a positive coefficient in the model (H4). Risk is captured by beta (BETA) and standard

errors (SE). SARA bonds should have a high level of SE and beta and hence a negative

coefficient in the model (H5).

The impact of the 2008 AAOIFI recommendations, captured by SS (2008) year dummy, were

particularly critical of prevalent IJV structures and suggested much more stringent

alternatives (Usmani, 2007). So, the number of IJV bond issues should decline in 2008,

implying a positive coefficient for their dummy variables (H6).

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

16

The Shariah advisor effect is captured by a Shariah committee dummy variable. Individual

Shariah advisors may prefer certain type of bonds (Karim, 1990; Grais and Pellegrini, 2006;

Karim and Archer, 2007; Dusuki and Mokhtar, 2010). Shariah committees with multiple

members, however, may reduce such individual preferences in reaching a final decision.

SARA bonds, unanimously agreed upon by scholars as Shariah compliant, should have a

higher probability of being approved by Shariah committees (H7). In contrast, debt bonds and

some structure of IJV bonds are less likely (AAOIFI, 2008; Dusuki and Mokhtar, 2010).

Hence, the coefficient of the Shariah committee dummy variables should be positive for IJV

bonds and negative for SARA bonds.

5.2 Robustness of the Model and Industry Effect

As a firm’s industry might impact on its bond choice, a robustness test of the results is

conducted by running Equation 3 with dummies for three industries: manufacturing, property

and agriculture.11

Given the nature of SARA bonds, they are used frequently in the property

industry as real property can serve as the underlying asset. Therefore, the sign of the property

dummy variable should be negative for SARA bonds.

11

The numbers of observations for SARA bonds issuers belonging to the energy sector are not large enough for

it to be included in the model.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

17

6. Results for Islamic Bond Type Determinant

This section begins by examining firm specific factors that may impact a firm’s Islamic bond

choice, followed by an examination of the impact of new Shariah standards in 2008, and

Shariah advisor effects.

Equation 2 is estimated using multinomial probit analysis with Huber (1964) and White

(1980) robust variance-covariance matrix. The results are generated for IJV and SARA bonds

using Islamic debt bonds as the base case and are reported in Table 5: Panel A for IJV bonds

and Panel B for the SARA bonds. Table 6 also shows the impact of firm specific variables

affecting issuer’s choice of IJV (H1 and H2), SARA and debt bonds (H3, H4 and H5). The

impact of changes in Shariah standards (H6) and Shariah advisor effect affecting issuer’s

choice of Islamic bond type (H7) are similarly reported in Table 5. Finally, the robustness

results after accounting for industry effect by estimating Equation 3 are presented in Table 6.

<Insert Table 5 about here>

<Insert Table 6 about here>

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

18

6.1 Impact of Firm Specific Variables Affecting Issuer’s Choice of IJV

Two types of firm specific variables affect the IJV choice, those that capture the firm’s

current position relative to the target debt to equity ratio (H1) and those that cause deviations

from the target (H2). The results for firm specific variables are reported in Panel A of Table

5, and the corresponding robustness tests, after addressing any industry effect are in Panel A

of Table 6.

H1 tests the impact of firms’ current debt level relative to its target debt to equity ratio. The

current’s position is captured by the total debt to asset ratio (DTA) and long term debt to

asset ratios (LTDA). Holding the target ratio constant, these variables should have a positive

impact (negative coefficient) on issuer’s choice of IJV bonds. According to Table 6, only

DTA is negative and significant at the 1% level. In contrast, LTDA is positive and significant

at the 5% level. This suggests that the long term debt as a proportion of total assets negatively

affects issuer’s choice of IJV while short term debt has a positive impact.12

Therefore, only

the DTA results support H1, while LTDA rejects it. A possible explanation for this is that

Islamic debt bonds have shorter maturities than IJV and SARA bonds. Therefore, holding

target debt to equity ratio constant, higher long term debt ratio (LTDA) should increase the

issuer’s propensity to issue Islamic debt bonds.

The target debt to equity ratio is determined by bankruptcy risk, firm performance and firm

size (Marsh, 1982; Hovakimian et al., 2001). Bankruptcy risk in Equation 2 is captured by

standard error (SE) and market beta (BETA) (expected impact on IJV bonds: positive,

coefficient in the model: negative). Firm performance is reflected in interest coverage (IC)

12

In the presence of long-term debt variable in the regression model, the coefficient of total debt represents the

effect of issuer’s choice assuming long-term debt is constant. Therefore, the total debt coefficient would only

capture the impact of short-term debt (Blume et al., 1998).

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

19

and operating income to sales ratios (OIS) (expected impact on IJV bonds: negative,

coefficient in the model: positive). Firm size addressed by LA (log of assets) (expected

impact on IJV bonds negative, coefficient in the model positive). Table 5 shows that BETA is

significant at the 10% level and has the expected negative coefficient. In contrast, standard

error (SE), which also measures bankruptcy risk, is insignificant. Nevertheless, once the

industry effect is considered, Table 6 shows that SE becomes significant but with an

unexpected positive coefficient. These results further indicate that the log asset (LA) variable

capturing firm size variable, insignificant in the original model in Table 5, becomes

significant at the 1% level with the expected positive sign given the industry effect. Table 5

also reflects that the IC result is contrary to expectations and the wrong sign (negative). The

result for OIS is significant at the 10% with the expected positive coefficient. Panel A of

Table 6 shows that both IC and OIS results are robust to any industry effect. Therefore, it can

be concluded that only the results for DTA, BETA, LA and OIS have the expected sign and

hence partially support H1. In contrast, LTDA, IC and SE reject the hypothesis.

H2 tested whether firms with high market to book value ratio (MtB) prefer IJVs over SARA

or Islamic debt bonds. A low MtB reflects that a stock is undervalued, (Ross, 1977; Myers

and Majluf, 1984), firms with lower market to book value are expected to issue less IJV and

more debt (H2). Panel A of Table 5 and Table 6 show that the MtB is significant at the 10%

level but with the incorrect positive sign. Hence, H2 is rejected. This implies that over or

undervalued stocks do not affect the issuer’s preference for IJV.

The lack of convincing support for H1 and H2 corroborate the assertion that IJV bonds have

little in common with equity and should be treated more like conventional debt bonds

(Usmani, 2007; Ali, 2008).

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

20

6.2 Firm Specific Variables Affecting Issuer’s Choice of SARA Bonds

The issuer’s choice of SARA and debt bonds is determined by variables capturing leverage

(H3), firm performance (H4) and risk (H5). The results for each are reported in Panel B of

Table 5 and their robustness results after accounting for industry effect are reported in Panel

B of Table 6.

The leverage ratios (H3) of long term debt to asset ratio (LTDA) and total debt to asset ratio

(DTA) were both expected to have a positive impact (negative coefficients) on issuer’s

choice of SARA bonds. Panel B of Table 5 and Table 6, however, show that LTDA and DTA

are statistically insignificant. Therefore H3 is rejected.

Firm performance (H4) as captured by operating income to sales ratio (OIS) and interest

coverage ratio (IC) was expected to have a negative impact (positive coefficients) on SARA

bond issues. Panel B of Table 5, though; show that IC is statistically insignificant. In contrast,

OIS is significantly positive at the 5% level. Panel B of Table 6 suggests that these results are

robust to changes in industry effect. IC remains statistically insignificant while OIS become

positively significant at the 1% level. This partially supports H4 suggesting that firms with

higher operating margin prefer debt bonds over SARA bonds.13

The issuer’s risk (H5), as captured by BETA and standard error (SE), was expected to have

positive impact on issuer’s choice for SARA bond. Panel B of Table 5, however, shows that

BETA is statistically insignificant. In contrast, SE is significantly negative at the 1% level.

The results partially support H5 suggesting that issuers with risk proxies of higher standard

13

The results are consistent with the signalling hypothesis (Ross, 1977).

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

21

errors (SE) prefer SARA over debt bonds. Thus firms with high bankruptcy risk may have a

preference for secured SARA bonds.

6.3 Results for Changes in AAOIFI Shariah Standards

The 2008 change in AAOIFI Shariah standards (H6) on instrument choice is captured by SS

(2008) year dummy. In line with expectations, PANEL A of both Tables 5 and 6 shows that

the SS coefficient is positive and statistically significant at the 1% level. This suggests that

holding other things constant, the issuer’s preference for IJV bonds declined in 2008. This

can be attributed to AAOIFI’s tougher Shariah standards which require IJV bonds to be

structured in equity like manner (Usmani, 2007). The results support H6.

6.4 Results for Shariah Advisor Effect

The Shariah advisor effect on the issuer’s choice of Islamic bond type (H7) reflects the

potential for personal preferences in Shariah interpretations (Karim, 1990; Grais and

Pellegrini, 2006; Karim and Archer, 2007; Dusuki and Mokhtar, 2010). The effect is captured

by a Shariah committee (SB) dummy variable, which takes on the value 1 for bond approved

by Shariah committees and 0 for those approved by individual advisors. Panel A of Table 5

shows that the coefficient for SB is significantly positive at the 1% level suggesting that

Shariah committees have an aversion for IJV bonds. This is in line with expectations as some

Shariah scholars have been very critical of the previous IJV structures (Usmani, 2007). The

SB variable for the SARA bonds as shown in Panel B of Table 5, however, is insignificant.

The results partially support the H7 suggestion that issuer choice of IJV bonds is determined

by Shariah advisor preferences.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

22

7. Conclusion

The paper examined the firm specific determinants of issuer choice of Islamic bond type. It

was expected in theory, that an issuer should view IJV bonds as equity instruments (Usmani,

2006). The empirical results however, suggested that issuers do not consider them as equity

replacement in that their choice is not being affected by the normal firm specific determinants

of equity. The impact of firm specific determinants of issuer’s choice of SARA bonds was

then tested. As SARA bonds are supposed to represent ownership of the underlying asset,

they should be more secured than conventional secured bonds. Our results though, implied

that not all SARA bonds in the sample actually represent ownership of the underlying asset

and so are no different from secured conventional bonds. The impact of changes in the 2008

AAOIFI Shariah standard on IJV bonds was then tested and the issuer’s aversion for IJV

bonds was found to have increased. In regards to Shariah committees, issuers disliked the

controversial IJV structures as this preference appears reflected in the issuer’s choice.

In conclusion, the findings suggest that an issuer’s choice of Islamic bond type is not the

same as for conventional bonds. This study extends the academic literature related to the firm

choice of debt to equity ratio to provide evidence that IJV bonds are not viewed by issuers as

equity instruments. Furthermore, it also the first to employ the literature on issuer’s

preference for secured versus unsecured bonds to show that SARA bonds are not different

from secured conventional bonds.

These findings have several implications for regulators, Islamic bond issuers and investors.

For regulators, the findings magnify the lack of Shariah harmonization in the Islamic finance

industry and suggest that for the Islamic finance to prosper policy makers / regulators must

support Shariah harmonization. For issuers, the findings suggest that firm specific variables

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

23

affect an issuer’s choice of Islamic bond type are not unlike that of conventional debt. This

implies that issuers should be more concerned about the bond’s security and seniority rather

than simply their IJV or SARA like structure. Issuers should nevertheless prefer Shariah

compliant structures to protect themselves from changes in Shariah standards and to avoid

conflict with their Shariah committees. For investors they may want to further investigate the

asset backing of SARA bonds before investing.

We would like to acknowledge comments from participants at the 15th

Malaysian Finance

Association conference, Kuala Lumpur, 2-4 June 2013.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

24

References

AAOIFI. (2008). Sukuk Resolution. Retrieved June 13, 2009, from

http://www.aaoifi.com/aaoifi_sb_sukuk_Feb2008_Eng.pdf

Ali, S.S., 2005. Islamic capital market products: Developments and challenges. Research

Paper No 9. Jeddah, Saudi Arabia: Islamic Research and Training Institute, Islamic

Development Bank.

Ali, S.S., 2008. Islamic capital markets: Products, regulation and development. Seminar

Proceedings. Jeddah, Saudi Arabia: Islamic Research and Training Institute, Islamic

Development Bank.

Ayub, M., 2007. Understanding Islamic Finance. Chichester, UK: John Wiley and Sons.

Bayless, M., Chaplinsky, S., 1993. Seasoned equity issuance in hot and cold markets.

Journal of Finance 48, 1062-1062.

Berger, A., Udell, G., 1990. Collateral, loan quality and bank risk. Journal of Monetary

Economics 25, 21–42.

Blume, M.E., Lim, F., MacKinlay, A.C., 1998. The declining credit quality of us corporate

debt: Myth or reality? Journal of Finance 53, 1389-1413.

Dusuki, A.W., Mokhtar, S., 2010. Critical appraisal of Shariah issues on ownership in asset-

based sukuk as implemented in the Islamic debt market. Research Paper No 8. Kuala

Lumpur, Malaysia: International Shariah Research Academy on Islamic Finance.

Flannery, M.J., 1986. Asymmetric information and risky debt maturity choice. Journal of

Finance 41, 19-37.

Graham, J.R., 1996. Debt and the marginal tax rate. Journal of Financial Economics 41, 41-

73.

Grais, W., Pellegrini, M., 2006. Corporate governance in institutions offering Islamic

financial services: Issues and options. Policy Research Working Paper 4052. Washington,

DC: World Bank, Financial and Private Sector Development Network.

Greene, W.H., 2003. Econometric Analysis, sixth edition. Upper Saddle River, New Jersey,

Prentice Hall.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

25

Hovakimian, A., Opler, T., Titman, S., 2001. The debt-equity choice. Journal of Financial

and Quantitative Analysis 36, 1-24.

Howladar, K., 2006. Shariah and Sukuk: A Moody’s Primer. International Structured

Finance Special Comment. New York, Moody’s Investors Service.

Jensen, M.J., Meckling, J.W., 1976. Theory of the firm: Managerial behavior, agency cost,

and ownership structure. Journal of Financial Economics 3, 305-360.

Julio, B., Kim, W., Weisbach, M.S., 2008. What determines the structure of corporate debt

Issues? Unpublished working paper. Ohio State University, Charles A. Dice Centre for

Research in Financial Economics.

Karim, R.A., 1990. The independence of religious and external auditors: The case of Islamic

banks. Accounting, Auditing and Accountability 3, 34-44.

Karim, R.A., Archer, S., 2007. Islamic Finance: The Regulatory Challenge. Singapore: John

Wiley and Sons.

Kumar, P., Chava, S., Warga, A., 2010. Managerial agency and bond covenants. Review of

Financial Studies 23, 1120-1148.

Marsh, P., 1982. The choice between equity and debt - an empirical-study. Journal of

Finance 37, 121-144.

Modigliani, F., Miller, M., 1958. The cost of capital, corporation finance and the theory of

investment. American Economic Review 48, 261–297.

Myers, S.C., 1977. Determinants of corporate borrowing. Journal of Financial Economics 5,

147-176.

Myers, S.C., Majluf, N.S., 1984. Corporate financing and investment decisions when firms

have information that investors do not have. Journal of Financial Economics 13, 187-221.

Ross, S.A., 1977. The determination of financial structure: The incentive-signalling

approach. The Bell Journal of Economics 8, 23-40.

Salah, O., 2010. Islamic finance: The impact of the AAOIFI resolution on equity-based

sukuk structures. Law and Financial Markets Review 4, 507-517.

Stulz, R.M., 1990. Managerial discretion and optimal financing policies. Journal of Financial

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

26

Economics 26, 3-27.

Stulz, R.M., Johnson, H., 1985. An analysis of secured debt. Journal of Financial Economics

14, 501-521.

Usmani, T.M., 2002. An Introduction to Islamic Finance. Hague, Netherlands: Kluwer Law

International.

Usmani, I.A., 2006. Shirkat-o-Mudaradat , Asr-e-hazir Mein. Karachi, Pakistan: Idaratul

Maarif.

Usmani, M.T., 2007. Sukuk and their contemporary applications. AAOIFI

Working Paper No n/a. Retrieved June 13, 2009, from

http://www.failaka.com/downloads/Usmani_SukukApplications.pdf

White, H., 1980. A heteroskedasticity-consistent covariance matrix estimator and a direct

test for heteroskedasticity. Econometrica 48, 817-838.

Wooldridge, J.M., 2009. Introductory Econometrics: A Modern Approach, fourth edition.

Mason, OH, South Western, Cengage Learning.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

27

Table 1: Issues by Industry 2002-2010

Industry Total

Manufacturing 144

Property and infrastructure development 102

Energy (electricity, gas and oil) 55

Agriculture, food and palm oil 86

Others* 21

Total sample 456

*These include shipping, transportation and technical

services (IFIS database).

Note the data is for corporate Malaysian Islamic

bond issues.

Source: IFIS data.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

28

Table 2: Variables for the Issuer’s Choice of Islamic Bond Type Model

This table details variable measurement for Equation 2:

where is the latent utility function.

Variable Description Measurement

LTDA Leverage Ratio I Long term debt / total assets

DTA Leverage Ratio II Long term debt + debt in current liabilities / total assets

BETA Beta Market beta

SE Standard Error

IC Interest Coverage (Operating income after depreciation + interest income) /

interest expense

OIS Operating Margin Operating income before depreciation / sales

LA Firm Size Log of total assets

MtB Stock Valuation Market value / net total assets

SS Shariah Standards Dummy variable that takes the value 1 for bonds issued in

2008

GFC Global Financial

Crisis

Dummy variable that takes the value 1 for bonds issued in

2009 and 2010

SB Shariah Committee Dummy variable that takes the value of 1 for Shariah

Committees

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

29

Table 3: Expected Impact of Variables in the Issuer’s Choice of IJV Bond Type Model This table provides the expected impact, hypothesized coefficient and associated hypotheses related to variables affecting issuer’s

choice of IJV bonds for Equation 2*:

where is the latent utility function.

Variable Effect Expected Impact

for IJV

Hypothesized Coefficient

Sign for IJV

Hypotheses

LTDA Leverage Ratio I + - H1: Higher current debt, relative to

targeted debt, results in firms

preferring IJV over SARA and debt

bonds (and vice versa).

DTA Leverage Ratio II + -

BETA Beta + -

SE Standard Error + -

IC Interest Coverage - +

OIS Operating Margin - +

LA Firm Size - +

MtB Stock Valuation + - H2: Holding other firm specific

variables constant, firms with higher

market to book value should prefer

IJV bonds over SARA and debt

bonds (and vice versa).

SS Shariah Standards - + H6: Issuer’s aversion for IJV bonds

increased in 2008, ceteris paribus.

SB Shariah Committee - + H7: SARA bonds, in contrast to debt

and IJV bonds are more likely to be

approved by Shariah committees

than individual advisors.

*GFC dummy variable included in Equation 2 is not discussed here because it is not related to any of the hypothesis. It is instead

used as a control variable to capture the effect of the GFC.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

30

Table 4: Expected Impact of Variables in the Issuer’s Choice of SARA Bond Type

Model This table provides the expected impact, hypothesized coefficient and associated hypotheses related to variables affecting issuer’s

choice of SARA bonds for Equations 2:

where is the latent utility function.

Variable Description Expected

Impact for

SARA

Hypothesized

Coefficient Sign for IJV

Hypotheses

LTDA Leverage Ratio I + - H3: Firms with high leverage (total

and long-term debt) prefer SARA

bonds over debt bonds and vice versa. DTA Leverage Ratio II + -

IC Interest Coverage - + H4: Firm having high profitability

(operating margin and interest

coverage ratio) prefer debt over

SARA bonds and vice versa.

OIS Operating Margin - +

BETA Beta + - H5: Firms having higher risk (market

beta and standard error) prefer SARA

bonds over debt bonds and vice versa. SE Standard Error + -

LA Firm Size - +

MtB Stock Valuation

2008 Shariah Standards + -

SB Shariah Committee + - H6: SARA bonds, in contrast to debt

and IJV bonds are more likely to be

approved by Shariah committees than

individual advisors.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

31

Table 5: Issuer’s Choice of Islamic Bond Type Model Results This table presents the estimation of the multinomial probit model show in Equation 2:

where, subscripts i denote individual bond issues. captures bond type. is the latent variable that links the

dependent variable with . DTA is total debt to asset ratio. LTDA is long term debt to asset ratio. BETA is the

market model beta. SE is standard errors. IC is interest coverage ratio. OIS is operating income to sales ratio. LA

is log of assets for 2002 - 2010. MtB is market to book value ratio. SS is a dummy that equals 1 for bonds issued

in 2008. GFC is a dummy variable that equals 1 for bonds approved in 2009 and 2010. SB is a dummy variable

that equals 1 for bonds approved by Shariah committees. In the result shown in this table debt bonds is used as

the base case. PANEL A provides the results for IJV bonds. Panel B provides the results for SARA bonds.

Adjusted P-values (Prob. Adj.) using Huber-White test are also provided.

Issue Coef. Std. Err. Z-Stat Prob. Prob.(Adjusted)

PANEL A (Issuers’ choice of IJV bonds vs. Islamic debt bonds)

DTA -8.27915 2.751596 -3.01 0.003 0

LTDA 6.113098 2.813355 2.17 0.03 0.035

BETA -1.09855 0.655874 -1.67 0.094 0.095

SE -33.8354 25.12348 -1.35 0.178 0.42

IC -0.12308 0.055935 -2.2 0.028 0.026

OIS 2.78282 1.577453 1.76 0.078 0.085

LA 0.069683 0.173934 0.4 0.689 0.758

MtB 0.273912 0.150924 1.81 0.07 0.099

SS 2.566097 0.629726 4.07 0 0

GFC 0.099917 0.43109 0.23 0.817 0.835

SB 2.10014 0.541412 3.88 0 0

_cons -0.43129 1.743708 -0.25 0.805 0.839

PANEL B (Issuers’ choice of SARA bonds vs. Islamic debt bonds)

DTA -2.68262 3.510771 -0.76 0.445 0.297

LTDA 3.743863 3.315118 1.13 0.259 0.125

BETA 0.841531 0.823889 1.02 0.307 0.15

SE -125.358 48.61427 -2.58 0.01 0.001

IC 0.005254 0.014146 0.37 0.71 0.641

OIS 2.987766 1.782678 1.68 0.094 0.031

LA -0.2187 0.205182 -1.07 0.286 0.301

MtB 0.165214 0.180923 0.91 0.361 0.304

SS 0.270215 0.887627 0.3 0.761 0.722

GFC -0.98672 0.545146 -1.81 0.07 0.034

SB -0.49824 0.561417 -0.89 0.375 0.379

_cons 1.229856 2.097052 0.59 0.558 0.582

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

32

Table 6: Industry Effect Results This table presents the estimation of the multinomial probit model shown in Equation 3.

where, subscripts i denote individual bond issues. captures bond type. is the latent variable that links the

dependent variable with . DTA is total debt to asset ratio. LTDA is long term debt to asset ratio. BETA is the

market model beta. SE is standard errors. IC is interest coverage ratio. OIS is operating income to sales ratio. LA

is log of assets. MtB is market to book value ratio. SS is a dummy that equals 1 for bonds issued in 2008. GFC

is a dummy variable that equals 1 for bonds approved in 2009 and 2010. SB is a dummy variable that equals 1

for bonds approved by Shariah committees. Property is a dummy variable that equals 1 if the issuer belongs to

the property industry. Manufacture is a dummy variable that equals 1 if the issuer belongs to the manufacturing

industry. In the result shown in the table debt bonds is used as the base case. PANEL A provides the results for

IJV bonds. Panel B provides the results for SARA bonds. Adjusted P-values (Huber-White test) are provided.

Coef. Std. Err. Z-Stat Prob. Prob.

(Adjusted)

PANEL A (IJV bonds vs. Islamic debt bonds)

DTA -15.7001 4.173263 -3.76 0 0

LTDA 9.992165 3.727618 2.68 0.007 0.029

BETA -1.45586 0.764833 -1.9 0.057 0.012

SE 37.99625 22.47332 1.69 0.091 0.04

IC -0.14879 0.054958 -2.71 0.007 0

OIS 3.555497 1.800129 1.98 0.048 0.079

LA 0.529083 0.216547 2.44 0.015 0

MtB 0.329576 0.171572 1.92 0.055 0.053

SS 2.860449 0.759338 3.77 0 0

GFC -0.45045 0.48809 -0.92 0.356 0.292

SB 3.166288 0.684501 4.63 0 0

Manufacture 2.738119 0.823975 3.32 0.001 0

Property 1.905076 0.665876 2.86 0.004 0.012

Agriculture 3.961999 1.018763 3.89 0 0

_cons -5.99666 2.242103 -2.67 0.007 0

PANEL B (SARA bonds vs. Islamic debt bonds) DTA -3.09623 4.194972 -0.74 0.46 0.359

LTDA 4.191526 3.909983 1.07 0.284 0.14

BETA 0.824404 0.875921 0.94 0.347 0.245

SE -83.043 46.79482 -1.77 0.076 0.007

IC 0.004852 0.015165 0.32 0.749 0.661

OIS 5.655009 2.259231 2.5 0.012 0.002

LA -0.08901 0.224404 -0.4 0.692 0.649

MtB 0.276328 0.191515 1.44 0.149 0.104

SS 0.289479 1.052435 0.28 0.783 0.772

GFC -1.193 0.638267 -1.87 0.062 0.041

SB -0.51802 0.686602 -0.75 0.451 0.412 Manufacture 0.063251 0.840112 0.08 0.94 0.918

Property -1.58731 1.178701 -1.35 0.178 0.114

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

33

Agriculture 1.253038 0.812069 1.54 0.123 0.089

_cons -1.42471 2.718931 -0.52 0.6 0.549

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

34

Issuer’s Choice of Islamic Bond Type

Highlights

Using conventional bond theory it is expected that firms would view IJV bonds as

equity.

Results suggest that issuers do not consider IJV bonds as equity equivalents.

SARA bonds did not appear to always represent the ownership of the underlying asset

as they should.

The 2008 AAOIFI Shariah standard on IJV bonds seem to have led to an aversion on

their issuance.

Issuers using Shariah committees as opposed to individual advisors avoided IJV bond

issuance.