» » » SUKUK VS CONVENTIONAL BONDS: A VALUE-AT-RISK BASED COMPARATIVE ANALYSIS
New publications!

SUKUK VS CONVENTIONAL BONDS: A VALUE-AT-RISK BASED COMPARATIVE ANALYSIS

BAKHSHI KOSTANDYAN
LOMONOSOV MOSCOW STATE UNIVERSITY, FACULTY OF ECONOMICS

Abstract. The present paper looks into the question of whether including sukuk (Islamic bonds) in the bond portfolio provides the investor with diversification benefits and how they what might be quantified. We analysed the sovereign bonds of Bahrain, Pakistan, Qatar, Malaysia, and the UAE. The conclusion is that using sukuk to diversify a portfolio yields substantial benefits.

Keywords: Islamic finance, sukuk, bonds, value-at-risk

JEL Classification: G11, G15, G39

KOSTANDYAN, BAKHSHI (2015) "SUKUK VS CONVENTIONAL BONDS: A VALUE-AT-RISK BASED COMPARATIVE ANALYSI". Journal of Russian Review (ISSN 2313-1578), VOL. 2(3), 36-49.


1. Introduction

Over the last decade, Islamic finance as an alternative to traditional finance has been growing at a significant pace. It has gained popularity not only in Islamic countries, but also in countries with a minority Muslim population like Great Britain or Japan. In theory, Islamic finances use the principles similar to those in traditional financing; however, the two differ in their application, structure, and content. In terms of issued volumes, one of the fastest growing instruments of Islamic finance is sukuk. Sukuk is a financial security based on Sharia principles that bears external resemblance to a bond. The peculiarities of sukuk’s structure, which is different from that of a bond, leave the question of its reliability and diversification benefit for an investor’s portfolio open. The present study, therefore, is undertaken to determine whether sukuk is a more reliable financial instrument than a conventional bond in terms of risk, and, specifically, whether it helps the investor to minimise portfolio losses.


2. Literature review

To present a review of literature on the subject separate searches for relevant studies in Russian and English were performed. Russian-language sources were searched for using the Russian Science Citation Index (RSCI) Electronic Library web site (elibrary.ru), English-language ones were searched for using the Social Science Research Network (SSRN) web site (ssrn.com), with the time horizon of 2005 to 2015.

The search at RSCI employed the following keywords (in Russian): Islamic finance, sukuk, Eurobonds, VaR analysis of sukuk. The web site’s search engine came up with 49 articles, out of which 3 were chosen by their titles and abstracts for further reading. However, upon reading the articles in full all 3 had to be dismissed as not relevant to the subject or having nothing of input for the present study, therefore leaving the Russian-language part of the Literature review section empty.

The search at SSRN employed the following keywords: sukuk, sukuk and Eurobonds, sukuk value-at-risk. The web site’s search engine came up with 102 articles, out of which 21 were chosen by their titles and abstracts for further reading. However, upon reading the articles in full 18 were dismissed as not relevant to the subject of or having nothing of input for the present study, producing 3 articles for the English-language part of the Literature review section.

The first of these, titled Comparison between Sukuk and Conventional Bonds: Value at Risk Approach (Khalid Abbasher Hassan, 2012) deals with various ways to combine sovereign Eurobonds with (international) sukuk of Malaysia and Dubai, as well as corporate Eurobonds and (international) sukuk of these countries. As data for his analysis, the author uses the bonds’ market price for the period of October 30, 2009 (or the earliest available data) to July 18, 2009 (or the latest available data).

The paper looks at three portfolios, one consisting of Eurobonds only (sovereign and corporate), one consisting of sukuk only (sovereign and corporate), and one combining all of the selected bonds (sovereign Eurobonds and sukuk, and corporate Eurobonds and sukuk). For each portfolio VaR calculations were performed, which allowed the author to claim that including sukuk in a portfolio yields diversification benefits. VaR was measured using the delta-normal method.

The study concludes that supplementing a portfolio of conventional bonds with sukuk yields diversification benefits, significantly lowering portfolio risks. At the same time sukuk are found to possess higher market and credit risks compared to conventional bonds owing to the limitations imposed by Sharia laws. Moreover, sukuk introduce additional risks due to a higher correlation between sukuk compared to internal bond correlation.

The paper Sukuk vs. Eurobonds: Is There a Difference in Value-at-Risk? (Selim Cakir and Faezeh Raei, 2007) also compares sukuk’s reliability in terms of risk to that of the bond. It analyses sovereign sukuk and Eurobonds of Malaysia, Pakistan, Qatar, and Bahrain. The data consisted of sukuk and Eurobonds market values starting from their date of issue or the earliest data available to the end of June, 2007.

This second paper analyses four portfolios, each containing securities (sukuk and Eurobonds) of the corresponding country. The study first calculates VaR for each country portfolio’s Eurobonds contents, and then adds the corresponding sukuk part to observe the change in VaR. Two methods are used: the delta-normal method, and the Monte-Carlo method (1% level of significance with a 5-day horizon).

The conclusion is that portfolios combining sukuk and Eurobonds definitely have a lower VaR than Eurobonds-only portfolios. Therefore, including sukuk in a portfolio of Eurobonds may provide diversification benefit. It is also said that the correlation between sukuk and Eurobonds yield is much lower than the correlation between Eurobonds. But the paper argues that sukuk yields are lower than bond yields, and that, owing to a small secondary market, sukuk is an instrument of lower liquidity.

The third paper, titled Differences and Similarities in Islamic and Conventional Banking (Muhammad Hanif, 2014), provides a theoretical angle on why, in terms of risk, sukuk is a more reliable financial instrument than the bond.

In points out that in Sharia-based Islamic finance usury is forbidden, therefore the use of bonds creating a creditor/debtor relationship (that is, pure indebtedness) is forbidden, too. Sukuk gives its holder the right to a part of the underlying asset, which the bond holder may lend to the issuer, receiving a rent for its use, and the right to coupon payments. When the bond expires, the sukuk issuer is obliged to buy back the corresponding part of the underlying asset, paying the sukuk off. Therefore, sukuk is creating an owner/renter relationship. The same principle, reminiscent of the lease agreement, lies behind ijara, a type of sukuk. Whereas the price of a bond is largely dependent on the quality (rating) of the borrower, the price of sukuk is also influenced by the value of the underlying asset itself, which is a more objective factor. Therefore the paper calls sukuk a more reliable financial instrument than the bond in terms of risks.


3. Data

Serving as data for the present study were market values of sovereign (international) sukuk and Eurobonds of the following countries: Bahrain, Pakistan, Qatar, Malaysia, and the UAE for the period starting from their issue or the earliest (first traded) date available to January 1, 2016. Quotations were taken from official Bloomberg financial databases and the website www.finanz.ru publishing trading summaries from German exchanges. For the list of all the securities used in this study see Appendix, Table 1.

The method for selecting the securities was as follows: first, choosing an issuer that has been issuing both Eurobonds (US dollar bonds listed on international exchanges) and sukuk listed in the US dollars on international exchanges. Therefore, all of the analysed securities share a common currency, the US dollar. The securities were picked on the secondary market, as it provides the data required for turther analysis. Each country had one sukuk and one Eurobond selected.

The present study is not without its limitations, owing to the fact that the number of emitters issuing both international sukuk and Eurobonds is limited, and most of them are sovereign nations. There are virtually no companies that issue both international sukuk and Eurobonds, so the analysis of corporate securities is restricted or impossible to carry out due to scarcity of data. Also noteworthy is the fact that, while the number of sukuk-issuing companies is greater than the number of sukuk-issuing countries, the former are traded either on the primary market (where quotations are unavailable), or on the domestic market (for which there is a limited secondary market, and time-series long enough to be analysed are unavailable). Owing to these restrictions, the scope of the present paper is limited to sovereign sukuk and Eurobonds listed on international exchanges.


4. Method

The present paper compares the reliability of sukuk in terms of risk against that of the bond by comparing VaR values of portfolios containing these securities in various combinations. Each portfolio consists of five securities, but with a different ratio of sukuk to Eurobonds. The description of the portfolios in question and the way they are built is as follows:

1)   5 Eurobonds and 0 sukuk. This combination can form only one portfolio of 5 Eurobonds and 0 sukuk.

2)   4 Eurobonds and 1 sukuk. This type of portfolio gets constructed by taking type 1 portfolio and substituting a single Eurobond for a sukuk of the same issuer. The total number of possible combinations is

3)   3 Eurobonds and 2 sukuk. This type of portfolio is constructed by taking type 1 portfolio and simultaneously substituting two Eurobonds for two sukuk of the same issuer. The total number of possible combinations is

4)   2 Eurobonds and 3 sukuk. This type of portfolio is constructed by taking type 1 portfolio and simultaneously substituting three Eurobonds for three sukuk of the same issue R. The total number of possible combinations is

5)   1 Eurobonds and 4 sukuk. This type of portfolio is constructed by taking type 1 portfolio and substituting four Eurobonds for four sukuk of the same issuer. The total number of possible combinations is

6)   Eurobonds and 5 sukuk. This combination can form only one portfolio of 0 Eurobonds and 5 sukuk.


In total, this makes for 32 different portfolios that provide the corresponding VaR figures for further analysis.

Analysis of the possibility of gaining diversification benefit from including sukuk into a portfolio and comparing it to that of the bond was be carried out using the Value-at-Risk model. A portfolio’s VaR value indicates its maximum possible loss of value over a given period with a certain probability. Therefore, by taking a portfolio and swapping one Eurobond for one sukuk of the same issuer while leaving other securities in place, it is possible to compare the two VaR values in order to determine which of the instruments, other things equal, is more reliable in terms of risk and yields greater diversification benefit. The lower a portfolio’s VaR, the more reliable in terms of risk it is.

VaR values were measured using the delta-normal method at a 1% level of significance (the possibility of a portfolio having its value fall lower than its VaR figure is 1%) for a time horizon of 1 and 5 days. The delta-normal method is one of the more popular parametric methods of calculating portfolio VaR. This method assumes that asset returns are normally distributed, which, in a stable economy with few major shocks, is consistent with the theory of finance. The present paper likewise assumes that returns on all securities to be analysed are normally distributed. For the returns distribution curves of the selected securities see Appendix, Diagrams 1 to 10.


Portfolio VaR was calculated in three stages. First, each sukuk’s and Eurobond’s return was calculated using a lognormal formula

where
P2 is the instrument’s new price
P1 is the instrument’s previous price


Then, within a portfolio, individual susuk and Eurobond’s VaR values were calculated using the formula

where
α is the standard normal deviation (that is, 2.33 for a 1% level of significance, or a 99% confidence level);
σi is the standard deviation of the security no. i
Posi is the investment in security no. i
T is the time horizon (in this case, either 1 or 5 days)
It should be noted that, given the assumed normal distribution of returns for these securities, their theoretical mean distribution would be zero. The proportion of each security in a portfolio is equal. This stage produced for each portfolio a column vector consisting of five VaR values (given the 5 securities in each portfolio).


The final stage saw the calculation of portfolio VaR using the formula

where
(VaR)T is the transposed column vector of individual VaR values;
R is the correlation matrix for all 5 securities in a portfolio;
(VaR) is the column vector of individual VaR values.

The study analysed all possible portfolio combinations, a total of 32 portfolios. See Appendix, Tables 2.1 and 2.2 for calculated portfolio VaR (by what percentage the value of a portfolio can fall within the selected time horizon, compared to its initial value at 99% probability).


5. Findings

  1. For the 1-day and 5-day time horizons, the sukuk-only portfolio was found to have a lower VaR than the Eurobonds-only portfolio. For the 1-day time horizon, sukuk-only portfolio VaR (0.4730%) was 0.0777 percentage points lower than a Eurobonds-only portfolio VaR (0.5507%). For the 5-day time horizon, sukuk-only portfolio VaR (1.0576%) was 0.1739 percentage points lower than a Eurobonds-only portfolio VaR (1.2315%). These VaR figures show that for 1- and 5-day time horizons, maximum value loss for a sukuk-only portfolio would, with a 99% probability, be lower than that of a Eurobonds-only portfolio. This indicates that a sukuk-only portfolio is more reliable in terms of risk than a Eurobonds-only portfolio. This may be explained by the fact that the return on a sukuk is underpinned not only by its issuer’s credit rating, but also by the underlying asset, a portion of which its holder is entitled to. This finding corroborates the theoretical conclusions of Hanif, 2014.
  2. By comparing average portfolio VaR for different sukuk and Eurobond combinations, it is possible to say that, on average, substituting sukuk for a Eurobond of the same issuer lowers the portfolio’s VaR (see Appendix, Tables 3.1 and 3.2). This indicates that, on average, substituting sukuk for a Eurobond of the same issuer makes a portfolio more reliable in terms of risk, that is, on average, it lowers the maximum possible loss in portfolio value. Nevertheless, it should be noted that this does not always apply to individual emitters, since in case of Bahrain, for example, substituting sukuk for one of its sovereign bonds leads to a rise in portfolio VaR. This is true for both the 1-day and 5-day time horizons. Also of note is that, analysing changes in average VaR, each subsequent same-issuer Eurobond to sukuk substitution seems to produce a greater drop in VaR, that is, the maximum loss in portfolio value. This means that the more sukuk-heavy a portfolio is, the more pronounced the drop in VaR from same-issuer Eurobond to sukuk substitution will be.
  3. Analysing the resulting portfolio VaR, it is possible to conclude that the lowest value corresponds to a portfolio consisting of four sukuk and one Eurobond (a Pakistani Eurobond plus sukuk of other nations). This result is shown in Tables 3.1 and 3.2. A slightly higher VaR corresponds to a portfolio of three sukuk and two Eurobonds and two sukuk and three Eurobonds. This allows us to conclude that, in terms of risk, the most reliable portfolio composition is a mixed one, that is, a portfolio containing both sukuk and Eurobonds, since its VaR is lower than that of sukuk-only or Eurobonds-only portfolios. This finding corroborates the conclusions of Hassan, 2012, and Cakir and Raei, 2007.


6. References

  1. Khalid Abbasher Hassan. Comparison between Sukuk and Conventional Bonds: Value at Risk Approach // University of Westminster, September 2012; available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2215194
  2. Muhammad Hanif. Differences and Similarities in Islamic and Conventional Banking // National University of Computer & Emerging Sciences, April 2014; available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1712184
  3. Selim Cakir and Faezeh Raei. Sukuk vs. Eurobonds: Is There a Difference in Value-at-Risk? // International Monetary Fund Working Paper, October 2007; available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1087156


7. Appendices

Appendix 1. Table 1. Sovereign sukuk and Eurobonds used in the study

Issuer Type ISIN Date of issue Expiry date Warrant
Bahrain
CBB International Sukuk Sukuk Ijara XS0708899272 20.11.2011 20.11.2018 6.273% fixed
Bahrain, 2010 Eurobond XS0498952679 31.03.2010 31.03.2020 5.5% fixed
Pakistan
Pakistan International Sukuk Sukuk Ijara XS1147732553 26.11.2014 26.11.2019  6.75% fixed  
Pakistan, 2014 Eurobond XS1056560763 15.04.2014 15.04.2019 7.25% fixed
Qatar

State of Qatar Sukuk Sukuk Ijara XS0801656330 18.07.2012 18.01.2023 3.241% fixed
Qatar, 2009 Eurobond XS0423038875 09.04.2009 09.04.2019 6.55% fixed
Malaysia
Malaysia Sukuk Sukuk Ijara USY5749LAA99 15.04.2015 22.04.2025 3.043% fixed
Malaysia, 2015 Eurobond MYBMO1500010 15.09.2015 15.09.2025 3.955% fixed
UAE
Dubai DOF Sukuk Sukuk Ijara XS0778097088 02.052012 02.05.2017 4.9% fixed
Government of Dubai Eurobond XS0546428144 04.10.2010 04.10.2020 7.75% fixed


Appendix 2. 

Table 2.1. Portfolio VaR for a 1-day time horizon (in % of initial portfolio value)

Portfolio composition  









16 eb:B,P,Q
suk:M,UAE
26 eb:M,UAE
suk:B,P,Q 



















15 eb:B,P,M
suk:Q,UAE
25 eb:Q,UAE
suk:B,P,M 
14 eb:B,P,UAE
suk:Q,M
24 eb:Q,M
suk:B,P,UAE
13 eb:B,Q,M
suk:P,UAE
23 eb:P,UAE
suk:B,Q,M
12 eb:B,Q,UAE
suk:P,M
22 eb:P,UAE
suk:B,Q,M
6 eb:B,P,Q,M
suk:UAE
11 eb:B,M,UAE
suk:P,Q
21 eb:P,Q
suk:B,M,UAE
31 eb:UAE
suk:B,P,Q,M
5 eb:B,P,Q,UAE
suk:M
10 eb:P,Q,M
suk:B,UAE
20 eb:B,UAE
suk:P,Q,M
30 eb:M
suk:B,P,Q,UAE
4 eb:B,P,M,UAE
suk:Q
9 eb:P,Q,UAE
suk:B,M
19 eb:B,M
suk:P,Q,UAE
29 eb:Q
suk:B,P,M,UAE
3 eb:B,Q,M,UAE
suk:P
8 eb:P,M,UAE
suk:B,Q
18 eb:B,Q
suk:P,M,UAE
28 eb:P
suk:B,Q,M,UAE
1 B,P,Q,M,UAE 2 eb:P,Q,M,UAE
suk:B
7 eb:Q,M,UAE
suk:B,P
17 eb:B,P
suk:Q,M,UAE
27 eb:B
suk:P,Q,M,UAE
32 B,P,Q,M,UAE
  Portfolio VaR   suk (0) eb (5)   suk (1) eb (4)   suk (2) eb (3)   suk (3) eb (2)   suk (4) eb (1)   suk (5) eb (0)
1  0.5507  0.5795 7  0.6096 17  0.4434 27 0.4811 32 0.4730
  3 0.5857 8 0.5702 18 0.4913 28 0.4391  
4 0.5422 9 0.5462 19 0.5208 29 0.4849
5 0.5179 10 0.5000 20 0.5498 30 0.5157
6 0.5001 11 0.5810 21 0.4561 31 0.5719
  12 0.5507 22 0.4800  
13 0.5346 23 0.5405
14 0.5132 24 0.5308
15 0.4815 25 0.5740
16 0.4587 26 0.6039

Key: eb – number of Eurobonds in a portfolio; suk - number of sukuk in a portfolio; B – Bahraini security; P – Pakistani security; Q – Qatari security; M – Malaysian security; UAE – UAE security.


Table 2.2. Portfolio VaR for a 5-day time horizon (in % of initial portfolio value)

Portfolio composition




































16 eb:B,P,Q
suk:M,UAE
26 eb:M,UAE
suk:B,P,Q





 
15 eb:B,P,M
suk:Q,UAE
25 eb:Q,UAE
suk:B,P,M
14 eb:B,P,UAE
suk:Q,M
24 eb:Q,M
suk:B,P,UAE
13 eb:B,Q,M
suk:P,UAE
23 eb:P,UAE
suk:B,Q,M
12 eb:B,Q,UAE
suk:P,M
22 eb:P,M
suk:B,Q,UAE
6 eb:B,P,Q,M
suk:UAE
11 eb:B,M,UAE
suk:P,Q
21 eb:P,Q
suk:B,M,UAE
31 eb:UAE
suk:B,P,Q,M
5 eb:B,P,Q,UAE
suk:M
10 eb:B,M,UAE
suk:P,Q
20 eb:B,UAE
suk:P,Q,M
30 eb:M
suk:B,P,Q,UAE
4 eb:B,P,M,UAE
suk:Q
9 eb:P,Q,UAE
suk:B,M
19 eb:B,M
suk:P,Q,UAE
29 eb:Q
suk:B,P,M,UAE
3 eb:B,Q,M,UAE
suk:P
8 eb:P,M,UAE
suk:B,Q
18 eb:B,Q
suk:P,M,UAE
28 eb:P
suk:B,Q,M,UAE
1 B,P,Q,M,UAE 2 eb:P,Q,M,UAE
suk:B
7 eb:Q,M,UAE
suk:B,P
17 eb:B,P
suk:Q,M,UAE
27 eb:B
suk:P,Q,M,UAE
32 B,P,Q,M,UAE
  suk (0) eb (5) suk (1) eb (4) suk (2) eb (3) suk (3) eb (2) suk (4) eb (1) suk (5) eb (0)
Portfolio VaR 1 1.2315 2 1.2957 7 1.3631 17 0.9915 27 1.0757 32 1.0576
  3 1.3097 8 1.2749 18 1.0987 28 0.9819  
4 1.2124 9 1.2213 19 1.1646 29 1.0842
5 1.1580 10 1.1181 20 1.2294 30 1.1531
6 1.1182 11 1.2991 21 1.0198 31 1.2789
  12 1.2315 22 1.0732  
13 1.1953 23 1.2086
14 1.1476 24 1.1870
15 1.0766 25 1.2836
16 1.0257 26 1.3504

Key: eb – number of Eurobonds in a portfolio; suk - number of sukuk in a portfolio; B – Bahraini security; P – Pakistani security; Q – Qatari security; M – Malaysian security; UAE – UAE security.


Table 3.1. Portfolio VaR and average VaR for different sukuk-Eurobonds portfolio combinations for a 1-day time horizon (in % of initial portfolio value)

Portfolio combination suk (0) eb (5) suk (1) eb (4) suk (2) eb (3) suk (3) eb (2) suk (4) eb (1)  suk (5) eb (0)
Portfolio VaR







0.5507 0.5795 0.6096 0.4434 0.4811 0.4730
  0.5857 0.5702 0.4913 0.4391  
  0.5422 0.5462 0.5208 0.4849  
  0.5179 0.5000 0.5498 0.5157  
  0.5001 0.5810 0.4561 0.5719  
    0.5507 0.4800    
    0.5346 0.5405    
    0.5132 0.5308    
    0.4815 0.5740    
    0.4587 0.6039    
Average combination VaR 0.5507 0.5451 0.5346 0.5191 0.4985 0.4730
Average VaR change (in percentage points) -0.0056 -0.0105 -0.0155 -0.0205  -0.0255
Min. VaR of the combination 0.5507 0.5001 0.4587 0.4434 0.4391 0.4730


Table 3.2. Portfolio VaR and average VaR for different sukuk-Eurobonds portfolio combinations for a 5-day time horizon (in % of initial portfolio value)

Portfolio combination suk (0) eb (5) suk (1) eb (4) suk (2) eb (3) suk (3) eb (2) suk (4) eb (1) suk (5) eb (0)
Portfolio VaR







1.2315 1.2957 1.3631 0.9915 1.0757 1.0576
  1.3097 1.2749 1.0987 0.9819  
  1.2124 1.2213 1.1646 1.0842  
  1.1580 1.1181 1.2294 1.1531  
  1.1182 1.2991 1.0198 1.2789  
    1.2315 1.0732    
    1.1953 1.2086    
    1.1476 1.1870    
    1.0766 1.2836    
    1.0257 1.3504    
Average combination VaR 1.2315 1.2188 1.1953 1.1607 1.1148 1.0576
Average VaR change (in percentage points) -0.0127 -0.0235 -0.0346 -0.0459 -0.0572
Min. VaR of the combination 1.2315 1.1182 1.0257 0.9915 0.9819 1.0576


Diagram 1. Distribution of returns, CBB International Sukuk


Diagram 2. Distribution of returns, Bahrain, 2010 bond


Diagram 3. Distribution of returns, Pakistan International Sukuk


Diagram 4. Distribution of returns, Pakistan, 2014 bond


Diagram 5. Distribution of returns, State of Qatar Sukuk


Diagram 6. Distribution of returns, Qatar, 2009 bond


Diagram 7. Distribution of returns, Malaysia Sukuk


Diagram 8. Distribution of returns, Malaysia, 2015 bond


Diagram 9. Distribution of returns, Dubai DOF Sukuk


Diagram 10. Distribution of returns, Government of Dubai bond

 

 

скачать dle 10.2 Авто Тюнинг кузова