Islamic versus Conventional Bond Announcements: An Investigation of the Wealth Effect

Transcription

1. Islamic versus Conventional Bond Announcements : An Investigation of the Wealth Effect Ghadeer M. Khartabiel School of Business Innovation & Technopreneurship Universiti Malaysia Perlis Kampus Kubang Gajah, 02600 Arau, Perlis Malaysia Email: ghadeer.khartabiel@gmail.com Ahmad Abu-Alkheil School of Management and Logistic Sciences German Jordanian University (GJU) P.O. Box 35247, Amman 11180, Jordan Email: ahmad.abualkheil@gju.edu.jo Tunku Salha Tunku Ahmad School of Business Innovation & Technopreneurship Universiti Malaysia Perlis Kampus Kubang Gajah, 02600 Arau, Perlis Malaysia Email: salha@unimap.edu.my Walayet A. Khan Schroeder Family School of Business Administration University of Evansville 1800 Lincoln Avenue, Evansville, IN 47722 Email: wk3@evansville.edu
2. Islamic versus Conventional Bond Announcements : An Investigation of the Wealth Effect Abstract We employ standard event study methodology to examine the stock price reaction to the issuance announcements of 237 Islamic bonds (sukuk) versus 231 conventional bonds issued in twelve financial markets from 2005 to 2017. We further examine the effects of issuance announcement on shareholder wealth in multiple economic settings, pre-crisis, during the crisis, and post-crisis, using the recent financial crisis (2007-2009) as a benchmark. Using multiple regression model, we also attempt to identify the potential explanatory variables for the wealth effects. Our findings support existing literature: there is an insignificant market reaction for the announcements of Sukuk as well as conventional bonds in the pre-global financial crisis period of 2008. During the crisis period, market reaction is significantly negative for both groups. In the post-crisis period (which is the longest period and the largest sub-sample), the market reaction for Sukuk is positive and significant, apparently due to market participants’ new look, awareness and increased demand for Islamic financial products, whereas for conventional bonds the market reaction is insignificant. Additionally, our study finds supporting results for the signalling theory and the asymmetric information theory. Offer size has a significant positive impact on stock returns. In the pre-crisis period, large firms (with a higher asset base) generated abnormal returns. For the periods during and post crisis, large Sukuk issuers had higher abnormal returns. In addition, significant leverage is found for issuing bonds in post and during crisis periods. Keywords: sukuk, bonds, event study, issuance announcement, returns and wealth. JEL classification: G14, G31, G32 1. Introduction Sukuk, the “Shari’ah compatible investment bonds” are an innovative and essential component of the global Islamic financial industry (GIFI). The Islamic Finance Bulletin/ King Fahd University (2016) defines sukuk as “certificates of equal value representing undivided shares in the ownership of tangible assets, usufructs and services or the assets of particular projects or special investment activity”. Although sukuk constitute 5% of the total size of the GIFI, they exhibit a 27.1% annual growth rate from 1995 through 2016 compared to 7.2% by 2
3. conventional bonds . In 2016, the global Sukuk issuance increased to $ 88.3 billion, a 44% jump in volume from 2015. The increase in volume was largely driven by the steady issuances from Asia, Gulf Cooperation Council (GCC) countries, Africa, Malaysia, Indonesia, and Turkey (Sukuk report, 2017). In 2017, the sukuk market increased by 45.3% to reach $97.9 billion primarily because of massive issuance in some GCC countries (Ratings Direct 2018). Rapid expansion of the Islamic banking sector has played a key role in the growth of sukuk issuances. To make their funding profiles more robust, Islamic banks frequently seek sukuk as an alternative to conventional bonds. Additionally, Qatar has initiated ambitious development plans, including the infrastructure required to host the 2022 FIFA World Cup. These plans have generated financing needs of $278 billion. The long lead times present a challenge that longer-duration sukuk are ideally suited to meet (Markaz Research, Oxford Business Group, 2018). The likelihood of an increase in the use of sukuk in the short to medium term in all GCC countries has significantly increased due to the decline in oil prices (beginning in the second half of 2014) as well as fiscal surpluses. According to Thomson Reuters (2017), the drop in oil prices is the main reason for sukuk market volatility. The revenue shortfall from such a drop led to more issuances of sukuk in GCC countries in order to meet current and capital spending needs. The drop in oil prices also correlates to credit downgrades for some sovereigns such as the GCC nations. With oil prices remaining low, the deficits are expected to grow and further downgrades are expected. Following the global financial crisis, the governments and companies felt a need to diversify funding options. The Western financial market has seen an increase in the use of sukuk by non-Shari’ah-compliant investors seeking diversification and attractive yields (Lee 2017). The UK issued £200 million sovereign sukuk in mid-2014, which was the first for a Western nation. The Sukuk issuance is gaining a foothold in non-Muslim Asian and African countries as well. The governments of Singapore, Luxembourg and Hong Kong issued their first sukuk in 2014. More recently, South Africa, Nigeria and the Ivory Coast have introduced legal and tax changes to make it easier for deficit units to issue sukuk (Oxford Business Group, The Report: Qatar 2015). Another factor for the growth of sukuk issuance i s the requirement of establishing a quality-funding base by Basel III rules. The Basel regime requires all banks, conventional and Islamic, to hold higher-quality capital to enhance their liquidity. The goal of Basel III is to prepare banks to absorb financial shocks more proficiently than during the global financial crisis period. The Islamic Finance Standard Board (IFSB) has recently drawn up capital adequacy standards for Islamic banks answering the unique requirements of Shariah compliant financing. Taken together, the Basel and IFSB initiatives will be instrumental for Islamic banks to use 3
4. sukuk for regulatory compliance purposes . Consequently, the full implementation of these international standards is widely expected to yield an increase in the issuance of sukuk by Islamic banks. Cakir and Raei (2007) claim that Sukuk are different from bonds as they comply with Islamic law and offer unique risk-reduction benefits when added to a portfolio of fixed income securities. They found that the reduction in value at risk (VAR) is not just due to the inclusion of an extra instrument in the portfolio, but rather is a result of the very different behavior of Sukuk prices in the secondary market compared to conventional bonds. Another difference between Sukuk and conventional bonds is related to asset ownership. Sukuk holders’ own shares of the underlying assets (generally referred to as asset-backed sukuk) while conventional bonds are considered debt obligations. Consequently, conventional bonds disallow such type of ownership (Jamaldeen 2012; Hussain et al. 2015). Nevertheless, Miller, Challoner, and Atta (2007) and Usmani (2007) argue that Sukuk replicate conventional bonds as both have the same levels of cash and risk. Wilson (2008) also argues also that sukuk issuers make efforts to render Sukuk identical to conventional bonds in order to allow unfamiliar investors to assess the risk of such new investments. Such Sukuk bonds just mirror conventional bonds, forcing the innovation notion in Islamic finance product. Moreover, Sukuk face compliance challenges. The Shari’a-based Dana Gas, the Middle East’s leading private sector natural gas company presents a recent example with respect to challenges Sukuk issuers and investors face. Dana Gas launched their sukuk al-Mudarabah in late 2007 for $1 billion. The Sukuk due on 30th October 2012, with a fixed profit rate of 7.5 per cent (LNG World News 2012). Dana Gas defaulted and was unable to pay Sukuk accrued profit due to liquidity difficulties triggered by payment delays from Egypt and Iraq’s Kurdistan region (Hekmatyar and Parkar 2018). Dana Gas used ambiguity over the standardized definition of the Mudarabah sukuk to get out (or to restructure) of its old debts. Dana Gas claimed that these issues were no longer considered Shariah-compliant under the UAE’s law due to changes in the AAOIFI standards of Islamic finance consequently, are unlawful and unenforceable, as the Sharia-based financial standards had changed since the sukuk were issued. Specifically, the company claimed that the sukuk market has evolved and pure Mudarabah have been outdated since 2007. Dana Gas was intensely involved in court battles over the legality of its Sukuk (Bloomberg 2017). The case is not only significant for Dana Gas and its sukuk holders but will also have broader implications for Islamic finance globally. This raises the questions of whether Sukuk are different from conventional bonds or if they play a vital role in the financial market as an alternative source of financing. We attempt to answer these questions in this study. 4
5. Alam et al . (2013) believe that Sukuk create a positive impact on the Islamic financial market. Yet, the impact of issuance of Sukuk announcements on stock prices is still unclear, particularly during economic shocks. Accordingly, this study contributes to the applied researches by examining the wealth effect of Sukuk issue announcements compared to that of conventional bond issue announcements from 2005 to 2017. Our results provide a comparative analysis based on the market's sensitivity to Sukuk and conventional bonds. In doing so, we attempt to resolve some of the controversies surrounding the differences between Islamic bonds and traditional bonds. The remainder of this paper is organized as follows. Section 2 discusses the literature review and theoretical framework. Section 3 explains the sample, data, and research methodology. Section 4 discusses the findings and section 5 presents the conclusion. 2. Literature Review Jobst et al. (2008) believe that Sukuk are instrumental in the development of the Islamic financial sectors. They argue that there is a strong demand for Sukuk from Islamic as well as conventional institutions despite the global financial crisis of 2008. Cakir and Raei (2007) compared the VAR for a portfolio containing both sovereign Sukuk and Eurobonds issued by the same entity. Findings showed a reduction in VAR, and Sukuk indeed offered diversification benefits to the investors. To test the difference in wealth effect of both Sukuk and conventional bonds, we rely on the traditional financial models of the asymmetric information based on the works of Ross (1977), Myers and Majluf (1984) and Jensen (1986). Signalling theory (Ross 1977) is based on information asymmetry – management is better informed about the company than outsiders, creditors or investors. Bond issuance announcements could suggest managements’ sound expectations regarding future cash flows, indicating their ability to fulfill their contractual obligations, thus enabling investors to create value. According to the pecking order theory (Myers 1984), equity announcements reflect management’s belief that their stock is overvalued, whereas bond issuance announcements show management’s confidence about the firm’s future prospectus. Jensen (1986) argues that issuance of debt obligates the firm to pay out cash (interest and principle), and thus management has less free cash flow to promote their self-interests at the expense of shareholders’ interests. Issuing debt has some disadvantages as well. Jensen (1986) and Lasfer (1995) point out that issuance of debt gives management incentive to take on risky projects. If returns prove to be greater than the face value of the project they pocket, the bondholders suffer. Also, debt increases borrowers’ interest payments and bankruptcy risk, and the market may react negatively to bond issuance announcements. 5 Given the advantages and
6. disadvantages of debt issuance , the market reaction becomes an empirical question. Moreover, it is quite possible that in a certain period some advantages outweigh the disadvantages and vice versa. Many researchers have examined the wealth effects of conventional bond announcements, and findings thus far are mixed. Stephen and Ward (1996) examined the stock price reaction to the announcement of 164 high yield straight bond issues. Findings show that the AARs are statistically not significantly different from zero on day +2. Moreover, they illustrated that the stock market does not react differently for the high-yield bond issues that default, as compared to those that do not default. Contradicting the models developed by Myers and Majluf (1984), Miller and Rock (1985), Krasker (1986) and Jensen (1986), they found that stock prices neither positively nor negatively are related to bond issue. Howton et al. (1998) found that the market reacts negatively to straight bond issuances and that the announcement day reaction is inversely related to the level of free cash flow and investment opportunities prior to the debt issue. Harvey et al. (2003) claimed that firms with high managerial agency costs at the time of bond issuance create value for shareholders. Ashhari et al. (2009) examined the wealth effect of Malaysian Sukuk and conventional bonds issuance announcements for the period 2001 to 2006. Results showed that the size of the bond offering was a significant factor in explaining stock returns. However, the sign was negative for Sukuk bonds as compared to conventional bonds. Alam et al. (2013) examined the market reaction of 87 bonds and 79 Sukuk issuance announcements in six developed Islamic financial market over the sample period from 2004-2012, and observed a negative market reaction following the issuance announcements of Sukuk before and after-crisis periods. However, they also found that cumulative abnormal returns for Sukuk issue during the post crisis period were positive and significant. Alam et al. (2013) confirmed the wealth creation for the shareholders of firms offering Sukuk. On other hand, the stock price reaction was positive for the announcement of conventional bonds during pre-crisis period but negative during the post crisis period. While the size of the bond offering had a negative impact on the abnormal returns for sukuk but it had positive impact positive for conventional bonds. Sherif and Erkol (2017) investigated the stock market reaction following the announcements of the issuance of 255 sukuk compared to 205 conventional bonds of non-financial Malaysian firms between 2000 and 2015. They used standard event study methodology on fourteen event windows. Results showed an insignificant difference in the market reaction to fixed-rate sukuk and fixed-rate conventional bond issuance announcements in the pre-crisis as well as the overall period. Findings showed also a highly significant difference in the stock market reaction to the issuance of the two types of bond in the post-global financial crisis period. Albeit our study is an enhancement of Ashhari et al. (2009), Alam et al, (2013), and Sherif and Erkol (2017), but it differs in the following aspects: 1) Our sample size is larger and covers a wide range of financial 6
7. markets . 2) The time period of this study is longer than Ashhari et al. (2009) and Alam et al, (2013) and is more recent. 3) We develop a more extensive country sample consisting of: I) four Pacific rim countries namely; Malaysia, Indonesia, Singapore, and Brunei Darussalam. II) Five GCC nations namely; Bahrain, KSA, UAE, Qatar, and Kuwait. III) Three other countries namely; Turkey, Pakistan, and Gambia. 4) We attempt to replicate Mikkelson and Partch’s (1986) results that state that straight and convertible debts lead to abnormal returns. Considering the above discussion, our study contributes to the literature in different ways. The study attempts to assess existing findings and contributes additional evidence that investigates the reaction of stock prices to two different types of bond announcements. Our study contributes to the current debate on the reliability of Sukuk as potential investment channels that offer risk reduction and create diversification benefits when adding Sukuk to conventional bond portfolio. It also tests their involvement in wealth creation and sheds the light on the relationship of cumulative abnormal returns (CAR) with leverage, free cash flow, firms size, and shariah compliance status. The conclusions attained in this paper enrich the literature in terms of the impact Sukuk issuance triggers in the stock market and create wealth for a longer period with the largest sample of Sukuk and bonds, in particular after the crisis period. 3. Data and Methodology We employ a sample of 468 issues (237 Sukuk and 231 conventional bonds). Data has been extracted from the Bloomberg and Thomson ONE databases. The sample consists of global, regional, corporate and sovereign Sukuk. The sample size is determined by available information on all requested variables especially closing stock prices for firms issuing debt for prior to and on and the actual announcement date of the issuance. This requirement is necessary for the beta estimation in order to calculate the abnormal returns. We also screen the sample to include only firms offering straight bonds and Sukuk. In addition, the sample of bond offerings must have data availability on the size of the bond offering, length of bond maturity, a firm’s debt ratio and a firm’s total assets. To ensure that we capture the pure effect from bond announcements, we excluded those issues bundled with the announcements of other corporate proposals made within seven days of the announcement date (Jothee 2005). The sample of conventional bonds and sukuk has approximately the same ratings and maturity. The sample was chosen from Malaysia, Indonesia, Singapore, Bahrain, Kingdom of Saudi Arabia (KSA), UAE, Qatar, Kuwait, Turkey, Pakistan, Gambia and Brunei Darussalam. These markets have the distinction of being developed Islamic capital markets. Furthermore, to study the effect of the global financial crisis on Sukuk 7
8. and bond issuance announcements on shareholder wealth , we divide the full sample into three windows: precrisis (2005–2007), during crisis (2008–2010), and post-crisis (2011–2017). Following Godlewski et al. (2011) and Alam et al. 2013, we use market model event study methodology to calculate the abnormal returns (AR) around the announcement of Sukuk and conventional bond issuance dates. Abnormal return for each firm i on day t is computed as follows: AR it = ( α i+ βi Rmt ) Rit − (1) where ARit is the abnormal returns on stock i during period t, Rit is the observed returns during period t, Rmt is the market portfolio returns during period t, αi is the constant average return of stock i and βi is the beta estimate of stock i. The beta is estimated by using returns from 60 trading days before and 60 days after the crisis periods (Harvey et al. 2003). Adjusted beta was computed using Blume’s method to ensure accuracy. We follow Alam et al. (2013) and use the date of announcement as day 0. Market model parameters are estimated over the period (−60, 60). This reduces the sample size to firms having at least 120 days of stock returns. Most event studies use daily data as financial markets respond notoriously quickly to new information. Thus, we estimate one-day [0,0], three-day [−1, +1] and seven-day [−3, +3] time event windows and calculate average abnormal daily returns. We obtain cumulative average abnormal returns (CAARs) by taking the sum of the daily excess returns over the respective event windows. Daily abnormal returns on event day t for all sample firms are totaled and divided by the number of observations to give the average abnormal returns on event day t, AARt as in equation (2). n AARt ∑ AR = i,t / nt i (2) To determine the nature of the relationship between variables and abnormal returns associated with debt announcements, the following multivariate regression model (which is free of multicollinearity, serial correlation and heteroscedasticity problems)1 was estimated: CAAR(−3,+3) = l og ¿ ¿ + δ FCF + ƞ LEV + ƛ SHCO + φ BOA +ƍ ¿ ω (3) where: where CAAR is the cumulative average abnormal returns for issuer i from the window period of t 1 to t2. BOA is the ratio of the size of the bond offering divided by the total assets in the pre-announcement period. SIZE 8
9. is the size of the issuing entity estimated by the natural log of its total assets . FCF is the level of free cash flows of the issuer. FCF = (operating income − interest − net taxes)/BV asset, where taxes = current tax − change in deferred tax (Shao et al. 2007). LEV is the level of leverage of the firm, where LEV = total firm debt/total in the pre-announcement period. SHCO is the classification of the issuer where SHCO = 1 if the firm is Shariah compliant bond and 0 otherwise. ω is error term. 4. Findings and Discussion 4.1. Event study analysis Table 1 presents the full sample descriptive statistics of the variables used in the multiple regression model. Statistics show that the mean leverage ratio of (0.6233) is higher for the firms issuing conventional bonds compared to those issuing Sukuk (0.4468). Sukuk-issuing firms are smaller than conventional bond issuing firms with respect to size as well as free cash flow. Additionally, the mean cumulative abnormal returns within the time window −3 to +3 of the announcement date for Sukuk issuers is negative −1.6124% with a higher standard deviation (09.572) compared to the conventional bond issuer firms. ________________________________________________________________________________________ 1 We utilize the Variance Inflation Factor (VIF) to assist in detecting the presence of multi-collinearity. VIF values more than 10 show presence of multi-collinearity in the data. The results of multi-collinearity show the following VIF values: CAAR (3.41), SIZE (2.32), FCF (3.07), and LEV (1.24), while the mean VIF value is found to be (1.37). Therefore, results indicate an absence of multicollinearity of the model. To test for the autocorrelation (serial) the Durbin-Watson test is applied which indicates that the dstatistic = 1.994 (nearly 2) indicating that there is no autocorrelation. Eventually, the Breusch Pagan test is used to test for heteroscedasticity. The chi-square value obtained from the analysis is 434.65 while the p-value is 0.000 therefore, we do not reject the null hypothesis which states the equality of variance, accordingly, we observe no heteroscedasticity. Table 1 Descriptive Statistics of Sukuk and Conventional Bonds for the Period 2005-2017 Variables CAAR SIZE FCF LEV Mean −1.6124 2.3611E8 1.7725E9 0.4468 Sukuk Std. dev. 09.572 6.44120E7 4.055E8 0.13882 N 237 237 237 237 Mean −1.2280 4.1123E7 1.8966E7 0.6233 Conventional bonds Std. dev. 5.39 7.00436E6 6.20160E7 0.19440 N 231 231 231 231 Table 2 presents the estimated cumulative average abnormal returns (CAAR) for the entire sample period and for the three sub-sample periods.2 Results show a lack of significant market reaction to either conventional bonds or Sukuk announcements for both the pre-crisis period and the full sample period. This result is consistent 9
10. with Eckbo (1986) and can be explained by the opposing effects of the stock market's reaction to the issue of bonds. In some periods, advantages of bond issuances are perceived to be almost equal to the disadvantages of bond issuances producing insignificant results. This reflects the indifference of market participants. Table 2 CAAR for Periods 0, 0; −1, 1; −3, +3 Days Event window Overall period (2005–2016) Mean CAAR Sukuk Bond (237)1 (231) −2.358 0.844 (0.496)2 (0.247) (−0.886)3 (0.217) (−0.119)4 (0.593) Pre-crisis period During crisis period (2005–2007) (2008–2010) Mean CAAR Mean CAAR Sukuk Bond Sukuk Bond (47) (39) (67) (55) [0,0] −1.101 0.661 −3.441 −1.044 (0.203) (0.351) (0.041) * (0.037)** (−0.773 (1.122) (−2.33)** (−1.33)* ) (0.383) (−2.24)** (−3.02)** (−0.405 ) [−1,1] −1.983 0.901 −1.212 1.112 −2.472 −2.441 (0.183) (0.244) (0.316) (0.184) (0.036)** (0.0322)** (−1.051) (0.478) (−0.681 (1.423) (−2.023)* (−2.442)** (−0.740) (1.211) ) (0.518) (−2.414)** (−1.973)** (−0.123 ) [−3,3] −1.612 1.283 −1.290 1.666 −4.133 −3.013 (0.444) (0.192) (0.144) (0.305) (0.012)* (0.0262)** (−0.612) (0.099) (−0.813 (0.486) (−1.980)** (−2.497)** (−0.953) (0.238) ) (0.160) (−2.037)** * (−0.633 (−2.238)** ) 1 = Number of issuances; 2 = p-values for t-tests; 3 = BMP-test; 4 = Patell Z. * ** = Significance at 5%; *** = Significance at 10%. Post-crisis period (2011–2017) Mean CAAR Sukuk Bond (123) (102) 0.521 −1.553 (0.06)*** (0.118) (3.136)** (0.721) (7.943)** (0.633) 0.611 (0.044)** (2.603)* (2.523)** −0.534 (0.329) (0.616) (0.511) 0.894 (0.050)* (1.453)** (2.123)** −0.461 (0.441) (0.801) (0.715) = Significance at 1%; _______________________________________________________________________________________ 2 We also follow Brown & Warner, (1985) to test and report the statistical significance of CAARs applying a simple timeseries test. The deviations from the assumption of underlying normal distribution of the time-series test are highly likely in event studies. Accordingly, we additionally apply various robust test statistics such as the Patell (1976) standardized residuals test that is robust to heteroscedastic event period abnormal returns and the standardized cross-sectional test introduced by Boehmer et al. (1991), which is robust to event-induced variance increases. The CAARs of Sukuk and conventional bond issues, however, are negative and significant in the crisis period (2008-2010). One possibility is that there is a downward trend in prices that the estimation for abnormal returns is not fully removing. In their literature, Myers and Majluf (1984) and Miller and Rock (1985) offer several explanations to explain this negative impact. Gosh et al. (1990) contend that an increase in debt is taken as a diversion of the future cash flows (in the form of periodic fixed payments and principle) to bondholders. Consequently, shareholders might perceive it negatively. Further, eventual reduction in ownership concentration may contribute a negative price reaction to debt issues (Gosh et al. 1990). 10
11. Negative market reaction during the announcements of debt securities confirms the loss of investors ' faith in debt financing. However, it should be noted that the crisis period was relatively short. Additionally, it was chaotic for all financial markets and products, with the most likely exception being credit default swaps (CDSs) which offer protection against the default of mortgage-backed bonds, which gained value in the turbulent market conditions. CAARs for Sukuk issues during the post-crisis period are positive and significant which confirms wealth creation for the shareholders of firms offering Sukuk. This finding supports the results of Mikkelson and Partch (1986). However, no wealth effect is found for conventional bond issuance announcements during the post-crisis period, which advocates the findings of Eckbo (1986). This evidence indicates a change in market perception toward sukuk from the pre to post-crisis periods. Explanations for the high demand include the more appealing tax incentives and pricing, recent liquidity advantage 3 and active local markets as well as government backing in Malaysia and the gulf region, both of which dominate the global market for sukuk (Sharif and Erkol 2017). 4.2. Cross Sectional Regression Analysis Table 3 presents the regression results for the overall period. For sukuk issuance, the coefficient for BOA is 0.0021 and significant at 1% while it is 0.0141 and significant at 5% for conventional bonds, indicating a significant positive relationship between issue size and CAAR for bond and sukuk issuance announcements. This supports the signalling theory suggesting that investors perceive a more substantial conventional bond issue as an indicator __________________________________________________________________________________________ 3 Historically, low liquidity has been a constraint in attracting sukuk investors. However, changing perception and rising visibility have boosted demand for sukuk thus improving their tradability and liquidity. Furthermore, more sukuk are being rated and listed on global exchanges, leading to a gradual increase in the liquidity (Aljazira Capital 2014). of a firm’s improved performance and ability to fulfil its payment obligations. Results also support agency conflict theory, which states that higher debt reduces agency conflict. That is to say, the larger the issuance, the less the free cash management will have to mess around, thus, minimizing the agency problem. Results further show a significant negative relationship between free cash flow (FCF) and CAAR for sukuk issuance, indicating issuers with more FCF transmit adverse signals to the financial market. Management has more cash at their disposal to pursue their own interests. Table 3 11
12. Regression Results for Sample 2005 –2017 Variable Sukuk Conventional bonds Constant 2.33 (0.254) 0.612 (0.551) BOA 0.0021 (0.0321)* 0.0141 (0.012)** Log SIZE 0.036 (0.543) 0.0333 (0.432) FCF −0.083 (0.037)** 0.0261 (0.352) LEV −8.116 (0.302) −3.041 (0.511) SHCO 0.001 (0.426) −0.878 (0.511) R-square 0.263 0.412 N 237 231 Note. Value in parentheses denotes significance value *Significance at 1% ** Significance at 5% Table 4 presents the results for the pre-crisis period. For sukuk, the coefficient of BOA is -0.0128 and is significant at 5%. The finding of negative significant stock market reaction could be attributed to market participants’ perception, though correct, that sukuk violated (in the pre and crisis periods) at least the prohibition of fixed returns, mirroring fixed-rate conventional bonds (Sherif and Erkol 2017). By contrast, the results show a significant positive relationship between the firm size and CAAR for sukuk and conventional bond issuance. The results support the trade-off theory, which states that the firms’ higher asset base can offer higher collateral, thus attracting lower interest rates and better terms. Table 4 Regression Results for Pre-Crisis Sample (2005-2007) Variable Sukuk Conventional bonds Constant 5.10 (0.327) 13.144 (0.324) BOA −0.0128 (0.0211)** 0.0327 (0.030)*** Log SIZE 0.985 (0.871)* 0.0047 (0.072)** FCF −0.223 (0.664) −2.921 (0.443) LEV −18.877 (0.120) −23.366 (0.132) SHCO 0.001 (0.443) −5.010 (0.223) R-square 0.321 0.466 N 47 39 Note. Value in parentheses denotes significance value * Significance at 1% ** Significance at 5% *** Significance at 10% Table 5 shows the results for the crisis period. The market reaction to sukuk issuance is significantly and positively related to the size of the issue. These results support signalling and agency cost theories. Findings also revealed a significant negative relation between LEV and CAAR in case of both sukuk and conventional bonds. This suggests that highly levered firm issuance announcements were not viewed positively, particularly during the uncertain market turbulence arising from the market’s perception of increased bankruptcy and financial distress costs. High leverage indicates a high distress risk. Firms in competitive industries can enjoy higher risk 12
13. adjusted returns by reducing their leverage levels (Adamia et al. 2010). In addition, financial leverage contributes more to the dynamics of stock volatility for a firm exposed to both idiosyncratic risk and market risk (Aydemir et al. 2007). Table 5 Regression results for crisis sample (2008-2010) Variable Sukuk Conventional bonds Constant −0.111 (0.432) −7.455 (0.025) BOA 1.033 (0.033)** 0.0943 (0.253) Log SIZE 0.017 (0.440) 0.787 (0.171) FCF −0.361 (0.662) −1.448 (0.721) LEV -1.447 (0.487)* −1.443 (0.031)** SHCO 0.001 (0.612) 1.604 (0.521) R-square 0.428 0.334 N 67 55 Note. Value in parentheses denotes significance value *Significance at 1% ** Significance at 5% Eventually, Table 6 shows regression results for the post-crisis period. Findings again show strong evidence supporting stronger market reaction to sukuk offering size compared to conventional bond issuances. Results also reveal that LEV again have a significant negative relationship with CAAR for both sukuk and conventional bond issuance. Table 6 Regression Results for Post-Crisis Sample (2011-2017) Variable Sukuk Conventional bonds Constant 3.339 (0.501) 1.337 (0.641) BOA 0.0422 (0.342)* 0.0822 (0.043) logSIZE 1.911 (0.666) 1.776 (0.476) FCF −1.412 (0.069) 2.112 (0.155) LEV −10.325 (0.451)* −8.002 (0.061)** SHCO 0.001 (0.316) 0.463 (0.943) R-square 0.257 0.283 N 123 102 Note. Value in parentheses denotes significance value *Significance at 1% ** Significance at 5% 5. Summary & Conclusion We utilize a standard event study methodology (with beta refinement using Blume’s method) to examine the effect of Islamic bond (sukuk) and conventional bond announcements on shareholders’ wealth from 2005 to 2016 and divided into three time windows: 2005–2007 (pre-crisis); 2008–2010 (during crisis); and 2011-2017 13
14. (post-crisis). The sample consists of 237 sukuk and 231 conventional bonds issued in 12 financial markets. Utilizing the multivariate regression approach, we also attempt to determine the nature of the relationship of firms’ size, leverage, level of free cash flows and the Shariah compliant status, with abnormal returns associated with debt announcements. Findings show an absence of significant market reaction to both Sukuk and conventional bonds announcements during the pre-crisis period and the full sample period from 2005 to 2017. In the crisis period, we observe significant negative CARRs whereas in the post-crisis period, the announcement effect of sukuk is significant and positive, but there is an insignificant effect for the announcements of conventional bonds. To understand the effects of bond issuance, we refer to numerous theories including signalling theory (Ross 1977), Pecking order theory (Myers 1984), Trade off theory (Modigliani and Miller 1950), and agency cost theory (Jensen 1986). There are advantages and disadvantages associated (empirically documented) with the issuance of debt which rely heavily on the market conditions and the perception of market participants. We observe no effect in the pre-announcement (normal) period due to market participants netting out the benefits and costs. In the crisis period, we observe significant negative market reaction. During this time, all financial markets were in chaos and experiencing heavy loss. Given that even in normal times debt issuance is considered a diversion of future cash flows in the form of payments, the market reaction was much tenser during the crisis period. However, the jolting experience of the crisis period gave a new awareness about the Islamic financial products, which were less vulnerable due to restrictive screening, prohibition on speculation and usury (interest) in equity mutual funds, and profit and loss sharing arrangements in debt instruments. After the crisis, there was a substantial increase in demand of sukuk, particularly from the rich Gulf States. The market participants showed added interest in Islamic financial products, and we observe significant positive market reaction on the announcements of sukuk during the post-crisis, whereas there was an insignificant reaction following the announcements of conventional bonds during this period. Both issuers and investors reassessed sukuk because of the certificate’s unique characteristics of profit & loss sharing (in a predetermined formula) and asset ownership. In the post-crisis period, many issuers, particularly from the rich gulf regions, showed an increased interest in issuing sukuk to raise funds. There was increase in demand for sukuk by investors as well (Alam et al. 2013). Ernst & Young (2012) note that the positive effect during the post-crisis period is due to the huge demand for asset-backed sukuk in the global market. 14
15. As far as determinants of CAARs are concerned , we find mixed results for the entire sample. The bigger the sukuk and conventional bonds issuance (offer) size, the higher the CAARs, which supports the signalling theory. Moreover, sukuk issuers with more FCF transmit adverse signals to the financial markets. In the pre-crisis, there still is a negative impact of Sukuk issuance announcements. This could be attributable to the market participants’ perception, though correct, that sukuk violated (in the period pre and during crisis) at least the prohibition of fixed returns, mirroring fixed-rate conventional bonds. By contrast, a significant positive relationship is found between the firm size and CAAR for conventional bond issuance. This supports the trade-off theory. The offer size is still significant and positive for sukuk during the crisis period. These results support signalling and agency cost theories. Additionally, in terms of the leverage (LEV) factor, findings revealed a significant negative relation between leverage and CAAR for sukuk and conventional bonds during the crisis and post-crisis periods. Our results overall have important policy implications for individual and institutional investors as well as the policy makers. Investors can use Sukuk to diversify their investment portfolios to mitigate risk. Another implication for international investors is that sukuk enjoy competitive advantages in comparison to other conventional bonds. They target a segment of the global markets that has not been penetrated by other rivals. Given that sukuk command greater liquidity (in Islamic countries), a steady stream of cash flow, anchored in actual tangible assets, and much safer than conventional bonds because large portion of sukuk are linked to governments, policy makers should design new strategies to modernize the operational aspects of Sharia'ah compliant sukuk markets. References Alam N, Hassan M K, Haque M A (2013) Are Islamic bonds different from conventional bonds? International evidence from capital market tests. Borsa Istanbul Review 13: 22-29. Aljazira Capital (2014) Sukuk: The Sharia Fixed Income Alternative Strategy. Strategy report http://www.aljaziracapital.com.sa/report_file/ess/spe-171.pdf. Accessed 24 June 2018 Ashhari Z M, Chun L S, Nassir A M (2009) Conventional vs Islamic Bond announcements: the effects on shareholders’ wealth. International Journal of Business and Management 4(6): 105-111. 15
16. Aydemir A C , Gallmeyer M., Hollifield B (2007) Financial Leverage and the Leverage Effect—A Market and Firm Analysis. Paper 142, Tepper School of Business, Pittsburgh. http://repository.cmu.edu/tepper/142. Accessed 24 June 2018 Bloomberg (2017) Dana Gas Is Said to Miss Payment on $700 Million Sukuk Today. By Matthew Martin, Archana Narayanan, and Arif Sharif. Available at: https://www.bloomberg.com/news/articles/2017-1030/dana-gas-is-said-to-miss-payment-on-700-million-sukuk-tomorrow. Accessed 24 June 2018 Boehmer E, Musumeci J, Poulsen A (1991) Event study methodology under conditions of event induced variance. Journal of Financial Economics 30: 253-272. Brown S, Warner J (1985) Using daily stock returns: the case of event studies. Journal of Financial Economics 14: 3-31. Cakir S, Raei F (2007) Sukuk vs. Eurobonds: Is there a difference in value-at-risk?. International Monetary Fund Working Paper WP/07/237. Dutordoir M, Strong N, Ziegan M (2014) Does corporate governance influence convertible bond issuance? Journal of Corporate Finance 24: 80-100 Eckbo B E (1986) Valuation effects of corporate debt offerings. Journal of Financial Economics 15: 119-151. Ernst & Young MENA (2012) Global Demand for Sukuk to Reach US$900bn by 2017. http://inaudit.com/economy/global-demand-for-sukuk-to-reach-us900bn-by-2017-ey-25080/ . Accessed 24 June 2018 Godlewski C J, Turk-Ariss R., Weill L (2010) Are Islamic investment certificates special? Evidence on the postannouncement performance of Sukuk issues. Working Papers of LaRGE Research Center Gosh C, Varma R, Woolridge R (1990) An analysis of exchangeable debt offers. Journal of Financial Economics 28(1): 251–263. Harvey C R, Lins K V, Roper A H (2003) The effect of capital structure when expected agency costs are extreme. Journal of Financial Economics 74: 3-30. Hekmatyar, M S and Parkar, E (2018) An Evaluation of Dana Gas’s Mudarabah Sukuk from Shariah and Legal Perspectives. European Journal of Islamic Finance 9: 1-9. Howton S D, Howton S W, Perfect S B (1998) The market reaction to straight debt issues: the effects of free cash flow. Journal of Financial Research 21:219-228. http://www.iifm.net/system/files/private/en/IIFM%20Sukuk%20Report%20%286th%20Edition %29_2.pdf Hussain M, Shahmoradi A, Turk-Ariss R (2015) An Overview of Islamic Finance. IMF Working Paper No. 15/120. https://ssrn.com/abstract=2623805. . Accessed 24 June 2018 International Islamic Financial Market (2017) Sukuk report: A comprehensive study of the global sukuk market. 6th Edition. Jamaldeen F (2012) Islamic finance for dummies. Hoboken, New Jersey: John Wiley and Sons, Inc. Jensen M (1986) Agency costs of free cash flow, corporate finance, and takeovers. American Economic Review 76:323-329. Jobst A, Kunzel P, Mills P, Sy A (2008) Islamic bond issuance e what sovereign debt managers need to know? International Monetary Fund Working Paper PDP/08/3. Jothee S (2005) Determining the effect of market microstructure changes on time to equilibrium: a Malaysian stock market experience. Paper presented at seminar of Faculty of Economics and Management, Universiti Putra Malaysia, Serdang, Malaysia. King Fahd University of Petroleum and Minerals (2016) Islamic Finance Bulletin. Center of Research Excellence for Islamic Banking and Finance. http://cim.kfupm.edu.sa/ibf2017/documents/Bulletin.pdf. Accessed 24 June 2018 Lasfer M A (1995) Agency costs, taxes and debt: the UK evidence. European Financial Management 1(3): 265– 285. Lee Y N (2017) Islamic finance is becoming so attractive that even non-Muslims want in. CNBC. https://www.cnbc.com/2017/12/19/sukuk-islamic-finance-grows-in-non-muslim-countries.html. . Accessed 24 June 2018 LNG World News (2012) Dana Gas Announces Terms of USD 1 Billion Sukuk Refinancing, UAE. Retrieved on 19 October 2018 from https://www.lngworldnews.com/dana-gas-announces-terms-of-usd-1-billionsukuk-refinancing-uae/ Markaz Research, Oxford Business Group (2018) Qatar's sukuk market looks promising. https://oxfordbusinessgroup.com/analysis/qatars-sukuk-market-looks-promising. . Accessed 24 June 2018 Mikkelson W H, Partch M M (1986) Valuation effects of security offerings and the issuance process. The Journal of Financial Economics 15: 31-60. Miller M H, Rock K (1985) Dividend policy under asymmetric information. The Journal of Finance 40(4):1031– 1051. Miller N D, Challoner J, Atta A (2007) UK welcomes the Sukuk. International Financial Law Review 26: 24-25. 16
17. 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. Oxford Business Group (2016) The Report: Qatar. https://oxfordbusinessgroup.com/qatar-2016. Accessed 24 June 2018 Patell J (1976) Corporate forecasts of earning per share and stock price behavior: empirical tests. Journal of Accounting Research 14: 246-276. Ross S (1977) The determination of financial structure: the incentive signaling approach. Bell Journal of Economics 8: 23-40. S&P global (2018) Global Sukuk Market Outlook: Another Strong Performance In 2018? RatingDirect.. https://www.spratings.com/documents/20184/86957/Global+Sukuk+Market+Outlook+2018/d1b1aa9b187f-44c0-8d9c-c1109f5c6890. Accessed 24 June 2018 Shao C C, Sheng S C, Hsing A, Chia W H (2007) Investment opportunities, free cash flow, and stock valuation effects of secured debt offerings. Review of Quantitative Finance and Accounting 28: 123-145. Sherif M, Erkol C T (2017) Sukuk and conventional bonds: shareholder wealth perspective. Journal of Islamic Accounting and Business Research, 8(4): 347-374. Sivaprasad S, Muradoglu Y G, Gough O, Adami R (2010) The Leverage Effect on Stock Returns (October 10). https://ssrn.com/abstract=1690183. Accessed 24 June 2018 Stephen P, David J (1996) Stock Price Effect of High-Yield Debt Issues. Journal of Economics and Finance 20 (1): 133-146. Thomson Reuters (2017) Sukuk Perceptions & Forecast Study: Poised For Growth. https://ceif.iba.edu.pk/pdf/ThomsonReuters-SukukPerceptionsForecastStudy2017.pdf. Accessed 24 June 2018 Usmani M T (2007) Sukuk and their Contemporary Applications [(Translated from the original Arabic by Sheikh Yusuf Talal DeLorenzo)]. In AAOIFI Shariah Council meeting, Saudi Arabia. 17