Amsterdam Business School
The influence of debt ratio and equity ratio on sukuk issuances
and the payment of interest
Name: Mohamed Talhaoui
Student number: 10548653
Thesis supervisor: Vincent O’ Connell
Date: 15 August 2016
Word count: 13265
MSc Accountancy & Control, specialization Accountancy Faculty of Economics and Business, University of Amsterdam
Statement of Originality
This document is written by student Mohamed Talhaoui who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
ABSTRACT: This study examines the influence of debt ratio and equity ratio on sukuk issuances and the likelihood of payment of interest. More specifically: first, I examine whether companies with a low debt ratio will have a higher equity/sukuk ratio than companies with a high debt ratio. Next, I examine whether companies with a low debt ratio issue a higher amount of sukuk than companies with a high debt ratio. Finally, I examine whether companies with a high equity ratio will borrow less and issue asset-backed sukuk and thus will pay dividend instead of interest; while companies with a low equity ratio will borrow more and thus pay interest. Using data from the London Stock Exchange, I find no evidence that companies with a low debt ratio, are more likely to have more equity with respect to the amount of sukuk they want to issue than companies with a high debt ratio. There is also no evidence that companies with a low debt ratio, are more likely to issue a higher amount of sukuk than companies with a high debt ratio. I could also find no evidence that companies with a high equity ratio are less likely to pay a fixed periodic distribution amount (interest) than companies with a low equity ratio.
Keywords: asset-based sukuk; asset-backed sukuk; sukuk issuance; debt ratio; equity ratio; equity/sukuk ratio.
Preface and acknowledgements
I want to thank Dr. Ir. Sander van Triest for his time, patience and effort during the initial fase of this Msc thesis. Special thanks go to Vincent O’Connell for his time, patience and insight in how to design this MSc thesis. I also want to thank my fellow student Hicham el Kaddouri for his time, contribution, patience and useful advices to write this Msc thesis.
During my search for scientific articles, I noticed that it took a large amount of time to find scientific articles on the topic “Islamic finance and accounting”. Much material on this topic is not available.
With this MSc thesis I want to contribute to the academic literature by (indirectly) extending the knowledge on the topic “Islamic finance and accounting”. By writing this MSc thesis I want to pave the way for future research on this topic.
Mohamed Talhaoui Amsterdam, August 2016
Contents
1 Introduction ... 6
2 Theoretical background ... 9
2.1 Islam ... 9
2.2 General information about sukuk ... 11
2.3 Sukuk market... 13
3 Hypotheses and empirical models ... 15
3.1 Debt ratio ... 15
3.2 Fixed periodic distribution amount (interest) ... 18
3.3 Empirical models... 21
4 Research Methodology ... 24
4.1 Overview of company data ... 26
5 Empirical results ... 28
5.1 Descriptive Statistics and Bivariate Correlations ... 28
5.3 Main findings ... 37
6 Conclusion ... 51
7 Limitations of the study ... 52
7.1 Contribution and future research ... 53
8 References ... 54
Appendix 1A ... 58
1 Introduction
Islamic finance is, according to the European Central Bank, on the rise in Europe (European Central Bank, 2013). The financial crisis, which started in 2008 with the collapse of Lehman Brothers, renewed the debate on the role that Islamic finance can play in the stabilization of the global financial system (European Central Bank, 2013). Globally, 20% of the people is muslim and yet Islamic compliant assets make up approximately 1% of the world’s financial assets (TheCityUK, 2015). Therefore considerable potential exist for expansion of Islamic finance (TheCityUK, 2015). At once restricted to the Middle East and Southeast Asia, today it pervades the markets of Europe, United States and Asia. Islamic compliant finance and investment are increasingly becoming part of the world economy as money of Muslims around the world are increasingly integrated into global capital markets (Abdel-Khaleq & Richardson, 2007). Over the last decade, Islamic banking has globally experienced an annual growth rate of 10 to 15 percent (Solé, 2007). According to Zakaria, Md Isa, & Abidin (2012), the Islamic financial services sector grew the last decade with more than 10 per cent on an annual basis and has, as for 2009, assets estimated to be worth USD 700 billion globally. Islamic finance is not only limited to countries with a majority of Muslim population, it is also expanding in other countries where Muslims are the minority. Examples are the United Kingdom and Japan. According to Solé (2007) conventional banks are increasingly becoming interested in entering the market for Islamic financial products. This is not only because of the financial crisis, which started in 2008, but also because of the growing demand by the Muslim population in Western countries and Islamic countries (mainly from the Gulf region) that want to diversify their portfolio.
Sukuk or Islamic bonds, as many people call it, is an essential part of Islamic finance (TheCityUK, 2015). Sukuk are not only issued by Islamic companies/institutions but also by non-Islamic companies/institutions. General Electric became in 2009 the first major US company that issued sukuk (Slater & Haywood, 2009). In 2014 the United Kingdom became the first Western country that issued sovereign sukuk (TheCityUK, 2015; World Islamic Economic Forum, 2013).
A distinguishing feature of sukuk with respect to conventional bonds is that they are backed by assets. This is the essence of sukuk: “representing an ownership interest in the underlying asset” (Zakaria, Md Isa, & Abidin, 2012, p. 662). However, the present sukuk market places sukuk in two groups: asset-based sukuk and asset-backed sukuk (Abdullah, 2011; Hidayat, 2013). Both structures impact the level of debt as well as the level of equity. The new developed structure (asset-based) gives rise to new challenges and concerns (Alqahtani, 2012). What is its impact on the solvency and the relation between equity and debt of the company? Will companies treat this type of sukuk as debt because they are no longer obliged to transfer assets to holders of sukuk? Will companies issue an amount of sukuk that is higher than its total equity? In practice it becomes clear that many companies prefer asset-based sukuk because they do not want to sell their assets to investors, which can lead to a decrease in equity versus total assets (Abdullah, 2011).
However, Alqahtani (2012) advocates minimizing, if not eradicating the use of asset-based sukuk. According to Alqahtani (2012) the use of asset-asset-based sukuk is against the teachings of Islam. Instead, Alqahtani (2012) advocates the replacement of asset-based sukuk asset-backed sukuk. This means that the total of equity is at least equal to the amount of sukuk issued, which also can lead to a lower debt ratio. The assed-based structure is still dominant in the market (Hidayat, 2013; Abdullah, 2011). Approximately 90 per cent of all sukuk can be treated as asset-based sukuk. What is the influence of these two structures on the solvency, debt ratio and equity versus amount of sukuk issued by a company?
Another distinguishing feature of sukuk with respect to conventional bonds is the absence of payment of interest. In Islam it is forbidden to lend money at interest (Lewis, 2001; Pomeranz, 1997; Vinnicombe, 2010). Holders of sukuk receive a share of the profit instead of interest. However, Zakaria et al. (2012) mention that in the present day sukuk may be treated as either a debt or equity instrument. According to Zakaria et al. (2012) the treatment of sukuk as a debt instrument affects the possbility of payment of interest, while according to Abdullah (2011) treatment of sukuk as a debt instrument is the same as replicating the conventional system. Abdullah (2011) argues too that the amount of debt that a company holds, affects the possibility of paying interest. The higher the debt, the more likely the company will pay interest.
An issuer with too much debt would technically be barred from issuing sukuk (Abdel-Khaleq & Richardson, 2007). The question that rises is whether the company’s level of equity ratio will influence the likelihood of paying interest.
To my best knowledge no previous study answered the above mentioned questions from an empirical perspective. Therefore, this study tries to answer the following research question: “What is the influence of debt ratio and equity ratio on sukuk issuances and the
payment of interest?”.
The remainder of this study is organized as followed: first, I discuss the theoretical background of sukuk, followed with the development of hypotheses and empirical models. The fourth section pays attention to the research methodology, while the fifth section discusses the empirical results. The sixth section concludes, while the seventh section pays attention to the limitations of the study, contribution to academic literature and future research.
2 Theoretical background
2.1 Islam
While the West has a separation of Church and state, this is not the case with Islam where all behaviour undertaken by Muslims, including business affairs, is dictated by Islam. Islam encourages honest commercial activity and forbids certain commercial activity that are common in other parts of the world. The most important and famous example is the ban on the charging of interest or riba as it is called in Arabic. Interest can be seen as an unjustified monetary advantage received by a party, without giving a counter-value in return.
The Shari’a (Islamic law) consists of the Quran, the Sunna (deeds and sayings of the Prophet Muhammad) and Fiqh (Islamic jurisprudence). Important to note is that four madhahib (Islamic schools of understanding) exists in Islam. Each of these four schools have their own geographical area of dominance. These schools have a broad consensus on the main prohibitions within Islamic law (Ang, 2013). Logically, there are also differences between these four schools which makes discussion about for example sukuk possible.
Over the years a number of different investment vehicles have been developed and approved by Islamic scholars. It must be noted that not all structures are accepted by all the four schools of Islamic jurisprudence (Abdel-Khaleq & Richardson, 2007). It is important to acknowledge the fact that Islamic financial products that are permissible in some countries, could be impermissible in other countries. This is not only because supervisory bodies in different countries tend to rely on their own Islamic law experts (who have their own way of interpreting things) but also because of the four schools of Islamic jurisprudence (Solé, 2007). According to Solé (2007) there are more than 300 Islamic financial institutions spread over 51 countries and 250 mutual funds that comply with Islamic principles. Conventional banks are increasingly becoming interested in entering the market for Islamic financial products. This is not only because of the financial crisis, which started in 2008, but also because of the growing demand by the Muslim population in Western countries and Islamic (mainly from the Gulf region) to diversify their portfolio (Solé, 2007).
conservative and are more guided by Islamic law. The structure of sukuk needs to adapt to the laws and regulation of that specific country. This raises the question whether sukuk is still Islamic law compliant (Abdel-Khaleq & Richardson, 2007).
There are two types of scholars: critical and constructivist. The first group argues that current Islamic financial products like sukuk cannot be truly Islamic without a revision of the interpretation methodology of the Islamic law. The second group seeks to build conventional oriented, but Islamic law oriented, financial products based on the present interpretation methodology (Ang, 2013). The second group gets support from Investment bankers, who state that Islamic law requirements achieve the same end result that conventional investment or finance products would achieve in a number of situations (Abdel-Khaleq & Richardson, 2007). According to Abdel-Khaleq & Richardson (2007), only a small number of Islamic scholars combine an in-depth knowledge of Islam with a good understanding of the conventional economy.
2.2 General information about sukuk
While conventional bonds play an major role in the Western financial economy in raising capital, sukuk play an major role in Islamic capital markets (Sukor, Muhamad, & Gunawa, 2008; Hidayat, 2013). Longman & Longman (2000) define bonds as certificates or documents of debt that are issued by a government or an organization for an amount of money they borrow from the bondholders, with the promise to pay them back the amount of money plus interest. In contrast, sukuk are defined by the AAOIFI as certificates of equal value representing undivided shares of ownership in tangible assets, usufruct and services in the ownership of particular projects or special investment activity (Mohamed, Masih, & Bacha, 2015; Zakaria et al., 2012, p. 662). Sukuk means deed or instrument and is the plural of the Arabic word sakk (Ang, 2013).
From the above we can conclude that fundamental differences exist between conventional bonds and sukuk. According to Zakaria et al., (2012) and the Islamic Finance and Sukuk Company (2014) the differences are:
- Bonds indicate a debt obligation while sukuk indicate ownership of an asset;
- Assets backing bonds may be prohibited in Islam because they contain products or services that are against its teachings, while assets that back sukuk are compliant with the rules of Islam;
- Bond pricing is based on credit rating while sukuk are priced according to the value of the assets backing them;
- Bonds contain the component interest which is prohibited in Islam, while sukuk are interest free. As a result bonds rise in value when the interest rises and sukuk increase in value when the assets increase in value;
- The sale of bonds means the sale of debt. The sale of sukuk means the selling of ownership in the assets backing them;
Important to note is that there are different types of sukuk. Each of them has its own characteristics. The AAOIFI identifies at least fourteen possible types (Mohamed et al., 2015). It is not necessary to mention all the fourteen possibilities, because not all fourteen types are important for my research.
The most important types are Mudaraba sukuk, Musharaka sukuk, Murabaha sukuk, Ijara sukuk, Istisna sukuk and Salam sukuk (Pomeranz, 1997; Abdul Rahman, 2003; Dar Al Istithmar, 2006). Of this six important types only two types are allowed to be traded on the secondary market (Dar Al Istithmar, 2006). These are the Mudaraba sukuk and Ijara sukuk (Dar Al Istithmar, 2006). For my research I will focus on sukuk that can be traded on the secondary market. Therefore it deserves attention to briefly explain the above two mentioned types.
Mudaraba sukuk: an agreement between two parties whereby one of the two parties acts as the capital provider and the other party acts as the one who will work with the provided capital (Mudarib). Profit and loss is to be shared between the parties according to a pre-agreed ratio. The Mudarib is the one who issues mudaraba certificates and the subscribers to these certificates are the capital providers (Pomeranz, 1997; Abdul Rahman, 2003; Dar Al Istithmar, 2006).
Ijara sukuk: a contract to which a party purchases and leases out equipment required by the client, in exchange for a rental fee. A ijara sukuk represents ownership of tangible assets that are tied up to a lease contract. This sukuk can be traded in the market at a price determined by market forces (Pomeranz, 1997; Abdul Rahman, 2003; Dar Al Istithmar, 2006).
The process of structuring sukuk deals is for more complicated outside the Islamic world, because those who deal with it are less familiar with this phenomenon than those who live inside the Islamic world (Pomeranz, 1997). The lack of knowledge about the Islamic law is an important issue. Law and regulation in the Western world do not necessarily align with Islamic law principles, making the process of structuring sukuk more difficult. Think about the fact that in the West it is allowed to invest in companies who produce alcohol or make gambling possible. Western companies that want to issue sukuk need the knowledge and expertise of an Islamic scholar. Management accounting and SPV’s (special purpose vehicle) are a vital ingredient in sukuk issuances (Solé, 2007).
2.3 Sukuk market
Sukuk, in essence an asset-backed security and somewhat similar to a trust certificate or bond, is witnessing an increase in popularity as an investment product. Sukuk is an attractive investment for Islamic banks and Islamic insurance companies that cannot invest in conventional sercurities because of the involvement of riba (interest). But also for Western banks and companies that want diversification in their portfolio. Despite the fact that sukuk were first issued in the 1980s, almost all growth has come within the past decade (Godlewski, Turk-Ariss, & Weill, 2013). Sukuk is becoming more popular, not only in Islamic countries but also in non-Islamic countries. In the year 2004 alone sovereign and corporate sukuk issuance rose at a rate of 300% (Dar Al Istithmar, 2006). Saxony Anhalt, part of Germany, was in 2004 the first provincial government that offered sukuk to investors. It raised EUR 100 million from European and Middle East investors (Abdel-Khaleq & Richardson, 2007). Sukuk has even been issued in the United States. In July 2006 a USD 165.7 million sukuk backed by oil and gas assets in the Gulf of Mexico was offered, marking the first instance of a sukuk (Abdel-Khaleq & Richardson, 2007). The authors (2007) even argue that sukuk is compatible with the US oil and gas law. It can contribute to the growth potential of the energy industry, as well as additional possibilities for Muslim capital investors. The annual sukuk issuances almost multiplied by three from USD 45 billion in 2011 to USD 118.8 billion in 2014 (Bank Negara Malaysia, 2015). According to Malaysia International Islamic Financial Centre (2011) the overall sukuk issuance volume increased to USD 179 billion at the end of Q2 2011 compared to 2010. The global sukuk market grew approximately 40 per cent year-by-year to the end of May 2012 (Kuwait Finance House, 2012).
As mentioned before, sukuk is an attractive source of capital for issuers outside the Muslim world. In the Islamic world it is Bahrain, Malaysia and the United Arab Emirates that compete with each other to grab the leading position in the sukuk market. In the Western world the leading position has been taken by London (Abdel-Khaleq & Richardson, 2007). While Malaysia has been recognized to be at the forefront of Islamic finance internationally (Mohamed et al., 2015), the British Prime Minister David Cameron announced on October 29th 2013 that the City of London has the intention to become home to the Islamic Finance market outside the Islamic world (World Islamic Economic Forum, 2013). London Stock
2015). Other landmark issuances followed from Hong Kong, Senegal, South Africa and Luxembourg (Bank Negara Malaysia, 2015).
3 Hypotheses and empirical models
3.1 Debt ratio
A distinguishing feature of sukuk with respect to conventional bonds is that they are backed by assets. This is the essence of sukuk: “representing an ownership interest in the underlying asset” (Zakaria et al., 2012, p. 662). With the rising popularity of sukuk it has become part of or at least an alternative to the conventional system. Because of its use in the conventional system (by Western and non-Western entities) a new structure was developed. Sukuk can now be placed in two groups: asset-based sukuk and asset-backed sukuk (Abdullah, 2011; Hidayat, 2013). Asset-based sukuk means that the underlying asset stays on the balance sheet of the issuer of sukuk i.e. the issuer transfers beneficial ownership of the asset but retains legal ownership of the asset (Hidayat, 2013; Lewis, 2001; Godlewski et al., 2013). One of the consequences of this structure is that sukuk holders cannot sell the underlying asset to a third party (Hidayat, 2013). In contrast, the asset-backed sukuk does not stay on the balance sheet of the issuer of sukuk but is separated and is normally transferred to a Special Purpose Vehicle (Hidayat, 2013; Lewis, 2001; Godlewski et al., 2013). Another important distinctive difference between asset-based sukuk and asset-backed sukuk is the source of payment. While payment in case of asset-based sukuk comes from the issuer’s cash flows, the source of payments in case of asset-backed sukuk is generated from the underlying asset (Hidayat, 2013; Abdullah, 2011). Table 1 provides a summary of the above mentioned facts.
Table 1: Difference between asset-backed and asset-based
Asset-based sukuk Asset-backed sukuk
Does the underlying asset remain on the balance sheet?
underlying asset stays on the balance sheet of the issuer of sukuk.
Underlying asset does not stay on the balance sheet of the issuer of sukuk but is separated.
Source of payments Comes from the issuer’s cash
flows.
Generated from the underlying asset.
Asset-backed sukuk and asset-based sukuk are also known as respectively “equity sukuk” and “debt sukuk” (Alqahtani, 2012). Both structures impact the level of debt as well as the level of equity. The new developed structure (asset-based) gives rise to new challenges and concerns (Alqahtani, 2012). What is its impact on the solvency and the relation between equity and debt of the company? Will companies treat this type of sukuk as debt because they are no longer obliged to transfer assets to holders of sukuk? Will companies issue an amount of sukuk that is higher than its total equity? From conventional bonds we know that the capital structure (i.e. debt versus total assets and equity versus total assets) of the company has important effects on the prices of bonds (Dumitrescu, 2007, p. 109). Alqahtani (2012) addresses these questions but does not provide empirical evidence for it.
Abdullah (2011) argues that holders of asset-backed sukuk face less risk. More specifically, they face asset risk instead of originator risk. If the underlying assets fail to pay dividends, sukuk holders can sell the underlying asset and recover their initial investment. Holders of asset-based sukuk however, are not entitled to the underlying assets but, like holders of conventional bonds, entitled to the cash flows. However, the vast majority of sukuk do not have proprietary rights but instead have beneficial ownership. As a result most sukuk (asset-based) have been structured to enable originators to legally retain ownership of the underlying assets. In practice it becomes clear that many companies prefer asset-based sukuk because they do not want to sell their assets to investors, which can lead to a decrease in equity versus total assets (Abdullah, 2011). A decrease in equity versus total assets means automatically an increase in debt versus total assets. From research on the conventional bond we know that not selling assets to investors, increases the liquidation value of the company and pledging these assets reduces the default probability (Morellec, 2001, p. 200-201).
In contrast, Alqahtani (2012) advocates minimizing, if not eradicating the use of asset-based sukuk. The use of asset-based sukuk leads people to using the term “Islamic bonds” for sukuk, which is one of the major misconceptions about this instrument (Alqahtani, 2012). Because of the use of this term, companies are likely to treat this instrument as (conventional) debt, which is against the teachings of Islam (Alqahtani, 2012).
Using the asset-backed sukuk instead of the asset-based means that the total of equity is at least equal to the amount of sukuk issued and it can also lead to a lower debt ratio. Alqahtani (2012) provides no empirical evidence for his statement. Hidayat (2013) argues that asset-backed sukuk are more compliant with the Islamic law than asset-based sukuk, because they meet the demands set by the Islamic law in both form and substance. A company that issues an asset-backed sukuk will have the necessary assets to back it up. Hidayat (2013, p. 27) views asset-based sukuk as an equivalent to the conventional bond.
Despite the critics of many Islamic scholars, the assed-based structure is still dominant in the market (Hidayat, 2013; Abdullah, 2011). Approximately 90 per cent of all sukuk are – although linked to an underlying asset – not secured instruments and should not be treated as such.
Based on the above mentioned literature we conclude that an asset-backed sukuk requires the availability of underlying assets for a true sale and an asset-based sukuk does not require this. We also conclude that the sukuk structure influences the solvency, debt ratio and equity versus amount of sukuk issued by a company. Therefore, we predict that companies with a low debt ratio will have more equity versus the amount of sukuk they want to issue than companies with a high debt ratio. Furthermore we predict that companies with a low debt ratio issue a higher amount of sukuk than companies with a high debt ratio.
More formally:
H1: Companies that have a low debt ratio are more likely to have more equity with respect to the amount of sukuk they want to issue than companies that have a high debt ratio.
H2: Companies that have a low debt ratio are more likely to issue a higher amount of sukuk than companies with a high debt ratio.
3.2 Fixed periodic distribution amount (interest)
The incentive for investors to hold sukuk is that they receive a share of the profit. The incentive for investors to hold bonds is that they receive a fixed amount of payment (interest) periodically no matter how well or poorly the business may perform. It is forbidden in Islam to lend money at interest. In contrast, earning profit from trading or other business activities is not forbidden; in fact it is encouraged (Abdullah, 2011). With sukuk, investors are exposed to the possibility of profits as well as losses, which is in line with the profit and loss sharing principle of Islamic finance. Sukuk are sometimes portrayed as “Islamic compliant alternatives” to fixed income, conventional bonds and debt securities (Abdullah, 2011). While a conventional bond is clearly a debt instrument, sukuk may be a debt or equity instrument (Zakaria et al., 2012). Because of this classification we distinguish between two types of transactions: interest-based based/debt sukuk) and profit-sharing (asset-backed/equity sukuk). In case of the first one profit is predetermined, fixed and essentially non-speculative while in case of the second one the profit level remains undetermined and is predicated on speculative risk taking (Abdullah, 2011). Both structures can have an influence on the amount of equity versus the amount of debt that a company holds.
According to Abdullah (2011) much of Islamic finance today is focused on replicating the conventional system. Sukuk are not only structured to imitate bonds; the vast majority are structured to mimic conventional bonds. Like a conventional bond, issuers are obliged to make fixed periodic payments to holders of these type of sukuk. This is also known as paying interest periodically (Abdullah, 2011). Only a small part is structured to have the necessary assets backed up in case of an sukuk issuance. In this case, profits fluctuate according to the profitability of the underlying assets. However, it must be noted that in case of sukuk issued to finance trade, this profit is fixed in advance and in case of sukuk issued to finance other activities, the profit varies with market conditions. By all means, holders share the losses. Abdullah (2011) argues that the amount of debt that a company holds, affects the possibility of paying interest. The higher the debt, the more likely the company will pay interest. Thus, accumulating excessive debt must be avoid. It must be noted that Abdullah (2011) provides no empirical evidence for his statement.
As for conventional bonds, Leland (1994) provides empirical evidences that debt value is explicitly linked to the amount of interest that the company pays1.
Ang (2013) mentions that the structure of numerous sukuk transactions is built to guarantee a fixed stream to sukuk holders, despite the fact that the underlying profits are variable and uncertain over time. According to Ang (2013) this raises the greatest concern because of its association with interest, which is prohibited in Islam. Sukuk is supposed to represent ownership in profit-earning assets or companies not debt owed (Ang, 2013). Most sukuk adopt predetermined rates of return, either capital-based fixed rates or floating rates based on benchmark interest rates (Ang, 2013).
According to Zakaria et al. (2012) asset security or corporate guarantees are vital in sukuk structures. There must be enough equity to back up the issuance of sukuk, because it is regarded as protection for investors in case of default (Zakaria et al., 2012). The possibility of default risk is also present in sukuk (Zakaria et al., 2012). However, sukuk claim to be safer than conventional bonds, because it theoretically transfers ownership of the underlying assets to the holders. Sukuk holders are in case of default protected. Despite this fact there is an increasing number of defaulted sukuk (Zakaria et al., 2012). This is because an increasing number of sukuk is classified as debt sukuk or loans, which means the payment of interest instead of dividend. Default on loan takes place when the borrower cannot longer pay the interest or principal on a certain date (Zakaria et al., 2012). The greater the default risk, the higher the interest rate charged by lenders (Zakaria et al., 2012, p. 665). The statement of Zakaria et al. (2012) imply that more companies have less equity to back up the issuance of sukuk and thus classify this as debt sukuk or loan. This leads to the payment of interest. Wang & Downing (2005) have a somewhat similar result for conventional bonds. The amount of equity that a company has is negatively correlated with the level of interest rates. In other words, the more equity a company has, the lower the interest rate will be.
In contrast, Abdel-Khaleq & Richardson (2007) mention that not the amount of debt that the company holds affects the possbility of paying interest, but the degree of compliance with Islamic law. An issuer with too much debt would technically be barred from issuing sukuk (Abdel-Khaleq & Richardson, 2007). Abdel-Khaleq & Richardson (2007) and Ang (2013) mention that sukuk is truly Islamic if it is Islamic law compliant in both form and substance. Therefore, the role of policy makers, government officials, bankers and issuers of sukuk in both the Islamic and Western world is very important. What is the extent to which they are willing to encourage the development of sukuk to align with the principles of Islamic law. For example, sukuk in non-Islamic countries do not face the traditional limitations of the Islamic law, like gambling, alcohol and more importantly the prohibition of interest. A properly structured sukuk (asset-backed) may generate a profit-based return that essentially replicates a prime rate or LIBOR2 (Abdel-Khaleq & Richardson, 2007, p. 412). Sukuk offerings will likely be increasingly backed by assets in non-Islamic jurisdictions (Abdel-Khaleq & Richardson, 2007, p. 425).
Abdul Aziz & Gintzburger (2009) mention that in jurisdictions such as the United States, financiers are required by certain regulations, whether conventional or Islamic, to report implicit interest rates in credit sales and leases3. In the current practice almost all financial institutions and companies calculate interest. The amount of debt does not matter.
Based on the above mentioned literature we conclude that a group of researchers argues that there is a correlation between equity versus total assets and the likelihood of payment of interest, while another group does not see that correlation. We predict that companies with a high equity ratio will borrow less and issue asset-backed sukuk and thus will pay dividend instead of interest, while companies with a low equity ratio will borrow more and thus pay interest. More formally:
H3: Companies with a high equity ratio are less likely to pay a fixed periodic distribution amount than companies with a low equity ratio.
3.3 Empirical models
The model that I will use to test whether or not there is evidence for our prediction is the regression model. I have three hypotheses that I want to test. I will use two models for these three hypotheses. Model 1 will be used to test hypothesis 1 and 2 and model 2 will be used to test hypothesis 3.
The formula of the regression model is as followed: Y = α + β1X1 + β2X2 + … + βiXi + ε
Where:
Y = dependent variable (DV) / outcome variable (OV) Xi = independent variable (IV) / predictor variable (PV)
α = constant
βi = coefficient of IV / PV
Model 1
Independent Variable Dependent Variable
H1 H2 Control variables H1 H2
Debt ratio Equity/Sukuk ratio
Liabilities Equity Proportion of sukuk
financing Amount of sukuk
Model 2
Independent Variable Dependent Variable
H3
Control variables
H3
Equity ratio Interest (fixed periodic
distribution amount)
4 Research Methodology
The London Stock Exchange (LSE) is one of the world’s oldest stock exchange (London Stock Exchange Group, 2016). It not only offers real time data and access to markets, it is also the most popular stock exchange for sukuk listings (World Islamic Economic Forum, 2013). For my research I analyse sukuk that can be traded on the secondary market. Because the LSE is the most popular stock exchange for sukuk listing, I obtained data from this stock exchange.
The database of LSE contains 45 sukuk prospectuses which I all collected. According to WebFinance, Inc. (2016) a prospectus is a legally mandated document published by every firm offering its securities to public for purchase. A prospectus contains financial information as well as non-financial information. The 45 sukuk are issued by 30 different companies. My sample is equal to the total population of 30 unique companies. I analyzed the 30 prospectuses regarding the sukuk, which were issued in the period 2006 – 2015 and have a maturity date that varies between 2011 – 2025. I obtained both financial and non-financial data. Important to note is that the financial data are obtained from the audited financial statements. Table 2 provides an overview of all variables that were collected.
Table 2: Overview of all collected variables
1. Issuer 2. Year of the
financial statement
3. Currency 4. Total revenues 5. Total expenses 6. Total equity 7. Total liabilities 8. Total assets 9. Amount of sukuk
issued
10. Amount raised 11. Date of issuance 12. Maturity date
13. Maturity 14. Interest 15. Country of origin
16. USD
Table 3 provides an overview of all variables that were calculated. Table 3: Overview of all calculated variables
1. Debt ratio 2. Equity ratio 3. Debt/Equity ratio
4. Proportion of sukuk financing
5. Equity/Sukuk ratio
4.1 Overview of company data Nr Issuer Tota l equit y * Tota l li abil it ies * Tota l assets * Tota l re ve nue s * Tota l expe nses * De bt ra ti o Equit y ra ti o Amount of sukuk * Equit y/S ukuk rati o P ropor ti o n of sukuk fina nc ing Inte re st (f ixed pe riodic dist ributi on amount )
1 ADCB ISLAMIC FINANCE LIMITED $ 5.329 $ 43.210 $ 48.540 $ 1.453 $ 1.344 0,89 0,11 $ 500 10,66 0,01 4,071%
2 ADIB SUKUK COMPANY LIMITED $ 549 $ 5.493 $ 6.042 $ 395 $ 105 0,91 0,09 $ 5.000 0,11 0,15 -
3 ALDAR FUNDING LIMITED $ 891 $ 500 $ 1.391 $ 455 $ 115 0,36 0,64 $ 2.530 0,35 5,18 5,767%
4 ALMANA SUKUK LIMITED $ 593 $ 603 $ 1.196 $ 272 $ 253 0,50 0,50 $ 215 2,76 0,36 4,794%
5 BSF SUKUK LIMITED $ 5.240 $ 32.213 $ 37.453 $ 1.218 $ 442 0,86 0,14 $ 2.000 2,62 0,06 2,947%
6 CBB INTERNATIONAL SUKUK
COMPANY $ 1.170 $ 4.092 $ 5.262 $ 272 $ 163 0,78 0,22 $ 350 3,34 0,06 -
7 DANA GAS SUKUK LIMITED $ 1.855 $ 8 $ 1.863 $ 231 $ 9.530 0,00 1,00 $ 1.000 1,86 123,56 7,500%
8 DAR AL-ARKAN INTERNATIONAL $ 72 $ 1.364 $ 1.436 $ 183 $ 111 0,95 0,05 $ 450 0,16 0,33 -
9 DIB SUKUK COMPANY LIMITED $ 2.403 $ 15.142 $ 17.545 $ 1.246 $ 336 0,86 0,14 $ 750 3,20 0,05 -
10 DP WORLD SUKUK LIMITED $ 9.131 $ 9.111 $ 18.242 $ 3.660 $ 3.450 0,50 0,50 $ 1.500 6,09 0,16 6,250%
11 EIB SUKUK COMPANY LIMITED $ 363 $ 4.253 $ 4.616 $ 262 $ 197 0,92 0,08 $ 1.000 0,36 0,08 -
12 EMAAR SUKUK LIMITED $ 7.862 $ 9.601 $ 17.463 $ 2.406 $ 1.854 0,55 0,45 $ 2.000 3,93 0,21 8,500%
13 FGB SUKUK COMPANY LIMITED $ 6.707 $ 31.619 $ 38.326 $ 1.717 $ 752 0,83 0,17 $ 3.500 1,92 0,11 3,797%
15 GFH SUKUK LIMITED $ 668 $ 774 $ 1.501 $ 346 $ 134 0,52 0,44 $ 1.000 0,67 0,26 -
16 IDB TRUST SERVICES LIMITED $ 1.570 $ 756 $ 2.326 $ 86 $ 52 0,32 0,68 $ 3.500 0,45 4,63 1,775%
17 IIG FUNDING LIMITED $ 285 $ 248 $ 533 $ 113 $ 31 0,47 0,53 $ 200 1,43 0,81 6,750%
18 JAFZ SUKUK LIMITED $ 1.583 $ 1.641 $ 3.224 $ 213 $ 53 0,51 0,49 $ 2.042 0,77 1,21 -
19 TDIC SUKUK LIMITED $ 4.192 $ 7.617 $ 11.809 $ 165 $ 511 0,64 0,36 $ 1.000 4,19 0,13 4,949%
20 KT SUKUK VARLÝK KIRALAMA A.Þ. $ 826 $ 5.384 $ 6.210 $ 345 $ 210 0,87 0,13 $ 350 2,36 0,07 5,875%
21 NIG SUKUK LIMITED $ 2.881 $ 2.159 $ 5.040 $ 345 $ 263 0,43 0,57 $ 1.500 1,92 0,21 -
22 QATAR ALAQARIA SUKUK
COMPANY $ 462 $ 581 $ 1.043 $ 90 $ 52 0,56 0,44 $ 300 1,54 0,50 -
23 QIB SUKUK FUNDING LIMITED $ 2.472 $ 4.511 $ 10.782 $ 662 $ 170 0,42 0,23 $ 750 3,30 0,17 3,856%
24 RAKIA SUKUK COMPANY LIMITED $ 178 $ 44 $ 222 $ 11 $ 7 0,20 0,80 $ 325 0,55 7,18 -
25 SAUDI ELECTRICITY GLOBAL
SUKUK COMPANY $ 14 $ 43 $ 57 $ 8 $ 8 0,76 0,24 $ 500 0,03 0,00 2,665%
26 SIB SUKUK COMPANY $ 574 $ 869 $ 1.443 $ 88 $ 38 0,60 0,40 $ 225 2,55 0,27 -
27 SOQ SUKUK A Q.S.C.
$ 59.975 $ 41.000 $ 100.976 $ 82.512 $ 12.479 0,41 0,59 $ 2.000 29,99 0,05 2,099%
28 TABREED 06 FINANCING
CORPORATION $ 312 $ 414 $ 726 $ 115 $ 96 0,57 0,43 $ 200 1,56 0,52 -
29 URC SUKUK LIMITED $ 354 $ 312 $ 667 $ 102 $ 51 0,47 0,53 $ 100 3,54 0,32 -
30 MAF SUKUK LIMITED $ 5.442 $ 4.395 $ 9.837 $ 5.350 $ 5.143 0,45 0,55 $ 400 13,60 0,09 5,850%
5 Empirical results
This section has three parts. The first and the second part report descriptive statistics for the variables and the correlations. The third part presents the regression models and the main findings.
My analysis has resulted in the elimination of company 14, which is GE Capital Sukuk Limited. The reason of its elimination is that its presents in the results leads to a highly skewed distribution of the population. GE Capital Sukuk Limited, a special purpose vehicle of General Electric, has a total equity of USD 104.7 billion, while the remaining 29 companies have a combined total equity of USD 124 billion. It also has total assets worth USD 797.8 billion, while the other companies have a combined total assets of USD 355.8 billion. GE Capital Sukuk Limited generates a total revenue of USD 182.5 billion, while the remaining 29 companies generate a combined total revenue of USD 104.3 billion. The numbers of GE Capital Sukuk Limited, if staying present in the results, will have a significant impact on the interpretation of the results4. Its elimination leaves us with 29 unique companies. These results will be shown below.
5.1 Descriptive Statistics and Bivariate Correlations
Table 4 presents descriptive statistics of the variables used in the analyses. These descriptive statistics portrait the 29 firms that were used in our sample. It contains 18 variables of which 5 variables deserve special attention: Total_revenues_mln, Total_expenses_mln,
Total_equity_mln, Total_liabilities_mln and Total_assets_mln. These variables are calculated on the basis of the currency of that particular company. Apart from the US Dollar there are 6 other currencies. In order to compare them with each other, I converted the numbers of these 6 currencies to the currency that is been used the most globally, which is the US Dollar (Middelkoop, 2007). The variables mentioned above are shown in the descriptive statistics in both its domestic currency and US Dollar. This allows the reader to get a better picture of the results.
The variables: USD, Total_revenues$_mln, Total_expenses$_mln, Total_equity$_mln, Total_liabilities$_mln, Total_assets$_mln and Sukuk_issued_mln all have a wide range statistic which is a clear indication of a high skewness. The mean for Total_revenues$_mln, for example, is USD 3.6 billion and its statistical range varies between USD 8.3 million and USD 82.5 billion, while the mean for Total_assets$_mln is USD 12.3 billion and its
statistical range even varies between USD 56.9 million and USD 101 billion.
The same applies to the ratios (variables): Debt_Equity_ratio, Proportion_sukuk_financing and Equity_Sukuk_ratio, with for example a mean of 3.6 and a statistical range that varies between 0.03 and 30 for Equity_Sukuk_ratio and a mean of 5.1 and a statistical range that varies between 0 and 123.6 for Proportion_sukuk_financing.
The variables: Debt_ratio, Equity_ratio and Interest are the exception. Debt_ratio and Interest both have a mean of 0.6 and a statistical range that varies between 0 and 1, while Equity_ratio has a mean of 0.4 and a statistical range that varies between 0.1 and 1.
Looking from another perspective, which is the indicator “Skewness” itself (see table 4), we also can make a statement. In a normal distribution the values of skew are 0 (i.e., the tails of the distribution are as they should be). However, looking to table 4 we can conclude that almost all variables are skewed. An explanation for this finding is the relatively small sample5 that we used in investigating this topic.
Table 4: Descriptive Statistics
N Range Minimum Maximum Mean
Std.
Deviation Variance Skewness
Statistic Statistic Statistic Statistic Statistic Std. Error Statistic Statistic Statistic Std. Error
USD 29 3,192 ,267 3,459 ,797 ,194 1,043 1,087 2,037 ,434 Total_revenues_mlna 29 82482,525 29,475 82512,000 5095,377 2863,442 15420,108 237779735,222 4,866 ,434 Total_expenses_mlna 29 18879,707 8,918 18889 2028,891 779,470 4197,573 17619616,048 3,086 ,434 Total_equity_mlna 29 59923,107 51,893 59975 8644,602 2397,004 12908,263 166623248,730 2,581 ,434 Total_liabilities_mlna 29 158669,000 29 158698 22697,179 7554,093 40680,038 1654865492,724 2,369 ,434 Total_assets_mlna 29 178116,789 154,211 178271 31820,933 9048,930 48729,982 2374611132,090 1,895 ,434 Total_revenues$_mlna 29 82503,741 8,259 82512 3597,275 2827,220 15225,047 231802041,072 5,332 ,434 Total_expenses$_mlna 29 12471,706 7,294 12479 980,315 461,490 2485,198 6176208,532 3,984 ,434 Total_equity$_mlna 29 59961,165 13,835 59975 4274,191 2044,454 11009,719 121213913,830 4,947 ,434 Total_liabilities$_mlna 29 43202,353 7,896 43210,250 7860,598 2328,410 12538,873 157223338,139 2,003 ,434 Total_assets$_mlna 29 100919,092 56,908 100976 12267,879 3929,528 21161,155 447794473,544 3,074 ,434 Sukuk_issued_mlna 29 4900 100 5000 1213,351 223,309 1202,556 1446141,149 1,598 ,434 Debt_ratioc 29 ,950 ,000 ,950 ,590 ,044 ,237 ,056 -,261 ,434 Equity_ratioc 29 ,950 ,050 1,000 ,397 ,044 ,237 ,056 ,414 ,434 Debt_Equity_ratioc 29 19,010 ,000 19,010 3,366 ,802 4,319 18,656 2,191 ,434 Proportion_sukuk_financingc 29 123,560 ,000 123,560 5,060 4,244 22,856 522,419 5,337 ,434 Equity_Sukuk_ratioc 29 29,960 ,030 29,990 3,649 1,095 5,895 34,747 3,606 ,434 Interestb 29 1,000 ,000 1,000 ,552 ,094 ,506 ,256 -,220 ,434
Table 5 presents Pearson’s correlation matrix, with Pearson’s correlation coefficient, r, as a way to measure the strength of relationship between two variables.
To interpret r better Cohen (1988, 1992) suggested a measuring scale which you can see below:
r
Value Effect Description.10 Small Effect explains 1% of the total variance .30 Medium Effect explains 9% of the total variance .50 Large Effect explains 25% of the total variance
As you can see the highest correlation can be found between Total_equity$_mln and Total_revenues$_mln with a r value of .984. This suggests a large effect and explains at least 25% of the total variance. A possible explanation for this value is that a company who is able to generate more revenue, is more likely to generate more profit and thus will have more equity. After all, profit is part of total equity on the balance sheet.
Logical high (large effect) correlations can be found between Total_expenses$_mln and Total_revenues$_mln with a correlation of .922, because if a company generates more revenue it is more likely that total expenses go up. Another logical high correlation of .825 between Total_assets$_mln and Total_revenues$_mln, has a plausible explanation that a company with more assets is more likely to generate more revenue. The third6 logical high correlation can be found between Total_assets$_mln and Total_equity$_mln and Total_assets$_mln and Total_liabilities$_mln with values of respectively .885 and .912. There is no need to explain this further. The above mentioned correlations indicate multicollinearity.
Interesting correlations can be found between Total_assets$_mln and Total_expenses$_mln with a value of .807. A possible explanation is that a significant amount of total assets is financed with debt, which usually results in the payment of interest. However, it must be noted that this high value could be influenced by the relation between Total_revenues$_mln and Total_expenses$_mln and Total_assets$_mln and Total_revenues$_mln. Another interesting correlation is between Total_liabilities$_mln and Total_expenses$_mln with a value of .539. The reason that it is interesting is because of the presence of payment of interest. The higher the amount of liabilities, the higher the payment of interest will be. It must be noted that in Islam it is forbidden to charge interest, which of course also counts for the issuance of sukuk (Zakaria et al., 2012). However, the poor correlation between Sukuk_issued_mln and Total_expenses$_mln with a value of .103 contradicts that. This could be an indication that the issuance of sukuk does not involve the payment of interest.
Table 5: Pearson’s Correlations among Variables (N=29)
USD Total_revenues$_mln Total_expenses$_mln Total_equity$_mln Total_liabilities$_mln Total_assets$_mln Total_revenues$_mln ,025 Total_expenses$_mln -,021 ,922** Total_equity$_mln ,001 ,984** ,941** Total_liabilities$_mln -,162 ,530** ,539** ,617** Total_assets$_mln -,099 ,825** ,807** ,885** ,912** Sukuk_issued_mln -,222 ,131 ,103 ,183 ,253 ,243
**. Correlation is significant at the 0.01 level (2-tailed).
Table 5 (continued): Pearson’s Correlations among Variables7
(N=29)
Debt_ratio Equity_ratio Debt_Equity_ratio Proportion_sukuk_financing Equity_Sukuk_ratio
Equity_ratio -,962**
Debt_Equity_ratio ,777** -,757**
Proportion_sukuk_financing -,516** ,530** -,170
Equity_Sukuk_ratio -,114 ,120 -,160 -,075
While Table 5 presents Pearson’s correlations matrix, table 6 presents Spearman’s correlation matrix. Spearman’s correlation coefficient (denoted by rs) can, among other things, be useful to minimize the effects of extreme scores/results (Field, 2013). It is a non-parametric statistic based on ranked data (Field, 2013). So it is used for variables on at least a ordinal level. If we look to our sample we can see that, for example, some companies have a revenue of hundreds of billions of US Dollars, while other companies have only a revenue of millions of US Dollars. We can see these big differences also if we look to other variables8. This indicates that the population is not normally distributed. Pearson’s r correlation makes the assumption that variables should be normally distributed. Spearman’s rs correlation does not make that assumption. Therefore Spearman’s correlation coefficient is a good alternative.
To make life easier we use the same measuring scale as for Pearson’s r correlation9. When we analyse table 6 we find a lot of similarities between correlation values according to Pearson and correlation values according to Spearman. If we observe for example the
correlation between Total_expenses$_mln and Total_revenues$_mln according to Pearson (.922) we can see that this is almost equal to the correlation value according to Spearman (.867). Another example is the difference in correlation value between Pearson and
Spearman, regarding Total_liabilities$_mln and Total_revenues$_mln. The difference here is bigger but both are higher than .50 and thus both have a large effect (respectively .530 and .775).
It becomes interesting if we look to correlation values that are very different from each other. The correlation value of .457 between Sukuk_issued_mln and
Total_revenues$_mln is much higher according to Spearman. This suggests that the issuance of sukuk has a positive medium effect on generating more revenue. A possible explanation is that the issuance of sukuk leads to an increase in total assets which in turn can lead to an increase in revenues. Another interesting correlation is between Sukuk_issued_mln and Total_expenses$_mln with a value of .359. Despite the fact that it is a medium effect, it is a significant increase compared to Pearson’s .103. Correlation between Sukuk_issued_mln and Total_equity$_mln is also interesting because of Spearman’s value of .550 which means a large effect compared to Pearson’s value of .183.
Table 6: Spearman’s Correlations among Variables 10 (N=29) USD Total_revenues $_mln Total_expenses $_mln Total_equity $_mln Total_liabilities $_mln Total_assets $_mln Total_revenues$_mln ,010 Total_expenses$_mln -,051 ,867** Total_equity$_mln -,147 ,795** ,816** Total_liabilities$_mln -,141 ,775** ,859** ,752** Total_assets$_mln -,155 ,819** ,853** ,882** ,951** Sukuk_issued_mln -,350 ,457* ,359 ,550** ,466* ,549**
**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).
Table 6 (continued): Spearman’s Correlations among Variables (N=29)
Debt_ratio Equity_ratio Debt_Equity_ratio Proportion_sukuk_financing Equity_Sukuk_ratio
Equity_ratio -,947**
Debt_Equity_ratio ,958** -,998**
Proportion_sukuk_financing -,581** ,601** -,607**
Equity_Sukuk_ratio -,103 ,079 -,077 -,416*
Interest -,249 ,199 -,199 -,178 ,344
**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).
5.3 Main findings
In order to conclude whether the three hypotheses are supported or not, these were tested by means of performing three different linear regression analyses. Every regression analysis consists of the following four parts: Pearson’s Correlations, Model Summary, ANOVA model and Coefficients11. Each part is presented in a table:
Table Refer to hypothesis
7 up to and including 10 1 11 up to and including 14 2 15 up to and including 18 3
The first regression analysis shows the results of whether companies with a low debt are more likely to have more equity with respect to the amount of sukuk they want to issue or not. Looking to table 7 we see one interesting correlation between Total_liabilities$_mln and Equity_Sukuk_ratio which is significant (p < .01) and has a correlation value of .625. This shows that there is a positive correlation between these two variables, which means that if total liabilities go up with USD 1 million this results in a .625 higher Equity/Sukuk ratio (more equity with respect to the amount of sukuk they want to issue). This does not support hypothesis 1.
The R2 and Adjusted R2 reported in model 1 and 2 of table 8 are both very high. Model 1 reports a R2 = 82.9% and a Adjusted R2 = 80.9%, while model 2 reports a R2 = 83.4% and a Adjusted R2 = 80.7%. In other words, looking to the more important Adjusted R2 of model 2, we can conclude that 80.7% of the total variance in the dependent variable (DV) Equity/Sukuk ratio is explained by the independent variable (IV) Debt ratio. This high percentage is a good sign. However, an explanation for these high R2 and Adjusted R2 values is because of a number of logical correlations. For example, the correlation between Total_equity$_mln and Equity_Sukuk_ratio. Despite this fact table 8 reports a Sig. F Change > .05 for model 2. This means that hypothesis 1 is not supported.
The ANOVA model (table 9) contradicts this by reporting a p value of 0.000 (p < .05) for model 2. However, table 10 states non-significant values for almost all variables. Debt_ratio for example is not significant at the .05 level (t = -.848, p > .05). Total_equity$_mln, which is an exception and is significant, has a logical explanation: if total equity goes up, Equity/Sukuk ratio will improve.
The last aspect we look at are the tolerance statistic and the variance inflation factor (VIF). A rule of thumb:
Collinearity Statistics Value Description Source
VIF Higher than 10 Cause of concern (Bowerman &
O'Connell, 1990; Myers, 1990)
VIF Average VIF
substantially Higher than 1 Regression may be biased (Bowerman & O'Connell, 1990)
Tolerance Below .1 Indicates a serious
problem
(Field, 2013)
None of the VIF values in model 2 in table 10 is higher than 10 but the average VIF is substantially higher than 1. This can be an indication that the regression is biased. The tolerance statistics are all above .1 so there are no problems for this part.
Overall we can conclude that there is no evidence that supports the statement that companies that have a low debt ratio, are more likely to have more equity with respect to the amount of sukuk they want to issue, than companies that have a high debt ratio. In other words, hypothesis 1 is not supported.
Table 7: Pearson’s Correlations among Variables regarding hypothesis 1
Equity_Sukuk_ratio Total_liabilities$_mln Total_equity$_mln Proportion_sukuk_financing Pearson Correlation Total_liabilities$_mln ,625 Total_equity$_mln ,907 ,617 Proportion_sukuk_financing -,075 -,140 -,053 Debt_ratio -,114 ,345 -,129 -,516 Sig. (1-tailed) Total_liabilities$_mln ,000 . Total_equity$_mln ,000 ,000 . Proportion_sukuk_financing ,349 ,234 ,393 . Debt_ratio ,279 ,033 ,252 ,002
Table 8: Model Summary regarding hypothesis 1
Model R R Square Adjusted R Square Std. Error of the Estimate Change Statistics R Square Change F Change df1 df2 Sig. F Change 1 ,911a ,829 ,809 2,577 ,829 40,487 3 25 ,000 2 ,913b ,834 ,807 2,592 ,005 ,718 1 24 ,405
a. Predictors: (Constant), Proportion_sukuk_financing, Total_equity$_mln, Total_liabilities$_mln b. Predictors: (Constant), Proportion_sukuk_financing, Total_equity$_mln, Total_liabilities$_mln,
Table 9: ANOVAa model regarding hypothesis 1
Model Sum of Squares Df Mean Square F Sig.
1 Regression 806,845 3 268,948 40,487 ,000b Residual 166,070 25 6,643 Total 972,916 28 2 Regression 811,672 4 202,918 30,203 ,000 c Residual 161,243 24 6,718 Total 972,916 28
a. Dependent Variable: Equity_Sukuk_ratio
b. Predictors: (Constant), Proportion_sukuk_financing, Total_equity$_mln, Total_liabilities$_mln
c. Predictors: (Constant), Proportion_sukuk_financing, Total_equity$_mln, Total_liabilities$_mln, Debt_ratio
Table 10: Coefficientsa regarding hypothesis 1 Model Unstandardized Coefficients Standardized Coefficients t Sig.
Correlations Collinearity Statistics
B Std. Error Beta Zero-order Partial Part Tolerance VIF
1
(Constant) 1,360 ,588 2,314 ,029
Total_liabilities$_mln 5E-005 5E-005 ,104 ,978 ,337 ,625 ,192 ,081 ,608 1,645
Total_equity$_mln ,000 6E-005 ,842 8,012 2E-008 ,907 ,848 ,662 ,618 1,617
Proportion_sukuk_financing -,004 ,022 -,016 -,195 ,847 -,075 -,039 -,016 ,978 1,022
2
(Constant) 2,799 1,797 1,557 ,132
Total_liabilities$_mln 8E-005 6E-005 ,165 1,281 ,213 ,625 ,253 ,106 ,417 2,398
Total_equity$_mln ,000 7E-005 ,789 6,417 1E-006 ,907 ,795 ,533 ,457 2,187
Proportion_sukuk_financing -,016 ,026 -,063 -,625 ,538 -,075 -,127 -,052 ,687 1,455
Debt_ratio -2,513 2,964 -,101 -,848 ,405 -,114 -,170 -,070 ,486 2,056
The second regression analysis shows the results of whether companies with a low debt are more likely to issue a higher amount of sukuk than companies with a high debt ratio or not. Looking to table 11, two correlations deserve special attention: the first one is the correlation between Total_expenses$_mln and Total_revenues$_mln, which is significant (p < .01) and has a correlation value of .922. It shows that there is a strong positive correlation between these two variables, which means that if total expenses go up with USD 1 million this results in USD 922.00012 more revenues. An explanation for this result could be that companies sold more and thus the cost of sales increased. The second correlation and more importantly, is between Debt_ratio and Sukuk_issued_mln, which is not significant (p > .01) and has a positive correlation value of .073. This correlation relates directly to hypothesis 2 and should have a strong negative value. Instead, it has a weak positive value. In other words, table 11 is an indication that hypothesis 2 is not supported.
The R2 and Adjusted R2 reported in model 1 and 2 in table 12 are both very low. The Adjusted R2 in both models is even negative. Model 1 reports a R2 = 13.9% and a Adjusted R2 = - 5.6%, while model 2 reports a R2 = 2.8% and a Adjusted R2 = - 8.9%. In other words, looking to the more important Adjusted R2 of model 2, we can conclude that - 8.9%. of the total variance in the DV Amount of sukuk is explained by the IV Debt ratio. Model 2 does an even worse job with respect to model 1. The Adjusted R2 goes down from - 5.6% to - 8.9%, which also is an indication that hypothesis 2 is not supported. Despite the fact that the Sig. F Change goes from down from .775 for model 1 to .648 for model 2, it is still not significant (p > .05). This means that hypothesis 2 is not supported.
The ANOVA model (table 13) confirms the conclusion drawn by table 12 also by reporting a p value of .869 (p > .05) for model 2.
Table 14 states non-significant values (p > .05) for all variables. Debt_ratio for example is here also not significant at the .05 level (t = .462, p > .05). As for the VIF, none of the VIF values in model 2 in table 14 is higher than 10, but the variables Total_revenues$_mln and Total_expenses$_mln are substantially higher than 1. The exception is the more importantly variable Debt_ratio with a VIF value of 1.023, which is close to 1. The tolerance statistics are all above .1 so there are no problems for this part.
Overall we can conclude that there is no evidence that supports the statement that companies that have a low debt ratio, are more likely to issue a higher amount of sukuk than companies with a high debt ratio. In other words hypothesis 2 is not supported.
Table 11: Pearson’s Correlations among Variables regarding hypothesis 2
Sukuk_issued_mln Total_revenues$_mln Total_expenses$_mln Pearson Correlation
Total_revenues$_mln ,131
Total_expenses$_mln ,103 ,922
Debt_ratio ,073 -,145 -,150
Sig. (1-tailed) Total_revenues$_mln ,248 .
Total_expenses$_mln ,297 ,000 .
Debt_ratio ,353 ,227 ,219
Table 12: Model Summary regarding hypothesis 2
Model R R Square Adjusted R Square Std. Error of the Estimate Change Statistics R Square Change F Change df1 df2 Sig. F Change 1 ,139a ,019 -,056 1235,756 ,019 ,258 2 26 ,775 2 ,167b ,028 -,089 1254,892 ,008 ,213 1 25 ,648
a. Predictors: (Constant), Total_expenses$_mln, Total_revenues$_mln
Table 13: ANOVAa model regarding hypothesis 2
Model Sum of Squares df Mean Square F Sig.
1 Regression 787564,831 2 393782,415 ,258 ,775b Residual 39704387,347 26 1527091,821 Total 40491952,178 28 2 Regression 1123116,647 3 374372,216 ,238 ,869 c Residual 39368835,531 25 1574753,421 Total 40491952,178 28
a. Dependent Variable: Sukuk_issued_mln
b. Predictors: (Constant), Total_expenses$_mln, Total_revenues$_mln c. Predictors: (Constant), Total_expenses$_mln, Total_revenues$_mln, Debt_ratio
Table 14: Coefficientsa regarding hypothesis 2 Model Unstandardized Coefficients Standardized Coefficients t Sig.
Correlations Collinearity Statistics
B Std. Error Beta Zero-order Partial Part Tolerance VIF
1 (Constant) 1201,743 258,968 4,641 ,000 Total_revenues$_mln ,019 ,040 ,243 ,484 ,632 ,131 ,095 ,094 ,150 6,680 Total_expenses$_mln -,059 ,243 -,121 -,241 ,811 ,103 -,047 -,047 ,150 6,680 2 (Constant) 920,058 664,479 1,385 ,178 Total_revenues$_mln ,020 ,040 ,247 ,484 ,632 ,131 ,096 ,096 ,150 6,682 Total_expenses$_mln -,054 ,247 -,111 -,217 ,830 ,103 -,043 -,043 ,149 6,693 Debt_ratio 467,358 1012,455 ,092 ,462 ,648 ,073 ,092 ,091 ,977 1,023
The last regression analysis shows the results of whether companies with a high equity ratio are less likely to pay a fixed periodic distribution amount than companies with a low equity ratio. One correlation in table 15 deserves special attention, which is the correlation between Equity_ratio and Interest. It is not significant (p > .01) and has a weak positive correlation value of .142. This correlation relates directly to hypothesis 3 and should have a strong negative value. Instead it has a weak positive value. In other words, table 15 is an indication that hypothesis 3 is not supported.
The R2 in model 1 and 2 of table 16 is respectively 13.9% and 21.5%. The Adjusted R2 reported in model 1 and 2 is respectively 7.3% and 12.1%. Despite the fact that both the R2 and Adjusted R2 are higher for model 2 than for model 1, the Adjusted R2 in model 2 decreased significantly with 9.4%13 to a level of 12.1%. In other words, we can conclude that only 12.1%. of the total variance in the DV Interest (fixed periodic distribution amount) is explained by the IV Equity ratio. The Sig. F change for model 2 in table 16 is 1% lower than model 1, but it is still not significant (p > .05). This means that hypothesis 3 is not supported.
The ANOVA model (table 17) confirms the conclusion drawn by table 16 also, by reporting a p value of .104 (p > .05) for model 2.
Table 18 states non-significant values (p > .05) for all variables. Equity_ratio, for example, is also not significant at the .05 level (t = 1.553, p > .05). As for the VIF, none of the VIF values in model 2 in table 18 is higher than 10, but the variables Total_liabilities$_mln and Total_equity$_mln are substantially higher than 1. The exception is the more importantly variable Equity_ratio with a VIF value of 1.422, which is relative close to 1. The tolerance statistics are all above .1 so there are no problems for this part.
Overall we can conclude that there is no evidence that supports the statement that companies with a high equity ratio are less likely to pay a fixed periodic distribution amount than companies with a low equity ratio. In other words hypothesis 3 is not supported.
Table 15: Pearson’s Correlations among Variables regarding hypothesis 3
Interest Total_liabilities$_mln Total_equity$_mln Pearson Correlation
Total_liabilities$_mln ,366
Total_equity$_mln ,282 ,617
Equity_ratio ,142 -,328 ,140
Sig. (1-tailed) Total_liabilities$_mln ,025 .
Total_equity$_mln ,069 ,000 .
Equity_ratio ,232 ,041 ,235
Table 16: Model Summary regarding hypothesis 3
Model R R Square Adjusted R Square Std. Error of the Estimate Change Statistics R Square Change F Change df1 df2 Sig. F Change 1 ,373a ,139 ,073 ,487 ,139 2,101 2 26 ,143 2 ,463b ,215 ,121 ,475 ,076 2,411 1 25 ,133
a. Predictors: (Constant), Total_equity$_mln, Total_liabilities$_mln
Table 17: ANOVAa model regarding hypothesis 3
Model Sum of Squares df Mean Square F Sig.
1 Regression ,998 2 ,499 2,101 ,143b Residual 6,175 26 ,237 Total 7,172 28 2 Regression 1,541 3 ,514 2,280 ,104 c Residual 5,632 25 ,225 Total 7,172 28
a. Dependent Variable: Interest
b. Predictors: (Constant), Total_equity$_mln, Total_liabilities$_mln c. Predictors: (Constant), Total_equity$_mln, Total_liabilities$_mln, Equity_ratio
Table 18: Coefficientsa regarding hypothesis 3 Model Unstandardized Coefficients Standardized Coefficients t Sig.
Correlations Collinearity Statistics
B Std. Error Beta Zero-order Partial Part Tolerance VIF
1 (Constant) ,436 ,107 4,057 ,000 Total_liabilities$_mln 1E-005 ,000 ,310 1,342 ,191 ,366 ,254 ,244 ,620 1,614 Total_equity$_mln 4E-006 ,000 ,091 ,392 ,698 ,282 ,077 ,071 ,620 1,614 2 (Constant) ,123 ,227 ,543 ,592 Total_liabilities$_mln 2E-005 ,000 ,529 1,992 ,057 ,366 ,370 ,353 ,445 2,250 Total_equity$_mln -4,160E-006 ,000 -,090 -,357 ,724 ,282 -,071 -,063 ,488 2,048 Equity_ratio ,702 ,452 ,328 1,553 ,133 ,142 ,297 ,275 ,703 1,422
