Received: 29 March 2017 | Revised: 26 February 2018 | Accepted: 12 March 2018

DOI: 10.1111/beer.12185

ORIGINAL ARTICLE

The effects of Shariah board composition on Islamic equity

indices’ performance

M. Kabir Hassan1 | Federica Miglietta2 | Andrea Paltrinieri3 | Josanco Floreani3

1
Department of Economics and Finance, Based on a sample of 54 Islamic indices over the period 2007–2014, we investigate the effect of
University of New Orleans, New Orleans,
Shariah board members’ educational background on Islamic indices’ risk and return characteristics
Louisiana
2 via the screening criteria. Using a capital asset pricing model benchmark analysis, we assess the
Department of Economics, Management
and Business Law, University of Bari, Bari, sensitivity of Islamic indices to their conventional peers in terms of beta and derive a measure of
Italy return (Jensen’s alpha). First, we observe that the higher the number of members in common
3
Department of Economics and Statistics, among the boards, the higher the risk–return profile of Islamic indices. Second, commonalities
University of Udine, Udine, Italy among board members lead to standardization of the screening criteria and to similar Islamic indi-
ces’ performance. Third, we show that different betas across providers depend on the screening
Correspondence
M. Kabir Hassan, Department of Economics criteria, while the economic educational background of board members affects performance in
and Finance, University of New Orleans, terms of Jensen’s alpha. Our study aims at contributing to the governance literature related to
New Orleans, LA 70148.
board composition and its importance as a possible driver of performance. In addition, given the
Email: [email protected]
impressive growth that Islamic finance has experienced during the last decade, this topic is of great
interest to the asset management industry.

1 | INTRODUCTION responsibility to certify the Islamic index methodology. For this reason,
the Shariah board members’ decisions have a direct influence on the
The impact of board composition on screening criteria and, through composition of Islamic indices. In fact, basing their judgment on their
them, on index performance is a question of paramount importance in expertise and knowledge of Islamic Law, the Shariah board members
the Islamic finance asset management industry, since the activities of decide which exclusion screens and criteria have to be imposed on con-
Islamic financial institutions (IFIs) reached $1.89 trillion at the end of ventional indices in order to derive the correspondent compliant Islamic
2016 (IFSB Stability Report, 2017). In particular, Islamic funds, which one. For this reason, in our study, we wonder whether the subjective
exclude from the investment portfolio those stocks that are not compli- characteristics of the board members, such as their education (BA, MSs
ant with Shariah (Islamic Law), are experiencing an impressive growth. or PhD in law or economics), can influence the decisions related to the
Islamic funds track as benchmarks (actively or passively, as in the case exclusion screens, leading to a different index profile in term of risk
of Islamic Exchange Traded Funds) Islamic equity indices, whose risk and return. Should the different education background of the board’s
and return profile are analyzed in the present work. During the eco- members be able to influence the risk–return profile of the Islamic indi-
nomic downturn and the recent financial crises, we have witnessed ces, mutual fund managers, who use them as benchmarks in their
Islamic equity indices outperforming their conventional counterparts investment decisions, could have an additional important set of
due to different composition and construction mechanisms (IFSB Sta- information.
bility Report, 2016). These different results raise many questions Building on this proposition, at first, we focus on the composition
related not only to the construction mechanism, but also to the choices and education of the Shariah board members and, then, move to ana-
made by the Shariah Board. lyze if and how the board members’ educational background (speciali-
As a specific governance body in each IFI (Ullah, Jamali, & zation in economics or law) can affect the performance and risk of
Harwood, 2014), the Shariah Board must certify that a financial product Islamic equity indices, through the screens. Each Shariah board member
or activity complies with the precepts and requirements of Islamic law, has to be skilled both on Shariah and financial issues and, unfortunately,
thus exerting substantial influence and control over IFIs. In the asset this is not a common pairing of expertise. As a result, many scholars sit
management industry, each Islamic index provider has to hire eminent on multiple boards and this shortage of suitable scholars leads to a
Shariah scholars to sit in the Shariah Board entrusted with the Shariah board concentration issue (Bassens, Derudder, & Witlox, 2011).

248 | V
C 2018 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/beer Business Ethics: A Eur Rev. 2018;27:248–259.
HASSAN ET AL. | 249

Many studies have discussed the general topic of corporate gover- 2 | ISLAMIC FINANCE PRINCIPLES,
nance in IFIs in terms of differences between IFIs and conventional SHARIAH BOARDS, AND SCREENING
financial institutions (Grais & Pellegrini, 2006; Ullah et al., 2014). Unal CRITERIA
(2011) reviews the boards of several IFIs throughout the world and
finds evidence that the distribution of the boards is heavily skewed The core of the Islamic economic system is influenced by a body of
towards a few scholars. None of the abovementioned studies investi- immutable rules defined by Islamic law, Shariah (Iqbal, 1997). While not
gates how board composition and screening criteria affect the risk– differing from other contracts, financial transactions must be compliant
return profile of Islamic equity indices. with Shariah rules as well. Five principles distinguish Islamic financial
We find that commonalities among board members lead to stand- products: the prohibition of explicit interest rates (riba); the prohibition
ardization of screening criteria among providers and to similar perform- of transactions that are subject to excessive uncertainty (gharar); the
ance of Islamic equity indices. Furthermore, we find that Shariah prohibition of specific markets or products (i.e., pork, alcohol, weapons);
scholars’ educational background affects the risk profile of Islamic indi- profit and loss sharing between contractual parties (Minhat & Dzolkar-
ces via the screening criteria, which is the novelty of our work. naini, 2016); a direct link of each financial operation to the real economy
Our paper aims to contribute to the extant literature in several (Usmani, 2002). To ensure Shariah conformity in all transactions and
ways. activities, each financial institution sets up a Shariah board, which is a
First, our paper contributes to the Islamic business ethics litera- part of a larger Shariah department, together with a Shariah compliance
ture, focusing on behavior of boards as a driver of performance in unit and a Shariah audit/review unit (Ullah et al., 2014). The main pur-
the context of the Islamic perspective of ethical investments. Some pose of the board is to guarantee that every financial institution oper-
researchers draw attention to the role that financial instruments ates in conformity with Islamic principles and ethics (Rammal, 2006).
can play to fulfill the maqasid, or goals and broader objectives, of The board is formed of Shariah scholars, world-renowned experts in
Shariah (Laldin & Furqani, 2013). Moreover, a number of authors Islamic commercial jurisprudence (fiqh al-Muamalat) and finance, with
studied the role of Shariah boards in relation to the ultimate goals the fundamental role of setting business standards in all situations in
of Islamic investing. In particular, Shariah departments should have which a company needs to ensure that its activities abide by Shariah.
an active role in the implementation of Islamic financial values in Since a deep and detailed knowledge of both Shariah law and
IFIs (Ullah et al., 2014) and the decisions of Islamic boards are finance is difficult to obtain, just a few scholars who meet these
important for instilling Islamic ethics and values in the markets standards exist throughout the world. This shortage has led to
(Hasan, 2014). Among all Islamic compliant financial instruments many such scholars simultaneously being members of the boards of
that can be used, mutual funds, in particular, allow small investors a large number of financial institutions as well as several Islamic
to participate in capital markets and may help spreading prosperity indices. Bassens et al. (2011) provide evidence of a “transnational
and serve economic development in Islamic sense (Ibrahim, Elatrash, Shariah elite,” with just a few scholars sitting on several boards. This
& Farooq, 2014). Since our study focuses on those indices that interlocking Shariah board network spans the globe and is articu-
Islamic fund managers use as benchmarks, we indirectly link the lated in a few Islamic Financial Services hubs in the Middle East and
board characteristics to the fulfilling of the maqasid of Shariah. Far East. Table 1 lists the scholars and their education for the major
Second, we contribute to governance and board composition index providers.
issues by explicitly considering the impact that the characteristics of The qualitative data we used for our analysis (names of the scholars,
Islamic scholars (Ullah et al., 2014) can have on Islamic portfolios their CVs, and the screens applied to the starting universe) have been
through the Islamic screening criteria they impose. In fact, while there obtained from the official websites of the indices providers. Changes in
are few studies investigating the impact of Shariah boards on Islamic the board members have happened outside the period under analysis.
bank performance (Mollah, Hassan, Al Farooque, & Mobarek, 2017; Islamic indices are derived from conventional indices in accordance
Mollah & Zaman, 2015), ratings (Grassa, 2016) and Islamic bonds, or with screenings established by Shariah boards. Starting from a conven-
sukuk (Godlewski, Turk-Ariss, & Weill, 2014), there is a dearth of litera- tional index and applying all the Islamic criteria, the index provider is able
ture dealing with the impact of Shariah boards on Islamic indices’ risk to obtain the corresponding Islamic equity index that meets Islamic
and performance. investment requirements. The screens applied to the starting universe of
Third, we try to shed light on the determinants of systematic risk stocks are set by scholars to ensure that portfolios comply to Shariah.
and risk-adjusted performance. The findings are, in our view, of effec- The asset managers, then, implement Shariah board decisions in order to
tive interest to investors and asset managers, especially those using derive a final portfolio that is Shariah compliant. Thus, the scholars’ deci-
indices for exchange-traded fund management. sions influence both industry screens and financial ratios and, through
The paper is structured as follows. Section 2 examines the function them, shape the indices’ compositions in terms of stocks and sectors.
of the Shariah board and the process of setting the screening criteria. The industry screens relate to a company’s main activities and reve-
Section 3 reviews the literature and derives the research hypothesis. nue allocation. First, the company’s main activities must be halal. All banks
Section 4 describes our data set and the methodology used. Section 5 and insurance companies, whose activities are interest based, are excluded
presents the empirical findings, while Section 6 discusses the results from the portfolio, as are companies involved in alcohol, tobacco, and so
and develops their implications. Section 7 concludes. forth. After these industry filters are applied, all remaining stocks are
250 | HASSAN ET AL.

TA BL E 1 Scholars and education

Name Education Nationality

Panel A—Dow Jones Islamic Market Indices, Shariah Board

Dr Abdul Sattar Abu Ghuddah BA in Islamic Law, Damascus University Syria
PhD in Sharia, Islamic Law and Comparative law from Al-
Azhar University, Cairo, Egypt
Yusuf Talal DeLorenzo MSc in Finance, McGill USA
Nizam Yaquby University BA in Economics and Comparative Religion, McGill Bahrain
University
Dr Mohd Daud Bakar Degree in Shariah, University of Kuwait PhD, University of Malaysia
St. Andrews, Bachelor of Jurisprudence, University of
Malaya, Malaysia
Dr Mohd A. Elgari PhD in Economics, University of California Saudi Arabia
Professor of Islamic Economics
Statistics: Three scholars have a background in Economics (Mr DeLorenzo, Dr Elgari, Mr Yaquby); two scholars have a background in Law
(Dr Gnuddah and Dr Bakar)

Panel B—S&P Shariah Indices, Shariah Board

Dr Abdul Sattar Abu Ghuddah BA in Islamic Law, Damascus University Syria
PhD in Sharia, Islamic Law and Comparative law
from Al-Azhar University, Cairo, Egypt.
Dr Nazih Hammad Faculty of Shariah, Damascus University Canada
PhD in Islamic Law, University of Cairo
Dr Mohammad Amin Ali Qattan PhD in Islamic Banking, University of Birmingham, UK Syria
Dr Mohd A. Elgari PhD in Economics, University of California Saudi Arabia
Professor of Islamic Economics
Statistics: Two scholars have a background in Economics (Mr Qattan and Dr Elgari); two scholars have a background in Law (Dr Gnuddah and
Dr Hammad)

Panel C—FTSE Shariah Indices, Shariah Board from Yasaar Limited

Dr Mohammad Amin Ali Qattan PhD in Islamic Banking, University of Birmingham, UK Syria
Dr Mohamed Akram Laldin BA Honors, Islamic Jurisprudence and Legislation, University Malaysia
of Jordan
PhD in Principles of Islamic Jurisprudence (Usul al-Fiqh),
University of Edinburgh
Dr Mohd Daud Bakar Degree in Shariah, University of Kuwait Malaysia
PhD, University of St. Andrews, Bachelor of Jurisprudence,
University of Malaya, Malaysia
Mr Essam Ishaq Bachelor’s Degree in Political Science from McGill University Bahrain
Mr Muhammed Noorullah Shikder Master’s in Law, London University UAE
Statistics: One scholar has a background in Economics (Dr Qattan); four scholars have a background in Law (Dr Laldin, Dr Bakar, Mr Ishaq
and Mr Shikder)

Note. Data as of December 2014.

Source of data: The names of the scholars are taken from the official websites of the indices providers, last accessed on December 2014. FTSE
Shariah Global Equity Series scholars, whose names are taken from official documents available on www.ftse.com, are from the company Yasaar
Limited. The Scholars’ CVs were available on the Yaasar Limited company website, as of December 2014.The S&P Shariah and DJ Islamic indices
board scholars’ names, screenings, and CVs are from the S&P Shariah Indices Methodology publication available on the S&P official website
accessed in 2014.

analyzed based on financial ratio screens related to debt, interest-bearing performance. However, we go a step beyond, because we explicitly
securities, and receivables and cash with the clear aim of avoiding riba. consider how the subjective educational characteristics of the scholars
The prohibition of hoarding, however, is the basis for the condemnation of influence the indices’ riskiness through the imposition of those screens
excessive cash (Elgari, 2002). Therefore, to properly explore and study the that ensure compliance to Shariah.
behavior of Islamic indices, we cannot avoid taking into account the board There is a bulk of literature engaging with manager qualification,
characteristics and the screening criteria. education and experience, and its impact on portfolio performance and
risk. Different portfolio performance and risk metrics are used, where
3 | LITERATURE REVIEW AND RESEARCH the most common specifications regarding manager education refer to
QUESTION the value of holding an MBA or a CFA. Most studies test for the value
of holding an MBA or a CFA separately, providing mixed results. Obvi-
We contribute to the Islamic corporate governance field of literature ously, there is not only a matter of education but of quality of it as
by investigating the impact of Shariah boards on Islamic equity indices’ well. Several papers address the quality-of-education issue. Chevalier
performance and risk. Our paper falls within the broad strand of litera- and Ellison (1999) and Gottesman and Morey (2006) examine MBAs
ture linking observable characteristics of managers to fund with SAT and GMAT scores as proxies of the quality of education.
HASSAN ET AL. | 251

Our study resembles that of Chevalier and Ellison (1999), who et al. (2017) show that the Shariah board’s supervision allows banks to
examine whether mutual fund performance is related to characteristics undertake higher risk and achieve better performance than conventional
of fund managers in terms of educational background and other attrib- banks.
utes. They show that managers with different characteristics provide We try to bridge the gap by investigating the role of Shariah boards.
systematically different returns. Dincer, Russell, Allen, and Shawky While complying with Shariah principles, Shariah boards can draw on the
(2010) are the first to explicitly control for holding both an MBA and a unrestricted portion of investment opportunities through the imposition
CFA and assess the marginal performance of fund managers that is due of the screens, so that the final portfolio allocation (and its risk–return pro-
to the type of education. They find that MBA increases market betas. file) is influenced by the subjective characteristics of the board members.
€ tz (2017) provide new insights on the investment
Andreu and Pu If one thinks at Shariah board members as skilled professionals whose job
behavior of mutual fund managers holding a second degree. They com- is setting down and applying screening criteria, it seems reasonable to
pare the investment risk of managers holding both an MBA and a CFA speculate that their educational background might be systematically
to that of managers with only one degree, finding that managers hold- related to performance. Decisions about screens are not intentionally
ing the second degree choose significantly lower levels of risk. When meant to derive a portfolio with certain risk–return characteristics; how-
looking at Islamic indices’ performance, one should take into account ever, the literature shows that the cultural dimension, intended as educa-
that governance arrangements of Islamic indices differ since Shariah tion, gender, nationality, is a factor that determines financial risk taking
board is in charge of the selection policies in compliance with the attitude, shaping the asset allocation of portfolios (Black, Devereux, Lund-
Islamic law. This leads to prune out a subset of forbidden investments borg, & Majlesi, 2017; Riley & Chow, 1992; Wang, Rieger, & Hens, 2017;
from the entire set of available investment opportunities. Zhong & Xiao, 1995). In this sense, portfolios allocations are reliable indi-
Up to now, the majority of the studies on this field have dealt with cators of attitude toward risk (Shooley & Worden, 1996).
performance of Islamic products. According to the portfolio theory, if we Given that Islamic indices’ screenings are first based on religious
restrict the menu of activities to be chosen, we obtain a smaller set of beliefs, our choice is to refer Shariah board member characteristics to
investment opportunities that could affect overall portfolio performance the type of education, distinguishing members holding a degree in eco-
(Derigs & Marzban, 2008, 2009). Therefore, a first strand of literature has nomics from those holding a degree in jurisprudence.
compared the performance of Islamic and conventional equity mutual Provided that scholars are obliged to comply with Shariah princi-
funds (Ashraf, 2013; BinMahfouz & Hassan, 2013; Hayat & Kraeussl, ples when selecting investments, the different degree might impact on
2011; Hoepner, Rammal, & Rezec, 2011; Merdad & Hassan, 2016); how- their risk–return profile. Building on these considerations, we investi-
ever, no overall supremacy of any of these categories is noted. gate how the search for performance leads the portfolio moving along
A second group of studies has performed similar analyses on the the risk frontier. Our choice is to refer to risk as the systematic risk
performance of Islamic indices (Hassan, Antoniou, & Paudyal, 2005) in captured by market betas and return as the Jensen Alpha, according to
comparison with conventional ones (Adamsson, Bouslah, & Hoepner, prominent previous literature.
2014; Guyot, 2011; Walkshausl & Lobe, 2012). Results were mixed, in That impact of Shariah boards on the risk–return relation might
a similar way to what the literature has found in relation to socially flow through various channels, that is, the choice between different
responsible investment (SRI) (Revelli & Viviani, 2014). In particular, Ash- types of screenings (i.e., market-based versus accounting-based screen-
raf (2016) studies the performance of Islamic portfolios versus their ings) and the investment strategies pursued, which can be thought as
conventional counterparts taking into account different Shariah criteria driven by the educational background of scholars.
in both a single and a multi-equation framework. Ashraf’s findings We argue that the limited number of renowned scholars implies
show that different Shariah criteria do not affect index performance. that boards share common members. Therefore, as opposed to the
Nevertheless, none of the abovementioned studies analyze the general education-performance literature, two attributes concur to fea-
source of the screening criteria, namely, the Shariah board. Indeed, to the ture Shariah boards, that is, education and interconnections.
best of our knowledge there are no papers investigating the relationship Since scholars represent a rather narrow circle of advisors (Bassens
between Shariah scholar board members and Islamic index characteristics. et al., 2011), which allows for commonalities in the composition of the
Moreover, there are just a few recent papers investigating Islamic Shariah boards of the index providers (Unal, 2011), we do not expect a stat-
corporate governance (Najeeb & Ibrahim, 2014; Ullah et al., 2014; Wooi istically significant difference in performance among stock indices given
& Ali, 2017). The only studies attempting to evaluate the impact of Shar- the boards’ interconnections. We also expect that the more members in
iah scholar reputation on Islamic finance involve sukuk (Godlewski et al., common the greater the standardization of screenings and overall asset
2014) and banks (Grassa, 2016; Mollah & Zaman, 2015; Mollah et al., management strategies. This would lead to dampened return differences.
2017). In the case of sukuk, Godlewski et al. (2014) show that the reputa- Of course, standardization can be achieved at higher or, alternatively,
tion and proximity of Shariah scholars have a beneficial influence on stock lower parts of the risk–return frontier. If we think at scholars holding a
market reactions to sukuk issuance. Grassa (2016) documents that Islamic strict jurisprudence background as more conservative in securities selec-
bank credit ratings are negatively associated with the supervisory role of tion strategies, we can assume them as accepting lower returns and lower
the Shariah board and positively associated with the Shariah board’s inde- risk exposure than members with a degree in economics. Therefore, we
pendence and expertise. Mollah and Zaman (2015) find that Shariah measure the effect of education as referred to the share of common mem-
boards have a positive impact on Islamic bank performance, while Mollah bers having a degree in economics. Shared memberships are proxies for
252 | HASSAN ET AL.

4 | DATA AND METHODOLOGY

We collect data for 54 Islamic stock indices delivered by the major pro-
viders (the Dow Jones, the Financial Times Stock Exchange—FTSE, and
S&P) for the period 2007–2014. To assure homogeneity across providers,
we select a sample of Islamic indices that provide global and regional cov-
erage and their corresponding conventional peers. We could have selected
FIGURE 1 Research design: Concept and thematic outcome. Solid
lines indicate main effects; dotted lines indicate interactions SRI indices as the benchmarks (Forte & Miglietta, 2011). However, the
selection of conventional indices allows us to directly measure the addi-
the propensity to standardize screenings, while educational background tional yield that an investor could earn choosing the investment portfolio
can capture the propensity to beat or to follow conventional peers. that a specific benchmark represents but following Shariah principles.
All that considered, we can assume that both screening criteria and Moreover, the time series of SRI indices are quite limited, and we would
board education can moderate the risk–return relation. This is why we not be able to select the correct benchmark for every Islamic index.
introduce our two interaction terms (i.e. the interaction of screening Our sample permits a wide coverage by economic and geographi-
criteria with returns and the interaction of education with returns, cal areas while differences between providers just reflect the respec-
respectively) in our model. The previous discussion leads us to state tive buckets of products. We, therefore, believe that our sample is
the following hypothesis: representative of the entire population and provides an unbiased repre-
sentation of indices’ performance.
Hypothesis 1: The commonalities between Shariah boards,
We collect weekly annualized returns for each benchmark and the
together with screening criteria, do not produce different
corresponding Islamic index from the Bloomberg database. Following
performance and risk among Islamic stock indices.
financial theory, we perform a variety of risk-adjusted performance
Hypothesis 2: Education positively moderates the alpha– analysis such as the Sharpe ratio and Jensen’s alpha (Bauer, Derwall, &
beta relation in the sense that systematic risk is more Otten, 2007). We develop our research design in three steps as
performance-sensitive for indices managed by providers detailed in the following subsections.
having a higher fraction of common members with a
degree in economics.
4.1 | Relative performance analysis
Hypothesis 3: Screenings positively moderates the alpha–
We start with a risk-adjusted performance analysis. We calculate the
beta relation in the sense that systematic risk is more
Sharpe ratio for each Islamic and conventional index:
performance-sensitive for indices adopting market-based
screenings. ðRi 2Rf Þ
Sharpe ratio5 (1)
ri
Figure 1 describes our research design. Education (1), screening
criteria (2), and alpha (3) represent the main effects (solid lines). We, where Ri is the return on the index (Islamic & conventional peers); Rf is
then, introduce the moderating effects of both education and the risk-free rate proxied by the 3-month Treasury bill; and ri is the
screening criteria (dotted lines), that is, we test how education (1a) standard deviation of the index returns.
and screening criteria (2a) moderate the relation of alpha with We then perform a benchmark analysis. First, we calculate the excess
betas. return of each Islamic index over the conventional peer. This approach

FIGURE 2 Board interlocking and screening criteria

HASSAN ET AL. | 253

allows us to directly measure the additional yield that an investor could 4.3 | ANOVA analysis
earn by choosing the investment portfolio that a specific benchmark rep-
Moving from Ashraf (2016), we take a step further by explicitly investi-
resents, but following Shariah principles. Then, through a pair-by-pair com-
gating the impact of board composition and screening criteria on index
parison of index providers we analyze the differences in means.
sensitivity. We test an ANOVA model where beta is the dependent
A two-tailed Student’s t test with a 5% significance level is
variable. Given that screenings differentiate on a board-by-board basis
employed to check whether providers belong to different populations.
while remaining stable over time, our model is implemented on a pure
To this end, we first perform the Bartlett’s test for the null hypothesis
of equal variance. For those variables for which we reject the null cross-section basis:

hypothesis, we calculate the t-statistic using the Satterthwaite’s bi 5EDU1SCREE1ALPHA1ALPHA3EDU1ALPHA3SCREE (4)

approximation of the degrees of freedom.
Scholars and researchers (Durham, Kirkham, & Lindholm, 2012;
We test for differences in mean performance between the various
Hasan, 2014) have widely discussed the importance of the different
economic and geographical areas represented in our sample. Should we
schools of thought and their impact on the Shariah board members’
find that there are not statistically significant differences in means that
attitudes toward risk. The general consensus affirms the need for
would provide evidence that performance is not affected by sample
standardization through accepted rules. Harmonization efforts were
selection biases and differences between providers are due to different
made by the Accounting and Auditing Organization for Islamic Financial
screening criteria.
Institutions (AAOIFI) and the Fiqh Academy of the Organization of
Islamic Cooperation (OIC). Thus, we have assumed that the impact of
4.2 | Sensitivity analysis
the different schools of thought on screenings is gradually declining.
We, then, provide more insight into the risk–return relation by estimat- Instead, we focus on the skills and education of Shariah board members
ing the sensitivity of Islamic indices to their respective conventional as possible drivers of the screening criteria. We assume that differing
peers using a standard CAPM: education in terms of economic or law expertise can result in differing


Ri;t 2Rf 5a1bðRibt 2Rf Þ (2) approaches to setting criteria (Berger, Kick, & Schaeck, 2014). In fact,
criteria are derived from the Shariah, and differing education may lead
where Ri,t is the return of the ith Islamic index for time t, Ribt is the return of
to dissimilar views. Hasan (2014) discusses the importance of interdis-
the corresponding benchmark for time t, & Rf is the risk-free rate. We esti-
ciplinary competencies in Shariah boards, finding that different compe-
mate for each index the abovementioned regression, namely, the parame-
tencies allow boards to function more effectively.
ters a and b. We then jointly estimate the following null hypotheses:
We, therefore, construct three variables as drivers of sensitivity,
1. The null hypothesis H0 (ai 5 0), indicating that Jensen’s alphas are namely, education (EDU), screening criteria (SCREE), and alpha (ALPHA).
not statistically different from zero; that is, the Islamic indices do EDU and SCREE are attributes of the boards of the index pro-
not perform differently from their corresponding indices. viders. EDU is defined as the ratio of scholars with a degree in econom-
ics who each board shares with another board, divided by the total
2. The null hypothesis H0 (bi 5 1), indicating that the betas are not
number of board members (board size). FTSE has only one connection
statistically different from one, meaning that Islamic indices track
their benchmarks in terms of systematic risk. —with S&P. The common FTSE–S&P scholar has a degree in econom-
ics. Therefore, one out of four FTSE board members has an economic
3. The null hypothesis H0 (ai 5 0; bi 5 1).
background and is shared with another provider. Apart from the con-
Instead of separately estimating several independent equations, one nection with the FTSE, S&P has two connections with Dow Jones.
for each Islamic Index, we use a seemingly unrelated regression (SUR) Only one of the two S&P–Dow Jones common members has an eco-
model (Zellner, 1962) as in Ashraf (2016). Basically, an SUR model nomic background. Therefore, two out of a total of four S&P board
allows for a simultaneous estimation of the parameters in all equations. members have an economic background and sit on other boards. The
Such a model leads to greater efficiency in parameter estimates, thus Dow Jones has only the two connections with S&P. Therefore, one out
allowing us to tackle contemporaneous correlation effects and take of the six total Dow Jones board members has an economic back-
into account differences in relative performance of indices. We can ground and a seat on other boards. There are no connections between
rewrite Equation 2, therefore, as a system of equations: the FTSE and Dow Jones.
8
 Therefore, EDU describes commonalities in both board composi-
>
> R1;t 2Rf 5a1bðR1bt 2Rf Þ
>
>
>
> tion, referred to as shared membership (scholars sitting on more than
>
> ...
<
 one board) and the educational background of board members. A high
R 2R 5a1bðRibt 2Rf Þ (3)
> i;t f
> value of the ratio means that the specific board has widespread con-
>
>
>
> ...
>
> nections with the other boards, and those connections have a common
:

Rn;t 2Rf 5a1bðRnbt 2Rf Þ
educational background.
where Ri,t, for i 5 1 to n is the return of the ith Islamic index for time t, SCREE is a dummy variable that categorizes the providers accord-
Ribt is the return of the corresponding benchmark for time t, and Rf is ing to the screening criteria adopted, divided in liquidity ratios, debt
the risk-free rate as in Equation 2. ratios, and interest ratios. Table 2 summarizes the screening criteria.
254 | HASSAN ET AL.

TA BL E 2 Screening criteria

FTSE (%) S&P (%) Dow Jones (%)

(Accounts and receivables 1 cash and cash equivalents)/total assets 50

Accounts and receivables/market capitalization 49 33

Total interest income or expense/total revenues 5

Cash and cash equivalents/market capitalization 33 33

Total debt/total assets 33

Total debt/market capitalization 33 33

Non-permissible income other than interest/total revenues 5 5

Source of data: The screening data have been provided by the official websites of the indices providers, and these data are as of end of 2014. Each
index provider publishes, on the official website, a document regarding the methodology of the index.

TA BL E 3 Average Sharpe ratios

FTSE indices S&P indices Dow Jones indices

Benchmark Islamic indices Difference Benchmark Islamic indices Difference Benchmark Islamic indices Difference

2.78% 0.90% 1.97% 3.26% 3.38% 20.12% 2.16% 0.96% 1.20%

While screenings are stable over time, they differentiate one board 5 | EMPIRICAL FINDINGS
from another. This allows to map each provider into a specific style of
screen. This is why we adopt a dummy variable instead of measuring 5.1 | Relative performance analysis
our SCREE variable in terms of the quantitative thresholds, which is
For each provider Table 3 reports the average Sharpe ratios for both
coherent with our aim of investigating the sensitivity of risk–return
the conventional and Islamic indices, together with the corresponding
profiles to between-group differences in terms of screens. While FTSE
differences. FTSE Islamic indices have, on average, poorer performance,
calculates liquidity and leverage ratios based on accounting measures,
while the opposite holds for both S&P and Dow Jones indices, where
S&P and Dow Jones’ ratios are market based. Moreover, FTSE is the
Islamic series generally have, on average, higher Sharpe ratios and this
only provider adopting restrictions as regards interest income. There-
is particularly true for Dow Jones series (1.71% average Sharpe ratio
fore, the variable SCREE takes the value of 1 for FTSE screenings and
for Islamic indices against a 0.47% average ratio for the conventional
0 for S&P and Dow Jones screenings.
benchmarks). The detailed Sharpe ratios for each index are available
The variable ALPHA is the Jensen’s alpha. We also introduce the
upon request.
moderating effects of both EDU and SCREE on ALPHA. The goal is to
Table 4 reports, pair by pair, the t-test results for the Sharpe ratios’
gain a better understanding of the predictive power of both education
differences in means. While rejecting the null hypothesis of equal var-
and screenings on the sensitivity of Islamic indices via the impact on
iance for each pair in our sample (see the Bartlett statistic), our results
ALPHA.

TA BL E 4 Sharpe ratio differences: t-test results

FTSE S&P Bartlett’s statistic tTest*

Mean(a) SD Mean(b) SD Diff (a–b) Statistic p-Value Statistic p-Value

2.05 .11 .008 .07 2.06 26.13 .000 24.5417 .000

S&P Dow Jones Bartlett’s statistic t-Test*
Mean (a) SD Mean (b) SD Diff (a2b) Statistic p-Value Statistic p-Value

.008 .07 .01 .08 2.002 4.45 .035 2.2768 .782

FTSE Dow Jones Bartlett’s statistic t-Test*
Mean (a) SD Mean (b) SD Diff (a2b) Statistic p-Value Statistic p-Value

2.05 .11 .01 .08 2.06 8.89 .003 24.4327 .000

Source: Our elaborations on Bloomberg data.

*Two-tailed test, with a 5% significance level.
HASSAN ET AL. | 255

TA BL E 5 Weekly excess returns, by provider The two-tailed Student t test suggests that the only significant differ-
ence in mean excess returns is that between FTSE and Dow Jones.
FTSE
Finally, we test for mean differences in weekly extra-returns
Mean (%) Median (%) SD (%) P5 (%) P95 (%)
between economic and geographical regions. To this end, we perform a
2007 0.03 0.12 1.42 22.11 1.56
one-way ANOVA at a 95% confidence level. Table 7 summarizes the
2008 0.04 20.26 1.82 22.16 3.36 results. We report the summary statistics (mean, standard deviation, and
2009 20.25 20.22 0.63 21.22 0.51 frequencies) by economic region. The chi-square leads us to reject the
null hypothesis of equal variance, while the F-statistic (1.85) and the
2010 0.00 0.01 0.29 20.57 0.47
related p-value (.0852) tell us that there is not statistically significant dif-
2011 20.08 20.11 0.31 20.53 0.48
ference between our groups of means at a 95% confidence level.
2012 20.10 20.15 0.32 20.48 0.64 As a robustness check we controlled for crisis, that is, we per-
2013 0.02 0.10 0.26 20.41 0.37 formed an ANOVA analysis to test whether there are differences in

2014 20.15 20.16 0.23 20.43 0.22 performance between groups in different market scenarios (crisis
periods versus non-crisis periods). We find confirmation that differ-
DOW JONES
ences in means are not statistically significant. Results are available
2007 0.13 0.17 0.40 20.57 0.69 upon request.
2008 0.09 0.20 0.69 21.36 1.01 Overall, we provide evidence that performance is not affected by
2009 0.06 0.02 0.57 20.97 1.09 sample selection biases and differences between providers are due to
the different screening criteria.
2010 0.04 0.05 0.23 20.32 0.45

2011 0.05 0.05 0.34 20.48 0.52

5.2 | Sensitivity analysis
2012 20.05 20.06 0.26 20.55 0.47
Differences in volatility lead us to perform deeper benchmark analysis
2013 20.11 20.15 0.29 20.63 0.36
through a standard capital asset pricing model. This model is estimated
2014 20.03 20.03 0.27 20.46 0.44 based on an SUR methodology that allows us to simultaneously derive
S&P the Jensen’s alpha and the betas. Table 8 reports the alpha and beta’s

2007 20.26 20.37 0.93 21.74 1.51 average SUR estimations for the Islamic indices of the three providers
and, at the bottom, the hypothesis tests [H0 (ai 5 0); H0 (bi 5 1); and
2008 0.29 0.25 0.72 21.01 1.26
H0 (ai 5 0, bi 5 1)].
2009 0.03 20.02 0.43 20.65 0.96 Looking at the values in Table 9, we account for different
2010 20.02 0.01 0.24 20.44 0.43 degrees of responsiveness of the Islamic indices to their respective
2011 20.01 20.09 0.33 20.49 0.68 peers. The FTSE Islamic indices present, on average, the lowest
betas (0.64 average betas), while S&P and Dow Jones have, on aver-
2012 20.07 20.10 0.21 20.40 0.40
age, higher betas (0.92 and 0.95, respectively). Just a few S&P and
2013 0.00 0.03 0.24 20.39 0.35
Dow Jones Islamic indices are above 1 in market betas. Our results
2014 20.03 20.14 0.46 20.53 0.77 paint a picture where Islamic indices represent defensive invest-
ment strategies. Similar results emerge when it comes to investigat-
show statistically significant mean differences between FTSE and the ing Jensen’s alpha. Almost all providers share an average negative
other two providers according to the t-test statistics. alpha value suggesting that Islamic indices, with few exceptions, fail
The contrasting behaviors of both Dow Jones and S&P indices ver- to beat the reference market. Again, FTSE Islamic indices present
sus FTSE indices resemble the differences we accounted for in screen- the poorest performance according to Jensen’s alphas (–0.0046
ing criteria, suggesting their role in driving performance. against 20.0023 and 20.0010 for S&P and Dow Jones indices,
Table 5 summarizes the distribution of the weekly excess returns respectively). The details of the alpha and beta’s estimates for each
of Islamic indices. During the period of our investigation, FTSE indices index are available upon request.
recorded the poorest performance and greatest volatility. There are sig- When jointly testing our coefficients [H0 (ai 5 0); H0 (bi 5 1); and
nificant year-by-year differences between providers. In particular, there H0 (ai 5 0, bi 5 1)], Jensen’s alphas are statistically different from 0,
are abnormal returns in 2008, at least for Dow Jones and S&P indices, while betas are statistically different from 1, according to p-values. By
followed by a plunge in subsequent years. In terms of median values, rejecting the third null hypothesis, we state that the alphas and betas
Dow Jones indices show greater resilience to the crisis in 2010–2011, are jointly different from 0 and 1, respectively. Therefore, Islamic
with performance worsening over the following three years. equity indices produce statistically different performance and different
Table 6 reports, pair by pair, the t-test results. We reject the null risk–return payoffs than the corresponding benchmarks. Overall the
hypothesis of equal variance for each pair in our sample. The differences lower the sensitivity to the conventional index, the poorer the
in means show that S&P indices have higher average excess returns. performance.
256 | HASSAN ET AL.

TA BL E 6 Excess results: t-test results

FTSE S&P Bartlett’s statistic t Test*

Mean (a) (%) SD (%) Mean (b) (%) SD (%) Diff(a–b) (%) Statistic p-Value Statistic p-Value

20.08 0.94 20.01 0.52 20.076 136.18 .000 21.4529 .1467

S&P Dow Jones Bartlett’s statistic t-Test*
Mean (a) SD Mean (b) SD Diff (a2b) Statistic p-Value Statistic p-Value

20.01 0.52 0.02 0.41 20.03 21.50 .000 2.9199 0.3579

FTSE Dow Jones Bartlett’s statistic t-Test*
Mean (a) SD Mean (b) SD Diff (a2b) Statistic p-Value Statistic p-Value

20.08 0.52 0.02 0.41 20.11 250.91 .000 22.117 .0347

Source: Our elaborations on Bloomberg data.

*Two-tailed test, with a 5% significance level.

5.3 | ANOVA analysis benchmarks. Overall, our model fits the data set well in terms of the F-
statistic and its 1% significance level (see Table 10).
To analyze the variance of the dependent variable for all our groups
Looking at the main effects, SCREE and ALPHA are statistically
combined and to detect which of the specific groups differs, we per-
significant at 1% level. The greatest impact on betas is to be traced
form a one-way ANOVA at the 5% significance level. The results are
to the effect of screenings. Since SCREE identifies different groups
reported in Table 9.
of providers, we infer that differing selection criteria will contribute
Overall, we find a statistically significant difference in the mean
to differentiating indices in terms of systematic risk exposure. Hav-
beta between the three different groups of the dependent variable,
ing similar screening criteria, S&P and Dow Jones indices do not
according to the F-statistic. The comparison of means shows that FTSE
show significant differences in betas; by contrast, differentiation in
is the group that differs from the others while there is not statistically
screenings explains diverging patterns of FTSE indices’ systematic
significant difference between Dow Jones and S&P indices, according
risk exposure.
to p-values.
The alpha is significant (at the 1% level) as well. In sum, we can say
We, finally, estimate the ANOVA model described by Equation 4
that the attempt to beat the reference market comes at the expense of
to investigate the drivers of Islamic indices’ sensitivity to their
increasing the sensitivity of the Islamic index to its benchmark, and that
this effect is significant. The variable capturing education is only mod-
TA BL E 7 Summary of extra-performance between economic erately significant (10% level). As for interactions, ALPHA*SCREE is not
regions and analysis of variance statistically significant, while the moderating effect of EDU on ALPHA
is significant at the 1% level.
Economic areas Mean SD Frequency
The interaction term tells us that systematic risk is more
Americas 21.59907 20.450346 1,668
performance-sensitive for indices managed by providers having a
Asia-Pacific 20.07506 56.248794 6,835 higher fraction of common members with a degree in economics. The
Developed 0.139634 14.537466 1,668 positive coefficient (0.0233) of the interaction term tells us that when
EDU increases the effect of ALPHA increases. Overall, our results sup-
Emerging 817.8962 28,216.491 3,336
port H1 and H2, while H3 is rejected.
Europe 1.424361 37.64187 2,502

UAE, Arab, Egypt 23,333.63 962,592.69 3,420

T AB LE 8 Average alpha and beta values
World 1.288598 41.247783 3,118
FTSE S&P DJ
Analysis of variance a b a b a b
Between groups
Average 2.0046 .64 2.0023 .92 2.001 .95
F-statistic (p-value)* 1.85 (.0852)
H0 (ai 5 0) 345.2 (.000) 224.2 (.000) 184.4 (.000)
v2 (p-value) 300,000 (.0000)
H0 (bi 5 1) 398.68 (.000) 888.7 (.000) 1121 (.000)
Notes: The table reports the summary statistics (mean, standard deviation H0 (ai 5 0, bi 5 1) 753.56 (.000) 1,115.9 (.000) 1,310.5 (.000)
ad frequencies) for extra-performance across different areas. The bottom
of the table reports the one-way ANOVA results, that is, the between- Notes: The table reports the estimates of the average alpha and beta for
group F-statistic and related p-value and the chi-square statistic and each Islamic index. At the bottom, we report the estimates for the tests
related p-value. of H0 (ai 5 0), H0 (bi 5 1), H0 (ai 5 0, bi 5 1). Coefficient tests: 5% confi-
*5% Confidence level. dence level. p-values are in brackets.
HASSAN ET AL. | 257

TA BL E 9 ANOVA by group, with dependent variable b conventional peers. Looking at Jensen’s alpha, the third finding is that
our sample of Islamic indices presents, on average, lower performance
Sum of the Mean
squares df squares F than their peers. We can therefore reasonably argue that Islamic indi-

Between-groups 0.7241 2 0.3620 12.36 (.000)

ces take a more conservative approach, which tends to dampen
returns. We, then, support our hypothesis that homogeneity in board
Within-groups 1.5523 53 0.0292
attributes tends to reduce differences in performance. Our results thus
Total 2.2765 55 0.0413 differ from those of other studies, in particular that of Ashraf (2016),
Comparison of means Bringing up educational factors, however, allows us to rethink the
FTSE vs S&P 20.2753 (.000) relations between risk and returns in view of behavioral patterns, cap-
FTSE vs DJ 20.2598 (.000)
tured by board composition and their selection criteria. Here, the
DJ vs S&P 20.0155 (1)
fourth finding and the major contribution of our paper is that both edu-
Notes: p-Values in brackets. We employ a 5% significance level. cation and screening criteria play a role in driving the systematic risk
exposure of Islamic indices. Looking at the ANOVA model, we observe
6 | DISCUSSION OF THE RESULTS AND that the closer the connections between the boards, coupled with the
ETHICAL IMPLICATIONS
economic background of scholars, the higher the risk–return profiles of
these Islamic indices. Consequently, these indices resemble their con-
As a first finding, we rank Islamic indices according to their risk–return
ventional peers more than the other indices do. These findings lead us
trade-off. S&P and, to a far greater extent, Dow Jones indices show
to wonder whether Islamic investments should be regarded as more
higher risk-adjusted returns than their benchmarks, while the opposite
conservative than conventional portfolios due to their inherent con-
holds for FTSE indices.
straints or whether the difference in performance is an effect of sub-
Moreover, the investigation of Islamic indices’ excess returns
jective attitudes and beliefs captured by the board members’
reveals the superior performance of Dow Jones indices, which proved
education. To answer this question, we need to consider the role of
to be more resilient during the crisis and post-crisis periods. However,
education and screening criteria. Screening criteria unequivocally iden-
Islamic indices tend to perform worse than their benchmarks in the
tify groups since there is a clear pattern in criteria among providers.
post-crisis period, but our results do not show statistically different
Our results prove that market-based screenings make a positive
excess returns between our three groups.
contribution to systematic risk. A market-based debt ratio implies that
As a second finding, our sensitivity analysis reveals that Islamic
growing market capitalization makes feasible a greater level of debt.
indices are, on average, less exposed to systematic risk compared to
Therefore, expanding financial risk, together with operational risk,
makes Islamic indices much more sensitive to their conventional peers.
TA BL E 1 0 ANOVA, with dependent variable b
Our analysis suggests that screening criteria alone cannot solve the
Partial sum of Mean problem without endogenizing the effect of education. Screenings, while
the squares df squares F
having a main effect on the betas, do not significantly moderate the rela-
Model 1.0116 5 0.2023 18.8*** tion between alpha and systematic risk. It is, by contrast, EDU that posi-
EDU 0.0361 1 0.0361 3.36* tively moderates the attempt to pursue higher returns. The major
contribution of scholars in driving betas develops via the impact on alpha
SCREE 0.3581 1 0.3581 33.28***
rather than the main effect. When brought together into a comprehen-
ALPHA 0.1096 1 0.1096 10.19***
sive picture, our results have significant implications for investors. Con-
ALPHA*EDU 0.0233 1 0.0233 2.17*** nections in terms of common members sharing a common background
ALPHA*SCREE 0.0974 1 0.0974 9.05 lead to standardization in the screening criteria and contribute to
smoothing differences in performance, albeit with higher systematic risk
Residual 0.5274 49 0.0107
exposure. Therefore, what actually differentiates index providers is board
Total 1.5390 54 0. 0285
composition, captured by education and commonalities. Essentially,
Number Obs. 55 increasing commonalities in the economic background of scholars (rather
R2
0.6573 than a jurisprudential background) leads to the standardization of screen-
ing, driving selection strategies toward a more aggressive approach.
Adjusted R2 0.6223
As a practical matter, asset managers and investors looking at risk–
Notes: The table reports the results of the ANOVA model where mar-
return profiles can differentiate their asset allocation on Islamic prod-
ket beta is the outcome variable. EDU is the education variable meas-
ured as the fraction of common board members having a degree in ucts primarily based on screening criteria in the sense that different
economics. SCREE is the screening variable, which identifies the index screenings affect performance and risk-taking behavior. The impact of
provider and the specific style of screenings. It takes value 1 for FTSE education is indirect. It acts in a way of moving indices along the risk–
indices, constructed on accounting-based screenings and 0 otherwise.
ALPHA is the Jensen Alpha. ALPHA*EDU and ALPHA*SCREE are the return frontier reinforcing the effect of ALPHAS on systematic risk,
interactions. F-statistics: p-values. ***1% significance level, **5% signif- that is, the attempt to beat the market leads to a greater impact on sys-
icance level, *10% significance level. tematic risks for those indices having a greater fraction of members in
258 | HASSAN ET AL.

common with a degree in economics. Therefore, board composition, €tz, A. (2017). Choosing two business degrees versus
Andreu, L., & Pu
together with screening criteria, represents a good case for portfolio choosing one: What does it tell about mutual fund managers’ invest-
ment behavior? Journal of Business Research, 75, 138–146.
diversification.
Ashraf, D. (2013). Performance evaluation of Islamic mutual funds rela-
tive to conventional funds: Empirical evidence from Saudi Arabia.
7 | CONCLUSIONS International Journal of Islamic and Middle Eastern Finance and Man-
agement, 6(2), 105–121.
In this paper, basing our analysis on three major providers (FTSE, S&P, Ashraf, D. (2016). Does Shariah screening cause abnormal returns?
and Dow Jones) we carry out a thorough investigation of how Shariah Empirical evidence from Islamic equity indices. Journal of Business
Ethics, 134, 209–228. https://doi.org/10.1007/s10551-014-2422-2
board composition, in terms of education, affects the risk–return profile
Bassens, D., Derudder, B., & Witlox, F. (2011). Setting Shariah standards:
of Islamic equity indices via the screening criteria that the board mem-
On the role, power and spatialities of interlocking Shariah boards in
bers impose on the unrestricted portfolios in order to derive Shariah
Islamic financial services. Geoforum, 42(1), 94–103.
compliant ones. There are slightly relative differences in the Sharpe
Bauer, R., Derwall, J., & Otten, R. (2007). The ethical mutual fund per-
ratios of Islamic indices among providers, with FTSE indices underper- formance debate: New evidence from Canada. Journal of Business
forming, on average, Dow Jones and S&P indices. On a between-group Ethics, 70(2), 111–124.
comparison, FTSE Islamic indices present the poorest performance. Berger, A. N., Kick, T., & Schaeck, K. (2014). Executive board composi-
Our benchmark analysis suggests that Islamic indices represent tion and bank risk taking. Journal of Corporate Finance, 28, 48–65.

defensive investment strategies compared to the reference market. BinMahfouz, S., & Hassan, M. K. (2013). Sustainable and socially respon-
sible investing: Does Islamic investing make a difference?. Humano-
Screening criteria play a significant role in driving Islamic indices’ sensi-
mics, 29(3), 164–186.
tivity to their peers and account for major differences between pro-
Black, S., Devereux, P., Lundborg, P., & Majlesi, K. (2017). On the origins of
viders. The attempt to beat the reference market comes at the expense risk-taking in financial markets. The Journal of Finance, LXXII, 2229–2278.
of increasing the sensitivity of the Islamic index to its peer. We observe
Chevalier, J., & Ellison, G. (1999). Are some mutual fund managers better
that the more members shared among the boards, the higher the risk than others? Cross-sectional patterns in behavior and performance.
taking of these Islamic indices. We find, moreover, that connections in The Journal of Finance, 54(3), 875–899.
terms of common members sharing a common background lead to Derigs, U., & Marzban, S. (2008). Review and analysis of current Shariah
standardization in the screening criteria and contribute to smoothing compliant equity screening practices. International Journal of Islamic
and Middle Eastern and Management, 1(4), 285–303.
differences in performance, albeit with higher systematic risk exposure.
Derigs, U., & Marzban, S. (2009). New strategies and a new paradigm for
Our paper contributes to the existing literature in several ways.
Shariah-compliant portfolio optimization. Journal of Banking and
First, we extend the literature on Islamic Finance and Islamic business Finance, 33(6), 1166–1176.
ethics, since this is the first study assessing the impact of Shariah board Dincer, O. C., Russell, B., Allen, G., & Shawky, H. A. (2010). Are you smarter
composition and scholars’ background on the performance of Islamic than a CFA’er? (1458219). Retrieved from http://www.ssrn.com
indices through the impact of the different education characteristics on Durham, W. C., Kirkham, D., & Lindholm, T. (2012). Islam and political-
the risk profile of the final portfolio. Second, we contribute to the gov- cultural Europe. London: Routledge.
ernance literature related to board composition and its importance as a Elgari, M. (2002). Islamic equity investment. In S. Archer & R. A. Karim
possible driver of performance. Third, we try to shed light on the deter- (Eds.), Islamic finance, innovation and growth (pp. 151–160). London:
Euromoney Books and AAOIFI.
minants of alpha and beta measures. The findings are, in our view, of
Forte, G., & Miglietta, F. (2011). A comparison of socially responsible and
effective interest to investors and asset managers, especially those
Islamic equity investments. Journal of Money, Investments and Banking,
using indices for exchange-traded fund management. 21, 116–132.
However, our paper has limitations as well. First of all, it focuses Godlewski, C. J., Turk-Ariss, R., & Weill, L. (2014). Do the type of sukuk
on board composition and education as the drivers of systematic risk and choice of Shariah scholar matter? (IMF Working Paper 14/147,
exposure, without assessing other macroeconomic factors. Moreover, International Monetary Fund).
the limited and fixed number of Shariah board members could affect Gottesman, A. A., & Morey, R. M. (2006). Manager education and mutual
our analysis. Finally, we chose to use conventional indices rather than fund performance. Journal of Empirical Finance, 13(2), 145–182.

SRI indices as benchmarks. Future research could delve into this aspect Grais, W., & Pellegrini, M. (2006). Corporate governance and Shariah compli-
ance in institutions offering Islamic financial services (World Bank Policy
of the analysis, given the fact that SRI indices often have an ethical
Research Working Paper No. 4054). Washington, DC: World Bank.
committee with consulting power.
Grassa, R. (2016). Corporate governance and credit rating in Islamic banks:
Does Shariah governance matter? Journal of Management and Gover-
ORCI D nance, 20, 875–906. https://doi.org/10.1007/s10997-015-9322-4

M. Kabir Hassan http://orcid.org/0000-0001-6274-3545 Guyot, A. (2011). Efficiency and dynamics of Islamic investment: Evi-
dence of geopolitical effects on Dow Jones Islamic market indexes.
Emerging Markets Finance & Trade, 47, 24–45.
RE FE RE NCE S Hasan, Z. (2014). In search of the perceptions of the Shariah scholars on
Adamsson, H., Bouslah, K., & Hoepner, A. G. F. (2014). An Islamic equity pre- Shariah governance system. International Journal of Islamic and Middle
mium puzzle (2390347). Retrieved from http://www.ssrn.com Eastern Finance and Management, 7(1), 22–36.
HASSAN ET AL. | 259

Hassan, A., Antoniou, A., & Paudyal, D. K. (2005). Impact of ethical Usmani, M. T. (2002). An introduction to Islamic finance. The Hague, the
screening on investment performance: The case of Dow Jones Netherlands: Kluwer Law International.
Islamic Index. Islamic Economic Studies, 12(2), 67–97. Walkshausl, C., & Lobe, S. (2012). Islamic investing. Review of Financial
Hayat, R., & Kraeussl, R. (2011). Risk and return characteristics of Islamic Economics, 21(2), 53–62.
equity funds. Emerging Market Review, 12, 189–203. Wang, M., Rieger, M. O., & Hens, T. (2017). The impact of culture on
Hoepner, A. G., Rammal, H. G., & Rezec, M. (2011). Islamic mutual funds’ loss aversion. Journal of Behavioral Decision Making, 30(2),
financial performance and international investment style: Evidence 270–281.
from 20 countries. European Journal of Finance, 17(9–10), 828–850.
Wooi, H. C., & Ali, R. (2017). Does a Muslim CEO matter in Shariah-
Ibrahim, A. A., Elatrash, R. J., & Farooq, M. O. (2014). Hoarding versus compliant companies? Evidence from Malaysia. Pacific-Basin Finance
circulation of wealth from the perspective of maqasid al-Shari’ah. Journal, 42, 126–141. https://doi.org/10.1016/j.pacfin.2016.08.010
International Journal of Islamic and Middle Eastern Finance and Man-
Zellner, A. (1962). An efficient method of estimating seemingly unrelated
agement, 7(1), 6–21.
regressions and tests for aggregation bias. Journal of the American
IFSB. (2016). Islamic Financial Services Industry Stability Report. Malaysia: Statistical Association, 57(298), 348–368.
Islamic Financial Services Board.
Zhong, L. X., & Xiao, J. J. (1995). Determinant of family bond and stock-
IFSB. (2017). Islamic Financial Services Industry Stability Report. Malaysia: holdings. Financial Counseling and Planning, 6, 1–8.
Islamic Financial Services Board.
Iqbal, Z. (1997). Islamic financial system. Finance & Development, 34(2),
1–4. AUT HOR B IOGRAP HIE S
Laldin, M. A., & Furqani, H. (2013). Developing Islamic finance in the M. KABIR HASSAN is Hibernia professor of economic and finance, Bank
framework of maqasid al-Shari’ah. International Journal of Islamic and one professor in business and professor of finance at the University of
Middle Eastern Finance and Management, 6(4), 278–289.
New Orleans, USA. He holds a PhD in Finance and his research topics
Merdad, H., & Hassan, M. K. (2016). Does faith matter in mutual funds
include Islamic Banking and Finance, Emerging Markets and Financial
investing? Evidence from Saudi Arabia. Journal of Emerging Markets
Finance and Trade, 52(4), 938–960. Development, Investments and International Finance.

Minhat, M., & Dzolkarnaini, N. (2016). Islamic corporate financing: Does FEDERICA MIGLIETTA is Associate Professor of Banking and Corporate
it promote profit and loss sharing? Business Ethics: A European Review,
Finance at Universit
a degli Studi di Bari Aldo Moro, Italy. She holds a
25(4), 482–497.
PhD in Business Administration and Management from Bocconi Uni-
Mollah, S., Hassan, M. K., Al Farooque, O., & Mobarek, A. (2017). The
governance, risk-taking, and performance of Islamic banks. Journal of versity and an undergraduate degree in Economics of Financial Markets
Financial Services Research, 51, 195–219. https://doi.org/10.1007/ and Institutions from the same university. Her research interests
s10693-016-0245-2 include Islamic Finance, socially responsible investments and perform-
Mollah, S., & Zaman, M. (2015). Shariah supervision, corporate gover- ance of local banks.
nance and performance: Conventional vs. Islamic banks. Journal of
Banking & Finance, 58, 418–435. ANDREA PALTRINIERI is Assistant Professor of Banking and Finance at the
Najeeb, S. F., & Ibrahim, S. H. M. (2014). Professionalizing the role of University of Udine, Italy. He holds a PhD in Business Administration
Shariah auditors: How Malaysia can generate economic benefits. from the University of Verona. Research topics include Islamic Finance,
Pacific-Basin Finance Journal, 28, 91–109.
asset management and institutional investors, with a particular focus
Rammal, H. G. (2006). The importance of Shariah supervision in IFIS. Cor-
on sovereign wealth funds, commodity markets and exchange traded
porate Ownership and Control, 3(3), 204–208.
products.
Revelli, C., & Viviani, J. L. (2014). Financial performance of socially
responsible investing (SRI): What have we learned? A meta-analysis. JOSANCO FLOREANI is Associate Professor of Corporate Finance at the
Business Ethics: A European Review, 24(2), 158–185. University of Udine, Italy. He holds a P.h.D. in Business Management
Riley, W. B., Jr., & Chow, K. V. (1992). Asset allocation and individual from the University of Udine. His main research interests cover Islamic
risk aversion. Financial Analysts Journal, 48(6), 32–37.
Finance, SMEs financing, and bank’s loan portfolio quality.
Shooley, D., & Worden, D. (1996). Risk aversion measures: Comparing
attitude and asset allocation. Financial Services Review, 5(2), 87–99.
Ullah, S., Jamali, D., & Harwood, I. A. (2014). Socially responsible invest- How to cite this article: Hassan MK, Miglietta F, Paltrinieri A,
ment: Insights from Shari’a departments in Islamic financial institu-
Floreani J. The effects of Shariah board composition on Islamic
tions. Business Ethics: A European Review, 23(2), 218–233.
equity indices’ performance. Business Ethics: A Eur Rev.
Unal, M. (2011). The small world of Islamic finance: Shariah scholars and
governance—A network analytic perspective, v. 6.0. Funds@Work, The 2018;27:248–259. https://doi.org/10.1111/beer.12185
investment industry’s Strategy Consultant, Kuala Lumpur.