Alhadi 2016

ACCOUN-00763; No of Pages 21
The International Journal of Accounting xxx (2016) xxxxxx
Contents lists available at ScienceDirect
The International Journal of Accounting
Ruling Family Political Connections and Risk Reporting: Evidence from the GCC
Ahmed Al-Hadi b,c,, Grantley Taylor c, Khamis Hamed Al-Yahyaee a
a
Sultan Qaboos University, College of Economics and Political Science, Department of Economics and Finance, Oman, Muscat
b
Sultan Qaboos University, College of Economics and Political Science, Department of Accounting, Oman, Muscat
c
School of Accounting, Curtin University Perth, Western Australia, Australia
a r t i c l e i n f o a b s t r a c t
Available online xxxx This study examines whether the presence of ruling family members on boards of directors inu-
ences the extent and the quality of risk reporting. Based on a sample of publicly listed nancial
Keywords: rms of the Gulf Cooperation Council countries between 2007 and 2011, our regression results
Political connection show that ruling family board members reduce the quality and extent of risk disclosures. Firms
Ruling family director with ruling family board members also disclose signicantly less during periods of nancial dis-
Risk disclosure tress and when they are subject to higher levels of risk. We nd that risk reporting is negatively
Gulf Cooperation Council associated with the existence of a ruling family director acting as the board chairperson, negative-
ly associated with increasing proportions of ruling family directors on the board, and negatively
JEL classication: associated with increasing numbers of board members who are connected to ruling family direc-
M40 tors. Our results suggest that politically connected directors seize private benets at the expense
M41 of their rms' shareholders. Our regression results hold after a series of robustness checks that
G15 and D72
control for endogeneity and for alternative measures of ruling family membership.
2016 University of Illinois. All rights reserved.
1. Introduction
This study examines whether the presence of ruling family members1 on boards of directors inuence the extent and the qual-
ity of risk reporting by Gulf Cooperation Council (GCC)-listed nancial rms from 2007 to 2011. The GCC2 provides an ideal set-
ting in which to examine the relationship between the political connections of board members and the reporting transparency of
rms. Many of the GCC-listed rms have at least one ruling family member on the board of directors (Halawi & Davidson, 2008).
For instance, in the United Arab Emirates (UAE) in 2008, 56 directors from 101 listed rms were ruling family members (Halawi
& Davidson, 2008). Furthermore, in Qatar, 78 (24%) of all listed companies had ruling family members on their boards. Similarly in
Kuwait and Oman, there were 45 (21%) and 31 (26%) listed rms that had ruling family members on their boards, respectively.
Leuz and Oberholzer-Gee (2006) and Chaney, Faccio, and Parsley (2011) call for further research into the effects that these polit-
ical connections that GCC rms have on reporting transparency.
This study is motivated by the growing interest in the relationship between political connections and corporate transparency
(Chaney et al., 2011; Leuz & Oberholzer-Gee, 2006). Polsiri and Jiraporn (2012) provide evidence that rms with ruling family
connections may obtain economic benets from their family ties. These researchers cite a case where rms belonging to the
Corresponding author.
E-mail addresses: alhadisam@yahoo.com (A. Al-Hadi), grantley.taylor@cbs.curtin.edu.au (G. Taylor), yahyai@squ.edu.om (K.H. Al-Yahyaee).
1
The term ruling family member is equivalent to royal family member.
2
The Gulf Cooperation Council (GCC) is an economic union of six countries in the Arab Gulf region, established in 1981. The GCC consists of Oman, Bahrain, Kuwait,
Qatar, the Kingdom of Saudi Arabia (KSA), and the United Arab Emirates (UAE).
http://dx.doi.org/10.1016/j.intacc.2016.10.004
0020-7063/ 2016 University of Illinois. All rights reserved.
Please cite this article as: Al-Hadi, A., et al., Ruling Family Political Connections and Risk Reporting: Evidence from the GCC, The In-
ternational Journal of Accounting (2016), http://dx.doi.org/10.1016/j.intacc.2016.10.004
2 A. Al-Hadi et al. / The International Journal of Accounting xxx (2016) xxxxxx
King of Thailand were less likely to fail during the 1997 Asian Financial Crisis due to their ruling family connections. Few studies,
however, have investigated the relationship between political connections and corporate risk transparency in emerging economies
such as the GCC. Survey results (see, e.g., IFC & Hawkamah, 2008) suggest that adequate disclosures are of paramount importance
in the GCC as a means to protect shareholders' rights, particularly because monarchial control is strongly entrenched. Effective dis-
closure of rms' risks can minimize rent-seeking activities by ruling family members and other controlling interests (Jaggi, Leung,
& Gul, 2009). Risk disclosures are considered to be an integral component of good corporate governance structure, and such dis-
closures are particularly important to market and bank regulators, international organizations, and development institutions in the
GCC, as these enterprises seek to consolidate rms' risk management frameworks and practices or to strengthen internal controls
(Hertog, 2012; IFC & Hawkamah, 2008). The provision of adequate risk disclosures is an important factor in the regulation of pub-
licly listed GCC rms.
Appointment of ruling family members on boards arise through their seniority among a monarchical group, being a founding
member of a rm, their large equity and controlling interest in a rm, and appointment by a nomination committee (Hertog,
2012; IFC & Hawkamah, 2008). Generally, the board structures and legal provisions regarding the appointment of directors and
their responsibilities follow those that are required under company law in a Western-based system, but there can be some
signicant local variations. The nancial reporting frameworks of GCC countries are based on both company laws and royal
decrees (Al-Shammari, Brown, & Tarca, 2008). Some rms are established through presidential decrees or other special statutes
that give them particular benets, including appointments of politically connected directors such as ruling family members.
Thus, ruling family appointments to the boards are likely to be driven by a mixture of Western-based legal principles and political
or family connections (Hertog, 2012).
The power of ruling family members over the GCC's economic and political regimes may undermine recent efforts by the
region's regulators to enforce and improve corporate governance and to enable nancial reporting transparency. First, to the ex-
tent that board memberships for ruling family members serve to protect these members' personal interests and the interests of
related companies, the management of rms characterized by ruling family directorships may fail to disclose risk-related informa-
tion, including disclosures mandated by accounting standards or regulatory bodies. Second, the rent-seeking behavior and the
business interests of politically connected rms are protected by the monarchy (Mazaheri, 2013), which makes it less likely
that management will resist the wishes of powerful ruling family directors (Jaggi et al., 2009). A setting in which there is no
real separation between the monitoring and controlling functions is likely to have a negative effect on rm reporting and trans-
parency. Third, ruling family board directors may not wish to attract negative market attention by disclosing information relating
to their rms' risk exposures (Certo, 2003). Furthermore, politically connected rms could have lower disclosure levels, even in
the face of higher risk levels or in times of distress, due to the protection that ruling family connections may offer these rms.
Ruling family directors may consider that their connections with key stakeholders, such as other groups in the royal family or im-
portant government ofcials, will protect their rms in difcult times. Such ruling family directors are able to transfer wealth to
themselves at the expense of other contracting parties and to make their risk-related disclosures in an opportunistic manner
(Emanuel, Wong, & Wong, 2003).
However, several recent developments in the GCC region are likely to stimulate an increased demand for transparency and
disclosure, particularly with respect to risk reporting (Al-Hadi, Hasan, & Habib, 2016; IFC & Hawkamah, 2008). Corporate gover-
nance codes3 and regulations are now well established in all GCC countries, with some of these countries holding their rms ac-
countable for non-compliance with business regulations (Al-Shammari et al., 2008). Some GCC countries have established
corporate governance task forces to monitor the adherence of rms to codes of conduct and good governance (Al-Hadi et al.,
2016; IFC & Hawkamah, 2008). Recently, regulation of GCC rms has made signicant progress toward establishing more inde-
pendent boards of directors. In all of the GCC countries, the corporate governance codes require that an audit committee must
review a rm's risk management systems and policies, including their disclosure practices. Adoption of the International Account-
ing Standards (IAS) or the International Financial Reporting Standards (IFRS) is mandatory for all listed nancial companies
throughout the GCC (Al-Shammari et al., 2008; IFC & Hawkamah, 2008).4 All GCC central banks have adopted Basel II, including
Pillar III: Disclosures. Finally, although board risk management is voluntary in practice, about 39% of all the GCC nancial rms
have adopted risk management policies that require board oversight and disclosure of risk (Al-Hadi, Taylor, & Hossain, 2015).
The GCC region has also seen a marked increase in foreign direct investment (Mina, 2007). As a whole, the GCC has experi-
enced unprecedented growth rates, and many of its companies trade with offshore partners or have subsidiaries incorporated
in countries outside the GCC (Lagoarde-Segot & Lucey, 2007). This internationalization of GCC-listed rms makes them subject
to greater scrutiny from stakeholders, regulators, and international institutional investors, who have recently been demanding
greater transparency and accountability from those rms (Abu-Nassar & Rutherford, 1996).
Although ruling family membership on boards is relatively high in the GCC region, the ruling family ownership of nancial
rms is relatively low compared to such ownership of non-nancial rms (Al-Hassan, Oulidi, & Khamis, 2010). Market risk
3
For instance, the Capital Market Authority (CMA) of Oman mandated a code of corporate governance for listed rms starting from 2002. In 2005, the Qatar Financial
Market Authority (QFMA) introduced a code of corporate governance based on the OECD's comply or explain requirements, which took effect in 2009 (Sharar, 2011).
Similarly, in 2011, the Central Bank of Bahrain issued a code of corporate governance for listed rms based on nine governance principles that rms are required to ad-
dress in their annual reports. In 2006, the CMA issued a code of corporate governance in the KSA, effective from 2009. In 2007, the Securities and Commodities Authority
(SCA) released a code of corporate governance, closely followed by its Ministerial Resolution No. 518-2009. This code mandates standards of corporate governance for
all listed rms in the UAE from 2009 forward. The code consists of 16 articles that are based on OECD corporate governance principles.
4
Compliance with the IFRS is not required for non-nancial rms in some GCC countries, such as the KSA.
A. Al-Hadi et al. / The International Journal of Accounting xxx (2016) xxxxxx 3
exposures are considered to be particularly important for GCC nancial rms (Al-Hadi et al., 2016; Beattie, McInnes, & Fearnley,
2004), given the recent emphasis of regulatory bodies on strengthening risk management and risk reporting systems (Al-
Shammari et al., 2008). Firms belonging to the nancial sector are generally more prone to issues related to risk disclosure, as
these rms are subject to greater regulatory constraints (e.g., Central Bank regulations, Basel, and IFRS). Prior research on the -
nancial industry and on market risk reporting shows that nancial rms disclose more risk-related information than other indus-
tries (e.g., Hirtle, 2007; Nier & Baumann, 2006; Perignon & Smith, 2010a). Financial rms represent a signicant part of total stock
market value in the GCC (Hammoudeh & Choi, 2006).5 Additionally, the shareholdings of GCC nancial rms are often dominated
by inuential royal families or government leaders (Al-Hadi et al., 2016; Al-Shammari et al., 2008). Halawi & Davidson (2008)
nds that royal family members are strongly represented on the boards of directors of GCC nancial rms.
Given this dominance of royal family ownership and the strong representation of royal family members on the boards of -
nancial rms, it is important to focus on these rms in our investigation. Especially in the case of nancial rms, there is a real
need to determine whether the presence of ruling family members on boards of directors inuences the extent and quality of
risk reporting. Overall, the risk disclosure patterns of GCC nancial rms are affected by both the existence of ruling family
board members and by the increasing requirements for greater transparency, such as the recent developments in governance
codes, the adoption of IAS/IFRS, the growing importance of regulatory bodies, and the internationalization of GCC-based rms.
Using 667 rm-year observations from between 2007 and 2011, we provide evidence that being a politically connected GCC
nancial rm (as identied by the presence of ruling family members on their boards of directors) is negatively and signicantly
associated with the extent and the quality of market risk disclosure. We further nd that rms with ruling family board members
disclose signicantly less risk reporting during periods of nancial distress and when they are subject to higher levels of risk.
Moreover, rms chaired by a ruling family member and boards with members who have a direct family connection to the ruling
family are signicantly and negatively associated with the extent and quality of market risk disclosures.
The empirical models used in this study control for rm-specic factors, including the reporting quality and the corporate gov-
ernance structures of rms (e.g., Chaney et al., 2011; Guedhami, Pittman, & Saffar, 2014). Furthermore, we include rm-specic
governance regimes and country-specic investor protection levels in our empirical tests (e.g., Boubakri, Guedhami, Mishra, &
Saffar, 2012b). We also include family (other than ruling family) and government ownership variables in certain models to control
for the effects of ownership structures (e.g., Boubakri et al., 2012b; Ding, Jia, Wu, & Zhang, 2014; Jaggi et al., 2009; Wan-Hussin,
2009). Consistent results are obtained after using alternative specications of rm reporting. We also use instrumental variable
(IV) techniques and ordinary least squares (OLS) estimations to mitigate endogeneity concerns. The results from the IV estima-
tions are very similar to those of the baseline OLS estimations, indicating that endogeneity is not likely to account for the observed
associations.
We contribute to the literature on political connection and risk disclosure in several important ways. First, to the best of our
knowledge, the literature has not thus far examined the association between political connections and the market risk disclosures
of nancial rms in the context of the GCC. We extend earlier studies (e.g., Polsiri & Jiraporn, 2012) by providing evidence that
the presence of ruling family members on a board of directors reduces the extent and quality of market risk disclosures. Anecdotal
evidence has suggested that the presence of ruling family members on boards is important or even critical to achieving business
outcomes. Parallel studies (e.g., Ali, Chen, & Radhakrishnan, 2007; Jaggi et al., 2009; Wan-Hussin, 2009) on the family and govern-
ment connections of board members also suggest that political links are likely to be important determinants of business decisions
by rms in the GCC, including decisions related to the nature and degree of risk reporting (e.g., Polsiri & Jiraporn, 2012).6 This
study explicitly contributes to the literature by quantifying the association between the existence of ruling family board members
and the quality of market risk reporting by nancial rms. The literature to date (e.g., Chaney et al., 2011; Guedhami et al., 2014;
Leuz & Oberholzer-Gee, 2006) provides mixed evidence regarding the direction of association between politically connected rms
and corporate transparency. We use multiple measures of political connection in this study7 and provide consistent evidence of a
negative and signicant association between the presence and the number of ruling family members in boards (or serving as
board chairs) and the extent or quality of market risk disclosures.
Second, despite the pivotal role of disclosure in enhancing a rm's value and its relations with shareholders (Healy & Palepu,
2001), comparatively little research has addressed the issue of risk disclosure from the perspective of less democratic, family
relations-based economies. Hence, in this study, we draw our sample from the six monarchical GCC countries, which are char-
acterized by low levels of democracy, the dominance of ruling family members in businesses, high levels of state ownership, a
lack of transparency, and weaker disclosure levels (Al-Hadi et al., 2016; Al-Yahyaee, Pham, & Walter, 2011; Mazaheri, 2013). The
political and economic setting of the GCC region provides us with an important opportunity to investigate the role of political
connections in corporate transparency. The ndings of this study can be extrapolated to other relations-based developing
countries.
This study proceeds as follows. Section 2 provides the theory and develops the hypothesis. Section 3 explains the research de-
sign, data sources, and sample selection. Section 4 presents the empirical results. Section 5 contains the results of our robustness
checks and additional tests, and Section 6 offers the study's conclusions.
5
For example, N82% of the listed rms on the UAE's stock market are nancial rms.
6
Polsiri and Jiraporn (2012) nd that nancial rms related to Thailand's Crown Property Bureau were less likely to fail during the Asian Financial Crisis of 1997.
7
Previous studies investigating the effects of political connections on disclosure tend to rely on a single measure of political connection (Chaney et al., 2011). We are
able to rely on several measures of ruling family members in this study, as the annual reports of nancial rms disclose the full family names and the tribal afliations of
directors.
2. Theory and hypothesis development
2.1. Market risk disclosures
Market risk is dened as the risk of uctuations in fair value, cash ows, or earnings due to changes in market prices (Jorion,
2002). Such risk may be initiated by changes in interest rates, exchange rates, equity prices, or commodity prices (Al-Hadi et al.,
2016). Various U.S.-based studies (e.g., Jorion, 2002; Rajgopal, 1999) and professional surveys (e.g., Chartered Financial Analyst
(CFA) Institute, 2013) argue that market risk disclosure allows investors to make informed decisions regarding the risks associated
with on- and off-balance sheet items and the effect that changes in interest rates, exchange rates, or fair values may have on -
nancial outcomes. Al-Hadi et al. (2015) nd that detailed disclosure of market risk, given in a way that shows its particular com-
ponents, leads to a reduction in information asymmetry and lowers the cost of equity capital for GCC rms that make such
disclosures.
Healy and Palepu (2001) document that increases in disclosure quality enhance shareholder wealth through reductions in in-
formation asymmetry and agency problems (Jensen & Meckling, 1976; Myers & Majluf, 1984). For instance, mandatory disclosure
can play an important role in protecting stakeholder interests when insiders have the opportunity to use company information for
their own objectives (Admati & Peiderer, 2000; Boot & Thakor, 2001). Disclosure of risk information allows investors to assess a
rm's future performance and evaluate uncertainties concerning future cash ows (Linsley & Shrives, 2006). Such disclosures also
form an important part of risk management and effective governance (Jorion, 2002). Furthermore, risk disclosures may reduce a
rm's idiosyncratic risks (Jorgensen & Kirschenheiter, 2003).
Rajgopal (1999) posits that there is often a deciency in mandatory risk reporting irrespective of the regulatory requirements
or pronouncements that mandate such disclosure. Under IFRS, rms have some level of discretion in their communications of -
nancial risk information (Dobler, 2008; Einhorn, 2005). Although disclosure of aggregate accounting numbers (including risk in-
formation) is mandatory, the decomposition of and disclosure of these amounts in detail is voluntary (Einhorn, 2005). Risk
disclosure increases diversity in nancial reporting, and it improves the precision of risk information made available to stake-
holders. Providing such information can be less costly to the preparers of nancial information than seeking to withhold it
(Dobler, 2008). Einhorn (2005) posits that discretion in rms' disclosure practices is inuenced by mandatory disclosure require-
ments and that such requirements have an important bearing on a rm's voluntary disclosure practices.8
2.2. Political connections of ruling family members on the board
Prior research demonstrates that rms can benet from the use of political connections in a number of ways. These ways in-
clude preventing competitors from entering a market (Bunkanwanicha & Wiwattanakantang, 2009), lowering corporate tax rates
(Faccio, 2006), gaining assistance for survival during periods of nancial distress (Faccio, Masulis, & McConnell, 2006), providing
favorable lending terms from creditors (Claessens, Feijen, & Laeven, 2008), enhancing nancial performance (Boubakri, Cosset, &
Saffar, 2012a), improving the performance of family rms (Xu, Yuan, Jiang, & Chan, 2015), or lowering the costs of capital
(Boubakri et al., 2012b). The ruling family members on the board may not necessarily behave in an opportunistic way or seek
to benet themselves at the expense of all other stakeholders. Khanna and Palepu (2000) and Khanna and Rivkin (2001) provide
evidence that political connections may actually enhance rm performance and reduce nancial risk. Furthermore, Hertog (2012)
asserts that ruling family members on boards may serve as a buffer against government interventions that could impede, for in-
stance, the rm's economic or international expansion. However, Chaney et al. (2011) argue that political connections commonly
discourage accounting transparency, as politically connected members on the board provide protection to management in the
event that full disclosure is not provided to stakeholders. Thus, politically connected rms may be characterized as having
lower accounting transparency compared to non-connected rms.
A rm is deemed to be politically connected if it has a shareholder with political links who controls at least 10% of the rm's
stocks (a large shareholder); if the rm is state owned; if a founder of the rm is involved in management (Ding et al., 2014); if at
least one of the executives or members of the board is also a member of the parliament, the state government (Chaney et al.,
2011; Faccio, 2006), or a municipal council (Amore & Bennedsen, 2013), is a high-ranking member of the military (Fan, Wong,
& Zhang, 2007), or is a member of royalty or a ruling family (Polsiri & Jiraporn, 2012). Firm directors may be said to have con-
nections with ruling family members if a leading politician or the ruler himself is a large shareholder or a director or if a
major politician is their close relative (e.g., brother, son, or daughter).
The GCC monarchical regimes control a region in which the political institutions have been systematically built to favor specic
classes (Amenta, 2000). In the GCC region, the relationships between government members, old trading families, and royal fam-
ilies overlap and are interdependent (Kshetri & Ajami, 2008). These relationships can be strengthened by shared mutual interests
between merchant and ruling families (Al-Rumaihi, 1986). Ruling family members tend to lack accountability, expect absolute
obedience, are generally unwilling to be questioned (Sidani & Al Ariss, 2013), and they control many of the state-owned rms
(Hertog, 2012). In this situation, non-ruling entrepreneurs may capitalize on their connections with ruling family elites through
forming relationships, business partnerships, and marriages (Mazaheri, 2013).
8
Firms may reduce the extent of voluntary disclosures when there is more discretion in the mandatory disclosure practices. Conversely, rms may increase the ex-
tent of their voluntary disclosure when the accounting standards provide greater latitude and exibility in their disclosure requirements (Einhorn, 2005).
The GCC region is also characterized by political regimes that are inuenced by favoritism toward personal and family connec-
tions (Atiyyah, 1992). Our empirical analysis is based on the fundamental premise that the political orientation of company direc-
tors is inuenced by their degree of political power, which in turn determines the degree of their nancial transparency. The
political power and inuence of ruling family members who sit on the boards of GCC rms may exacerbate agency-related prob-
lems between the board and the management (Type I) or between the minority and majority (controlling and non-controlling)
shareholders (Type II).9 In general, the control of formal power in relation to wealth is strategically determined between ruling
family members and trading families in the GCC region (Crystal, 1995).10 Ruling family members are also key members of
state institutions, so they are commonly able to control top positions and centralize decision- making in those institutions
(Hertog, 2012).
Ali et al. (2007) nd that ruling family directors have more power to obtain and hold a rm's information, such as corporate
governance information. These directors often have more capacity to act in ways that benet themselves rather than the share-
holders. According to Hertog (2012), the board members of some GCC rms appear to have been chosen based on their seniority
among ruling elites or their ruling family connections. Hertog (2012) cites numerous examples in which directors are appointed
due to their political connections, without having the technical or nancial knowledge that is normally expected of someone
working in the industry. In some GCC rms, the boards may comprise members who cannot be identied as outside or indepen-
dent directors, and these members may combine numerous other senior functions with their directorships. In particular, Hertog
(2012) notes that most GCC rms have boards that are recruited based on seniority from a fairly small circle of elites. These
boards are staffed with directors who often have little spare time. Although they often have a wide range of general experience,
their specialized expertise is often quite limited. For example, Joshi and Wakil (2004) suggest that in some GCC countries such as
Bahrain, it is difcult to nd a director who meets the criteria of being fully independent. Thus, it is not surprising that rms may
lack conventional mechanisms of accountability and governance, particularly in relation to the disclosure of critical risk
information.
On balance, we see an increasing regime of accountability and governance in the GCC with adoption of the IFRS and various
codes of good corporate governance. Risk disclosure is an important component in these codes of governance, and such disclo-
sures are widely seen as particularly important for expansion into offshore markets and for reducing the region's reliance on par-
ticular commodities such as oil. However, the ruling family members on some boards may not be motivated to ensure that
adequate risk disclosures are provided in annual reports. Like in the situations that Khanna and Palepu (2000) and Khanna and
Rivkin (2001) describe regarding business groups in the Indian context, the ruling party members in the GCC nations have the
capacity to offer benets for rms (i.e., improved performance, reduced intervention, reduced risk) or to limit their costs
(i.e., reduced reliance on risk reporting).
2.3. Political connections and market risk disclosures
Although previous research supports the notion that political connections can add value for all shareholders, such connections
can also create Type II agency conicts between controlling and non-controlling shareholders (Schipper, 1989). Politically con-
nected directors or major family shareholders often enjoy substantial control due to their concentrated equity holdings in
rms, such that their voting rights exceed their cash ow rights.11 Ruling family members are able to exert signicant inuence
on the board owing to their ownership interests as major shareholders and to their political connections. Thus, ruling family di-
rectors are likely to have the power to obtain private benets at the expense of minority shareholders. The expropriation of ben-
ets from other shareholders can be achieved via a range of Type II agency mechanisms, such as partitioning minority
shareholders to make them ineffective (Gilson & Gordon, 2003), engaging in related-party transactions (Anderson & Reeb,
2003) practicing managerial entrenchment (Shleifer & Vishny, 1997), or promoting government risk-seeking (Xu, Xu, & Yuan,
2013).
Politically connected rms may be reluctant to improve their accounting transparency. Prior research provides evidence that
rms with ruling family connections have greater access to low-interest loans from state-owned banks (see, e.g., Claessens
et al., 2008). Moreover, Leuz and Oberholzer-Gee (2006) provide evidence from a potentially analogous situation in which polit-
ically connected Indonesian rms are able to source capital under non-arm's length arrangements to conceal transactions that
benet controlling insiders and their political backers, rather than seeking other sources of capital that would mandate higher
levels of transparency and disclosure. Piotroski, Wong, and Zhang (2010) explain that transparency will limit the ability of pol-
iticians and managers to consume their private benets of control by exposing poor governance (p. 3).
Stulz (2005) nds that the threat of public exposure has a moderating effect on whether politicians and insiders collude to
extract private benets. As evidence that smaller investors are marginalized in these situations, Fan et al. (2007) nd that Chinese
rms with politically connected CEOs seldom appoint directors representing minority shareholders. Therefore, a large proportion
of directors are afliated with the largest shareholders or with government agencies. Politically connected rms in Canada also
9
Type I agency costs arise from the separation of ownership and control, and Type II agency costs arise from conicts of interest between controlling (majority) and
non-controlling (minority) shareholders.
10
For example, the Halawi & Davidson (2008). reports that in Kuwait, the Khara family holds 5% of the Kuwait stock market, and the ruling family (the Sabah family)
has 4.6% ( Halawi & Davidson, 2008)).
11
Expropriation theory suggests that a controlling director is entrenched due to his or her signicant voting rights and frequent involvement in management and,
hence, tends to abuse this power through extracting corporate resources for the director's own interests (Shleifer & Vishny, 1997).
tend to have concentrated ownership (e.g., Morck, Stangeland, & Yeung, 1998), and such concentrations of control reinforce the
concerns of outside investors that the insiders in these rms may consume private benets to the detriment of others.
In the political environment of the GCC, ruling family members have control over both major shareholders and the govern-
ment (Crystal, 1995). Ruling family members can also take high-level positions in state institutions, through which they control
key decisions (Hertog, 2012). Politically connected rms are thus likely to have much lower market risk due to the protection
they may get from royal patronage. Given the lower risks faced by rms with ruling family members on their boards, these
rms may correspondingly reduce their risk-related disclosure. In support of this assertion, Mazaheri (2013) highlights that
non-ruling entrepreneurs gain a major advantage in the market if they are able to get a royal patron on board with a project
(p. 315). Non-ruling entrepreneurs are also able to capitalize on ruling family protection and branding through forming personal,
familial, and business relationships. Dobler (2008) argues that uncertainties regarding information endowment, issues of veri-
ability of information and threats of economic disadvantages all tend to promote reduced risk reporting. The study's results sug-
gest that the domination by ruling family members over operational and nancial strategies or arrangements not only facilitates
their own welfare, but it can also facilitate the businesses of non-ruling family entrepreneurs, which are therefore collectively
more likely to suppress incentives for disclosing risk information. All of these factors are likely to provide incentives for ruling
family directors to avoid disclosure of risk information, which is likely to exacerbate agency-related problems.
The vast majority of non-controlling shareholders in the GCC have weak protection against the major or ruling family share-
holders, who may take many of the rm's benets from the non-dominant shareholders (Crystal, 1995; Hertog, 2012). Agency
problems are thus likely to inuence the risk disclosure practices of GCC rms. Such agency problems could lead to a more
opaque reporting environment in which risk information is selectively omitted or sufciently aggregated12 so that specic risk ex-
posures are not completely disclosed. Ruling family directors are thus likely to promote their interests at the expense of minority
shareholders and thereby solidify their substantive control (Andres & Vallelado, 2008; Barclay & Holderness, 1989; Jaggi et al.,
2009). This discussion therefore suggests that rms with ruling family members on their boards, or with board representatives
who are connected through family ties to a royal family, are less likely to disclose market risk information. This suggestion
leads to our directional hypothesis:
H1. All else being equal, a politically connected rm, as evidenced by having at least one ruling family member on the board of
directors, is negatively associated with the extent and quality of market risk disclosures.
3. Research design
3.1. Sample and data
We drew our sample from the population of nancial rms listed in Bahrain, the Kingdom of Saudi Arabia (KSA), Kuwait,
Oman, Qatar, and the UAE between 2007 and 2011. We hand-collected data related to market risk disclosures, the ruling family
members on the boards, and other governance characteristics from annual reports. All other data were obtained from Capital IQ.
Denitions of the variables used in the study are provided in Table 1.
The sample initially consisted of 1375 rm-year observations (Table 2, Panel A). We then excluded joint-listed rms (15 rm-
years), rms without annual reports (670 rm-years), and rms with missing data (23 rm-years). This yielded a nal sample of
677 rm-year observations. Panel B of Table 2 shows that the UAE represented the highest number of rm-year observations
(150), followed by Kuwait with 137 and Oman with 124 rm-year observations.
3.2. Dependent variables
We used the IFRS 7: Financial Instrument Disclosure statement to construct our market risk disclosure index. Pursuant to the
IFRS 7 disclosure requirements, discretion in format, type, and number of quantitative disclosures is permissible. As such, we
followed Al-Hadi et al. (2015, 2016) to construct the disclosure index based on the types (qualitative or quantitative) and
reporting requirements (mandatory or voluntary) of market risk disclosures. The resulting qualitative section of our disclosure
index consisted of 14 mandatory and voluntary disclosure items, which were related to the Value-at-Risk (VaR) and Sensitivity
Analysis (Sen) formats.13 The quantitative section of the disclosure index comprised the ve subcategories of VaR disclosure:
VaR characteristics (four items), summary VaR statistics (ve items), inter-temporal comparisons (one item), back testing (two
items), and daily VaR gures (two items). The index also had three subcategories of Sen disclosures: Sen characteristics (four
items), summary Sen statistics (ve items), and intertemporal comparisons (four points). Like Prignon and Smith (2010b), we
12
Aggregated risk disclosures cause individual investors to react differently than disclosures in which the risk information is disaggregated (Hodder, Koonce, &
McAnally, 2001).
13
Sensitivity Analysis (hereafter Sen) is a disclosure format that measures the potential loss in future income, fair value, and cash ows related to market risk expo-
sures that may arise from a hypothetical change or reasonably possible change over a short timeframe (Hodder, Koonce, & McAnally, 2001). The effects of the potential
loss should involve declines in prots, equity, or cash ows. Value-at-Risk (VaR) is a disclosure format that measures the highest potential loss in future cash ows,
earnings or fair value over a selected period, with a likelihood of occurrence or probability in most cases at the 5% level (or the 95% condence level) (Hodder et al.,
2001).
Table 1
Variable denitions and measurement.
Variables Denition and measurement
Dependent variable: disclosure proxies

MRDFACTOR Mandatory market risk disclosures obtained from component factor analysis of both qualitative and quantitative Value-At-Risk (VaR)
and Sensitivity Analysis (Sen) market risk disclosures after Miihkinen (2012).
EMRD Number of qualitative and quantitative VaR and Sen market risk disclosure items divided by the maximum number of items possible.
This variable is a measure of the extent of mandatory market risk disclosures.
QMRD Sum of the qualitative and quantitative market risk disclosures scaled by the number of VaR and Sen risk exposures. This variable is a
measure of the quality of market risk disclosures.
Political connection proxies

RoyalF_Dir A rm with a ruling family member on the board of directors is scored as 1, otherwise 0.
RoyalNumDir The number of ruling family directors on the board of directors as a proportion of the number of board members.
RoyalChairman A rm with the board chair that is a ruling family member is scored as 1, otherwise 0.
RoyalRank A rm with a ruling family director related to royalty is scored as 1, otherwise 0.
Control variables
CG_Index Firm's corporate governance index comprising seven items: a. CEO and chairman duality, b. BOD busyness, c. chairman is a member
of at least one board monitoring committee, d. majority of independent directors on the BOD, e. existence of an audit committee, f.
existence of a remuneration or nomination committee, and g. existence of a risk committee. Each of these seven items are scored as 1
if present, otherwise 0.
LNMVAL Natural log of a rm's end-of-year market capitalization.
Leverage Total short and long-term debt divided by total assets.
ROE Return on equity.
BVMV Growth measure of book value divided by market value of stock at market end-of-year.
AuditQ A dummy variable that takes on the value of 1 if the rm is audited by one of the Big Four auditors, 0 otherwise.
Family_Dir A dummy variable that takes on the value of 1 if the rm is owned by a family company and has family representatives on the board
of directors, 0 otherwise.
GovDir A dummy variable that takes on the value of 1 if the rm is owned by the government (or one of its agencies) and has at least one
government representative on the board of directors, 0 otherwise.
Ownership_Con Percentage government and institutional shareholdings.
Earning_Volatility The standard deviation of return of assets (ROA) for the three previous years.
Beta A rm's beta for at least 12 months.
Ruling-Female = Dummy variable if rm board includes women from a ruling family on the board,
Royal_C = The country average of ruling family director.
CGovPoli_Stabi = The country's political stability index according to the World Bank.
Altman Z-Score Altman's (1968) prediction of rm bankruptcy as a nancial distress measure, calculated as:
(1.2 * Woking Capital / Total Assets. + 1.4 * Retained Earnings / Total Assets + 3.3 * Earnings Before Interest and Taxes /
Total Assets + 0.6 * Market Value of Equity / Book Value of Total Liabilities + 0.99 * Sales / Total Assets).
allocated equal weight to each disclosure item.14 Appendix A lists the items that comprise the risk disclosure index. We closely
followed previous disclosure studies in selecting the index items (e.g., Al-Hadi et al., 2015; Prignon & Smith, 2010b).15
3.2.1. Component factor analysis

Following Miihkinen (2012), we undertook composite factor analysis to measure the quality of mandatory market risk disclo-
sures by GCC nancial rms and to examine the underlying patterns for a number of variables. Hence, we condensed the data into
a smaller set of factors (Hair, Anderson, Tatham, & Black, 1995). The full results are not reported for the sake of brevity. First, we
constructed the qualitative items for market risk disclosures (e.g., policy, objectives, limitations) for both VaR and Sen disclosure
formats. Second, we constructed the quantitative items for VaR and Sen based on three potential risks to which GCC nancial
rms are exposed. Finally, based on the score for each format, we conducted principal component and factor analyses and obtain-
ed an eigenvalue of 1.52 (50.71% proportion). Hence, our eigenvalue represented our observable components:
MRDFACTOR Composite Highest Score of Eigenvalue 1
14
Perignon and Smith (2010b) nd that the results from using unweighted disclosure indices are very similar to those obtained using weighted disclosure indices.
Furthermore, Hooks and Staden (2011) provide similar arguments in support of using unweighted indices.
15
In constructing the mandatory (voluntary) part of the MRD index, we relied on IFRS 7 (or both IFRS 7 and Basel II, Pillar III) for several reasons. First, Pillar III is vol-
untary in nature, and it aims to facilitate market discipline by requiring institutions to disclose their capital and risk exposures, risk assessment processes, and the capital
adequacy of banks. The IFRS 7, however, is mandatory in nature, and it applies to all listed rms (Al-Shammari et al., 2008). As our sample comprised banks plus invest-
ment, nancial, and insurance rms, we constructed the mandatory part of our index based on IFRS 7, as it covers all rms. Moreover, the voluntary part of the risk
disclosure index was developed based on IFRS 7 (for all rms) and on Basel II, Pillar III (for banks only). Second, although the central banks' circulars and requirements
in all of the GCC countries strongly advise the adoption of Basel II, Pillar III, the Pillar III disclosures are (unlike those of the IFRS 7) not required to be audited by an ex-
ternal auditor unless otherwise required by the IFRS, the Capital Market Authority, or the central bank standards. Third, prior studies (e.g., Li, 2010) document that IFRS
improves the quality, reliability, transparency, and comparability of nancial information. We considered all of these aspects when constructing our index to improve its
accuracy, validity, and consistency.
Table 2
Sample specications.
Panel A: Sample selection
Number of observations available for nancial rms in S & P Capital IQ for the GCC 1375
Less:
Joint-listed rms (15)
Firms with unavailable annual report for disclosure items (670)
Firms with missing values in control variables (23)
Total rm-year observations 677
Panel B: Sample distribution
Bank Financial Insurance Investment
Bahrain 55 0 15 15 85
KSA 45 0 9 34 88
Kuwait 50 72 10 5 137
Oman 29 65 10 20 124
Qatar 39 10 25 9 83
UAE 80 20 50 0 150
Total 298 167 119 83 667
3.2.2. Extent of market risk disclosures

Based on studies by Francis, Khurana, and Pereira (2005) and Miihkinen (2012), and following Al-Hadi et al. (2015, 2016), we
constructed an additional market risk disclosure (EMRD) index, which was calculated based on the number of qualitative and
quantitative items gathered from both market risk disclosure formats (VaR and Sen), divided by the maximum expected
score:
nX ni14
i14
qualitative IndexVaR and SenX ij X
3133
Quantitative indexVaR and SenX ij
EMRD 2
t1
ne; j t1
ne; j
3.2.3. Quality of market risk disclosures

Based on the approach used by Miihkinen (2012),16 we constructed an index based on the quality of market risk information
disclosed (QMRD), which was the sum of the score of qualitative and quantitative market risk disclosures scaled by the number of
risk exposures reported in the annual report (e.g., interest rate risk, currency exchange risk, and equity risk from both the VaR and
Sen indices):
nX ni14
i14 X3 133
QUANTIndexVaR and SenX ij
QMRD Qualitative index VaR and Sen X ij ; 3
t1 t1
No: of market risk exposuree; j
where Xij equals 1 if the ith item is disclosed by the jth rm, nej is the expected score of each disclosed individual market risk
exposure, nj1 is the total score (nj1 14) from the qualitative disclosure of the jth rm, and nj2 is the total score from the quan-
titative disclosure of the jth rm. QMRD is the sum of the total qualitative disclosures (scaled by 14 items) and the quantitative
market risk disclosures scaled by the number of market risk exposures. The total score from the quantitative risk disclosure was
calculated as the sum of the scores from the VaR and Sen formats, scaled by the number of market risk exposures.
3.3. Independent variables
Ruling family membership (RoyalF_Dir) was set to equal 1 if a rm had at least one ruling family director on the board and 0
otherwise. Table 3, Panel A shows the mean and the total numbers of ruling family directors in our sample. As in the Halawi &
Davidson (2008) the Qatari RoyalF_Dir mean was the highest, with approximately 92.3% (79) directors, and the Saudi RoyalF_Dir
number was the lowest, at only 4%.17
16
Miihkinen (2012) uses a risk quality measure based on coverage that measures a rm's risk concentration across various risk factors.
17
This result was consistent with that of Al-Hassan et al. (2010). We found that ruling family directors had a comparatively low representation on the boards of -
nancial rms in the KSA. We tested whether this representation held for non-nancial rms. As an additional check, we also collected ruling family memberships in
boards from listed industrial rms (non-nancial rms) from 2007 to 2011 for all GCC countries, including the KSA. We found that the number of ruling family directors
on the boards of KSA nancial rms had increased to 84 directors among the 188 available observations. These results suggested that ruling family directors in the KSA
were concentrated in non-nancial rms.
3.4. Control variables
As suggested by the literature related to risk disclosure and corporate transparency, we controlled for several rm-level factors.
First, following Chaney et al. (2011) and Wan-Hussin (2009), we controlled for the level of rm governance (CG_Index).18 We in-
cluded rm size, measured as the natural logarithm of market capitalization (LNMVAL), as previous studies have shown that rm
size is consistently and positively related to increased disclosure levels (Al-Shammari et al., 2008). We controlled for Leverage,
measured as total liabilities scaled over total assets, and for return on equity (ROE), measured as pre-tax prot scaled by total eq-
uity (Aljifri & Hussainey, 2007). Firm growth was included as a control variable, due to the likelihood of differences in the extent
of information asymmetry at different points in the life cycle of the rms. Firm growth was measured as the ratio of the book
value to the market value of equity (BVMV) (Aljifri & Hussainey, 2007).
Furthermore, Al-Shammari et al. (2008) nd that compliance with accounting standards in GCC countries is driven by auditor
quality, and thus we controlled for auditor quality by incorporating a variable for the presence of an external Big-4 audit rm
(AuditQ). Following Ali et al. (2007) and Wan-Hussin (2009), we controlled for the presence of family directors (Family_Dir), mea-
sured as a dummy variable equal to 1 if there was at least one family member on the board and 0 otherwise. We also controlled
for government representation on the board (GovDir), measured as 1 if present and 0 otherwise (Cheng & Courtenay, 2006).19 We
followed Al-Shammari et al. (2008) in controlling for ownership concentration (Ownership_Con.) and followed John, Litov, and
Yeung (2008) to control for earnings volatility (Earning_Volatility). Finally, year, industry, and country were included as dichoto-
mous variables in all regressions. Continuous variables were winsorized at the 1st and the 99th percentiles.
3.5. Empirical model
3.5.1. Ordinary least squares (OLS)

We used the following OLS model to examine the relationship between the existence of risk disclosures and political connec-
tions of the ruling family directors (RoyalF_Dir):
MRDFACTOR=EMRD=QMRDi;t 0 1 RoyalF Dir i;t 2 CG Indexi;t 3 LNMVALi;t 4 Leveragei;t a5 ROEi;t

6 Big4i;t; j 7 BVMV 7 Family Diri;t 8 Gov Dir i;t 9 Ownership Con: i;t
9 Earning Vol 3yi;t Year Dummies Industry Dummies Country Dummies i;t 4
4. Empirical results and discussion
4.1. Descriptive statistics
Table 3, Panel B reports the descriptive statistics for the variables included in the regression models. The mean (standard de-
viation) values for the MRDFACTOR, EMRD, and QMRD indices are 0.760 (0.270), 0.566 (0.282), and 7.736 (3.768), respectively. Our
results are similar to those of Miihkinen (2012). On average, 31.2% of the listed nancial rms have at least one ruling family
member on their board of directors. This result is similar to the lower bound of ruling family membership on boards that is re-
ported by Boubakri et al. (2012b). In addition, GovDir and Family_Dir have relatively high means of 40% and 65%, respectively.
This nding is consistent with those of the Halawi & Davidson (2008), which indicates that GCC countries are controlled by dom-
inant families and state ownership. Moreover, Table 3, Panel B shows a large dispersion among the sample rms in terms of the
control variables, indicating considerable diversity in our sample.
4.2. Univariate t-test
Table 3, Panel C reports the mean differences and t-statistics of MRDFACTOR and EMRD for rms with political connections
(RoyalF_Dir). Halawi & Davidson (2008) suggests that the signicance of political connections varies by industry. Hence, we report
the mean differences and t-statistics for both MRDFACTOR and EMRD by industry. The mean differences in the MRDFACTOR and
EMRD of banks with political connections have t-statistics of 1.50 and 2.37, respectively. Furthermore, the mean differences for
the MRDFACTOR of nancial industry rms, investment industry rms, and the pooled sample have t-statistics of 2.28, 1.57,
and 2.87, respectively, for politically connected rms. However, we fail to nd a signicant mean difference (t-statistic) for
EMRD in the non-bank rms. Our ndings support the hypothesis that although banks are risk-oriented and should disclose
more comprehensive risk information, the effects of ruling directors on their boards negate these factors. Furthermore, results
18
The CG_Index comprised seven items: a) four dummy variables related to board composition ([1] CEO and Chairman duality, [2] BOD [board of directors] external
business, [3] Chairman's membership in at least one board monitoring committee, and [4] independent directors' majority status on the board (Chaney et al., 2011;
Wan-Hussin, 2009)); b) rms with an audit committee were scored as 1 and 0 otherwise; c) rms with a remuneration or nomination committee were scored as 1
and 0 otherwise; and d) rms with a risk committee were scored as 1 and 0 otherwise. We then divided the total score by the maximum expected score of the seven
items.
19
Ali et al. (2007) nd that family directors and family ownership provide incentives to reduce a rm's governance disclosures. In a sense, these managers tend to be
more deeply involved in their rms, possess their rm's information more completely, prevent its disclosure, and use it for self-benet.
Table 3
Political connection and descriptive statistics.
Panel A: Political connection based on country
Country Mean Sum
Bahrain 0.153 13
KSA 0.044 4
Kuwait 0.201 30
Oman 0.344 43
Qatar 0.923 79
UAE 0.265 45
Total 0.304 214
Panel B: Descriptive statistics
Variable N Mean S.D. Min Mdn Max
MRDFACTOR 667 0.760 0.27 0.170 0.730 3.800

EMRD 667 0.566 0.282 0.000 0.550 1.280
QMRD 667 7.736 3.768 0.000 7.700 17.000
RoyalF_Dir 667 0.312 0.464 0.000 0.000 1.000
CG_Index 667 0.344 0.252 0.000 0.429 0.857
LNMVAL 667 6.12 1.738 0.000 6.110 9.736
Leverage 667 0.666 0.32 0.003 0.719 2.393
ROE 667 8.58 15.408 53.100 11.000 41.500
BVMV 667 1.842 2.448 0.000 1.253 19.357
AuditQ 667 0.9 0.301 0.000 1.000 1.000
GovDir 667 0.406 0.492 0.000 0.000 1.000
Family_Dir 667 0.651 0.477 0.000 1.000 1.000
Ownership_Con 667 2.838 12.709 0.000 0.541 103.600
Earning_ Volatility 667 1.456 3.554 0.000 0.242 35.034
Panel C: Comparison of mean MRDFACTOR and EMRD across political connected (RoyalF_Dir) rms based on industry.
MRDFACTOR
1
RoyalF_Dir Mean Difference t-statistic
Connected =1 Connected =0
Bank 0.689 0.576 0.112 1.50

Financial 0.728 0.523 0.204 2.28**
Insurance 0.532 0.458 0.073 0.66
Investment 0.728 0.532 0.195 1.57
Pool 0.668 0.532 0.135 2.87***
2 EMRD
RoyalF_Dir Mean Difference t-statistic
Connected =1 Connected =0
Bank 0.712 0.781 0.068 2.37**

Financial 0.446 0.471 0.024 0.73
Insurance 0.38 0.396 0.016 0.52
Investment 0.39 0.276 0.114 1.95*
Pool 0.56 0.555 0.004 0.21
*, **, and *** denote signicance at 1%, 5%, and 10%, respectively.
not reported here suggest that Family_Dir (which has been used in the corporate transparency and political connection literature,
e.g., Chaney et al., 2011) is negative and signicant, with a t-statistic of 1.60.
4.3. Correlation analysis
Table 4 provides the Pearson correlation matrix for the main variables included in the regression analysis. There are positive
correlations between the proxies for market risk disclosure, namely MRDFACTOR, EMRD, and QMRD. We also nd a negative and
signicant correlation between MRDFACTOR and RoyalF_Dir (p b 0.10) as well as a positive but insignicant correlation between
EMRD and QMRD for disclosures and RoyalF_Dir. Moreover, EMRD and QMRD are signicantly and positively correlated with
CG_Index, LNMVAL, Leverage and GovDir (p b 0.01), and Earning_Volatility is negatively and signicantly correlated with all of
the market risk disclosure proxies (at p b 0.10 or better).
4.4. Regression analysis
Political connections and market risk disclosures.
Table 4
Pearson correlation matrix.
A. Al-Hadi et al. / The International Journal of Accounting xxx (2016) xxxxxx

1 2 3 4 5 6 7 8 9 10 11 12 13 14
1 MRDFACTOR 1
2 EMRD 0.289*** 1
3 QMRD 0.287*** 0.976*** 1
4 RoyalF_Dir 0.09** 0.003 0.017 1
5 CG_Index 0.15*** 0.186*** 0.172*** 0.258*** 1
6 LNMVAL 0.06 0.492*** 0.496*** 0.003 0.114*** 1
7 Leverage 0.032 0.301*** 0.286*** 0.039 0.009 0.358*** 1
8 ROE 0.019 0.073* 0.06 0.171*** 0.012 0.167*** 0.072* 1
9 BVMV 0.055 0.052 0.06 0.134*** 0.109*** 0.304*** 0.357*** 0.21*** 1
10 AuditQ 0.059 0.253*** 0.259*** 0.182*** 0.169*** 0.199*** 0.196*** 0.115*** 0.045 1
11 GovDir 0.144*** 0.214*** 0.196*** 0.122*** 0.171*** 0.283*** 0.09* 0.103*** 0.023 0.165*** 1
12 Family_Dir 0.02 0.008 0.01 0.009 0.085** 0.034 0.008 0.012 0.079** 0.13*** 0.062 1
13 Ownership_Con 0.034 0.029 0.027 0.103*** 0.112*** 0.051 0.061 0.018 0.138*** 0.059 0.067* 0.054 1
14 Earning_ Volatility 0.065* 0.072* 0.077** 0.092** 0.042 0.276*** 0.12*** 0.076* 0.028 0.078** 0.186*** 0.066* 0.001 1
MRDFACTOR is mandatory market risk disclosures obtained from component factor analysis of both qualitative and quantitative Value-At-Risk (VaR) and Sensitivity Analysis (Sen) market risk disclosures after Miihkinen
(2012); EMRD is the number of qualitative and quantitative VaR and Sen market risk disclosure items divided by the maximum number of items possible; this variable is a measure of the extent of mandatory market
risk disclosures. QMRD is sum of the qualitative and quantitative market risk disclosures scaled by the number of VaR and Sen risk exposures; this variable is a measure of the quality of market risk disclosures. RoyalF_Dir
is a rm with a ruling family member on the board of directors, scored as 1 and otherwise 0; CG_Index is rm's corporate governance index comprising seven items (see Table 1). LNMVAL is natural log of a rm's end of year
market capitalization; Leverage is total short and long-term debt divided by total assets; ROE is Return on Equity; BVMV is growth measure of book value divided by market value of stock at market end-of-year; AuditQ is a
dummy variable that takes on the value of 1 if the rm is audited by one of the Big Four auditors, otherwise 0; Family_Dir is a dummy variable that takes on the value of 1 if the rm is owned by family company and has
family representatives on the BOD, 0 otherwise; GovDir is a dummy variable that takes on the value of 1 if the rm is owned by the government (or one of its agencies) and has at least one government representative on the
board of directors, 0 otherwise; Ownership_Con is percentage government and institutional shareholding; Earning_Volatility is the standard deviation of return of assets (ROA) for the previous three years.
*, **, and *** denote signicance at 1%, 5%, and 10%, respectively (two-tail).
11
Table 5
Association between political connections (RoyalF_Dir) and disclosure types (MRDFACTOR, EMRD, and QMRD) using ordinary least squares (OLS) regression.
MRDFACTOR (OLS) EMRD (OLS) QMRD (OLS)
Intercept 1.134*** 0.204*** 2.509***

(5.73) (2.94) (2.73)
RoyalF_Dir 0.187*** 0.069*** 0.736***
(3.72) (3.41) (2.71)
CG_Index 0.300** 0.109** 1.407**
(2.23) (2.45) (2.34)
LNMVAL 0.056** 0.059*** 0.866***
(2.31) (6.95) (7.55)
Leverage 0.112** 0.094*** 1.098***
(2.43) (3.88) (3.47)
ROE 0.001 0.001 0.014*
(0.70) (0.98) (1.93)
BVMV 0.023*** 0.024*** 0.351***
(2.97) (5.29) (5.51)
AuditQ 0.023 0.029 0.506
(0.61) (1.22) (1.63)
GovDir 0.082*** 0.016 0.040
(2.79) (0.80) (0.14)
Family_Dir 0.038 0.037** 0.436*
(0.93) (2.21) (1.92)
Ownership_Con 0.002*** 0.000 0.002
(3.82) (0.61) (0.32)
Earning_ Volatility 0.004 0.007*** 0.095***
(1.07) (3.25) (3.06)
Year Dummies Yes Yes Yes
Industry Dummies Yes Yes Yes
Country Dummies Yes Yes Yes
N 667 667 667
adj. R-sq 6.8% 47.3% 46.4%
Robust t-statistics are in parentheses.
The variable denitions are provided in Table 1.
Our main independent variable of interest is RoyalF_Dir, which is a proxy for politically connected rms, and our dependent
variables are the extent and quality of mandatory risk disclosures (MRDFACTOR, EMRD, and QMRD). We predict a1to be negative
according to H1. Table 5 presents the OLS estimates for H1.
The OLS coefcients of RoyalF_Dir for each of the dependent variables (MRDFACTOR, EMRD, and QMRD) are negative and statisti-
cally signicant (with a1 = 0.187, 0.069, and 0.736, respectively, at p b 0.01). These results are consistent with the assertion by
Chaney et al. (2011) that the managers of politically connected rms practice discretion in their disclosure practices. This discretion
is manifested as a reduction in the extent and quality of risk disclosures despite the cruciality of risk management and risk reporting
for nancial rms. In terms of economic signicance, the reported coefcient implies a 0.093.2 basis point change in market risk
reporting for one standard deviation change in the existence of RoyalF_Dir, as calculated, for example, by the rst model (Model
1) [0.464 (SD of RoyalF_Dir) * 0.187 (regression coefcient on RoyalF_Dir)/0.27 (SD of MRDFACTOR). These results are consistent
with the view that politically connected rms dedicate fewer resources for identifying, monitoring, managing, and reporting nancial
risks. With respect to our control variables, rms with higher levels of LNMVAL, CG_Index, Leverage, and GovDir are positively and sig-
nicantly (p b 0.01) associated with the proxy measures of risk disclosure, but BVMV is negatively and signicantly (p b 0.01) asso-
ciated with risk disclosure proxies. However, family directorship (Family_Dir) is an impediment to both the extent and quality of
market risk disclosures, as is evidenced by the negative and statistically signicant coefcients for this variable. These results are con-
sistent with those of prior research (e.g., Ali et al., 2007), indicating that family directors may use their power to restrict the dissem-
ination of information to the public, thereby controlling their rm's information environment and acting more out of self-interest
than concern for the benet of shareholders. The adjusted R-square for the model that includes MRDFACTOR is 6.8%, but the adjusted
R-squares for the models that include EMRD and QMRD are 47.3% and 46.4%, respectively.20
5. Robustness and additional tests
5.1. Robustness test: two-stage least squares (2SLS)
It is possible that the statistically signicant and negative association between risk disclosure proxies and the presence of
ruling family members on the boards is a consequence of endogenous factors. In other words, a certain level or quality of risk
20
The variance in adjusted R-squares between the models may be related to the use of particular factors in some models (e.g., MRDFACTOR). Therefore, compared to
the EMRD and QMRD models, the use of current control variables and the OLS model for MRDFACTOR may not sufciently improve the explanatory power of this model.
Furthermore, due to including country dummies, some of the country-year common factor observations are not accounted for in the factors.
Table 6
Association between political connections (RoyalF_Dir) and disclosure types (MRDFACTOR, EMRD, and QMRD) using two-stage least-squares (2SLS) regression.
MRDFACTOR (2SLS-2nd stage) EMRD (2SLS-2nd stage) QMRD (2SLS-2nd stage)
Intercept 1.075*** 0.159* 1.924*

(4.86) (1.93) (1.82)
RoyalF_Dir 0.211*** 0.307*** 3.463***
(3.00) (5.98) (4.97)
CG_Index 0.308** 0.174*** 2.259***
(2.09) (3.30) (3.25)
LNMVAL 0.033 0.065*** 0.935***
(1.41) (6.63) (7.25)
Leverage 0.091** 0.145*** 1.599***
(1.99) (3.78) (3.31)
ROE 0.000 0.000 0.003
(0.46) (0.26) (0.37)
BVMV 0.018** 0.030*** 0.402***
(2.52) (5.55) (5.53)
AuditQ 0.097** 0.088*** 1.139***
(2.37) (3.04) (3.05)
GovDir 0.053* 0.023 0.138
(1.95) (0.96) (0.45)
Family_Dir 0.019 -0.050** -0.545**
(0.54) (2.44) (2.02)
Ownership_Con -0.002*** -0.001** -0.015*
(3.41) (2.29) (1.86)
Earning_ Volatility -0.005 0.006** 0.082**
(1.35) (2.23) (2.32)
Year Dummies Yes Yes Yes
Industry Dummies Yes Yes Yes
Country Dummies Yes Yes Yes
N 591 618 618
adj. R-sq 04.4% 35.4% 37.3%
Instrumental variables
Ruling-Female 0.635*** . .
(11.26) . .
Royal_C 0.942*** . .
(20.50) . .
CGovPoli_Stabi . 0.490*** 0.490***
. (13.81) (13.81)
Board Size 0.001 0.0008 0.0008
(0.34) (0.21) (0.21)
Under-indemnication test
Kleibergen-Paap rk LM statistic 135.283 88.580 88.580
P-value 0.0000 0.0000 0.0000
Weak-identication test
Kleibergen-Paap rk Wald F statistic 167.663 96.317 96.317
10% maximal IV size 22.30 19.93 19.93
Over-identication test
Hansen J statistic 3.670 1.284 1.335
Chi-sq.(6) P-Value 0.1596 0.2572 0.2479
Endogeneity test
Durbin-Wu-Hausman tests 3.429 29.166 19.599
Chi-sq.(1) P-Value 0.0640 0.0000 0.0000
Robust t-statistic are in parentheses.
The variable denitions are in Table 1.
Ruling-Female dummy variable if rm board includes women from a ruling family on the board, Royal_C is calculated as the country average of ruling family di-
rector, and CGovPoli_Stabi is the country's political stability index according to the World Bank.
disclosure may be determined by board-related factors other than the presence of ruling family members. To consider this pos-
sibility, we re-ran each regression model to test for endogeneity, using a 2SLS estimator. We also report the results of a two-
stage least-squares (2SLS) regression, which was conducted to mitigate the endogeneity bias of the OLS model that would
arise if RoyalF_Dir and the error term were correlated. Three factors that may contribute to such endogeneity are reverse causality,
omitted unobservable characteristics, and measurement error (Aivazian, Ge, & Qiu, 2005). To formally address these concerns, we
Table 7
Association between ruling family characteristics and disclosure types (MRDFACTOR, EMRD, and QMRD).
MRDFACTOR EMRD QMRD
Intercept 1.143*** 0.978*** 1.135*** 0.201*** 0.203*** 0.208*** 2.466*** 2.500*** 2.499***
(5.78) (7.18) (5.70) (2.90) (2.92) (3.00) 2.7 2.71 (2.72)
RoyalNumDir 0.092*** 0.038*** 0.455***
(4.22) (3.72) (3.06)
RoyalChairman 0.086** 0.055** 0.365
(2.27) (2.56) (1.27)
RoyalRank 0.072 0.025 0.485*
(1.29) (1.22) (1.71)
Country FE Yes Yes Yes Yes Yes Yes Yes Yes Yes
Year FE Yes Yes Yes Yes Yes Yes Yes Yes Yes
Industry FE Yes Yes Yes Yes Yes Yes Yes Yes Yes
All controls Yes Yes Yes Yes Yes Yes Yes Yes Yes
N 667 667 667 667 667 667 667 667 669
Adj. R-sq 06.5% 05.5% 04.6% 6.50% 5.50% 4.60% 46.6% 45.8% 45.7%
*, **, and *** denote signicance at 1%, 5%, and 10%, respectively (two-tail). The control variables have been included in this model but are not shown for reasons of
brevity. The variable denitions are provided in Table 1. Robust t-statistic in provided in brackets.
adopted a two-stage instrumental variable (IV) approach and re-examined the ndings reported along with the OLS estimates for
each model. The rst stage of this procedure involved regressing the endogenous variables (RoyalF_Dir) on selected instrumental
variables (IVs) and on the exogenous variables from our main analyses:
X X
^
RoyalF Dir i;t a0 a1 IV INSTRUMENTS Control Variablesi;t U Year firm 5
i;t
^ Control Variables
DMRDFACTOR or EDMRDi;t or QDMRDi;t a0 a1 U 6
i;t i;t
where U ^ = is the error term from Eq. (5), and the control variables are the same set of controls included in Eqs. (4) and
(5).
In the second stage, we replaced the endogenous variables with tted values from the rst-stage regressions. Studies of cor-
porate transparency have extensively adopted this two-stage IV approach to address endogeneity concerns.21 The approach, how-
ever, is appropriate only if the IVs are correlated with the endogenous regressor but uncorrelated with the error term of the
second-stage regression. In this context, effective IVs are exogenous variables that are economically related to the political connec-
tion proxy but are uncorrelated with the error term of the second-stage regression. We adopted the approach used by Larcker and
Rusticus (2010) and applied a mix of rm-specic and country-level variables as IVs, which included BoardSize, Ruling-Female,
Royal_C, and CGovPoli_Stabi.22
Table 6 reports the second-stage regression results for each model in which the endogenous variables are regressed on
the IVs and the exogenous variables from our previous estimates. The negative and signicant relationship between political
connections (RoyalF_Dir) and the extent and quality of market risk disclosure proxies (MRDFACTOR, EMRD, and QMRD) re-
mains robust after accounting for endogeneity. The coefcients on RoyalF_Dir between MRDFACTOR, EMRD and QMRD are
a 1 = 0.211, 0.307, and 3.463, respectively, with p b 0.01. This suggests that endogeneity cannot account for the
21
See Larcker and Rusticus (2010) for more detail on concurrent accounting studies that adopt the two-stage IV approach.
22
First, our use of BoardSize as an IV is justied on the basis that a larger board has greater diversity. Bradbury (1990) argues that boards have an incentive to increase
their size and choose new board directors, which suggests that the size of a board is correlated with the selection of RoyalF_Dir. Second, female directors in the GCC
region have recently appeared in many rms, specically women of ruling families. Halawi & Davidson, (2008) nds that although GCC rms are among the lowest-
ranked countries in terms of the presence of women on boards of directors, they report higher results than many developed countries such as Japan. This pattern
may be due to the social and cultural norms of rulers in the GCC. For example, Al Walid Bin Talal, the richest ruling family board member in the Middle East supports
his sister, Sultana Nurul Al Walid, who owns 16% of the Savoy Hotel in London (Mazaheri, 2013). The presence of women from a ruling family (Ruling-Female) correlates
positively with the presence of RoyalF_Dir on a board. Therefore, we selected female directors from ruling families as an IV for RoyalF_Dir. We then included Royal_C
(country average of ruling family directors). The use of country RoyalF_Dir as an instrument can be justied on the grounds that the country-level representation of
ruling family directors of rms indicates the potential existence and power of RoyalF_Dir in the rms that belong to each particular country. We expect a positive as-
sociation between RoyalF_Dir and Royal_C. Finally, Larcker and Rusticus (2010) suggest that country-level economic factors are reasonable IV measures. We included
CGovPoli_Stabi, which represents each country's political stability index according to the World Bank (Kaufmann, Kraay, & Mastruzzi, 2009). This variable captures
the likelihood that the country will become politically destabilized. We expect that a politically destabilized country has implications for the ruler and his or her rela-
tives, in the sense that destabilization decreases their economic and political benets. The GCC is the most stable region in the Middle East. Therefore, a higher
CGovPoli_Stabi index reects the greater benets of RoyalF_Dir in country i.
Table 8
Association between political connections and disclosure types (MRDFACTOR and EMRD) based on presence of family and government directorships.
FamilyDir =1 FamilyDir =0 GovDir = 1 GovDir = 0 FamilyDir =1 FamilyDir =0 GovDir =1 GovDir =0

DV MRDFACTOR EMRD
Intercept 1.124*** 0.951*** 1.759*** 0.806*** 0.066 0.368*** 0.287* 0.297***

(6.37) (4.42) (3.40) (18.88) (0.73) (3.14) (1.88) (3.47)
RoyalF_Dir 0.077** 0.325*** 0.386*** 0.054*** 0.060** 0.049 0.104*** 0.063**
(2.023) (3.625) (3.388) (3.003) (2.31) (1.51) (3.25) (2.56)
Country FE Yes Yes Yes Yes Yes Yes Yes Yes
Year FE Yes Yes Yes Yes Yes Yes Yes Yes
Industry FE Yes Yes Yes Yes Yes Yes Yes Yes
All controls Yes Yes Yes Yes Yes Yes Yes Yes
N 434 233 271 396 434 233 271 396
adj. R-sq 06.8% 19.6% 09.7% 13.1% 50.5% 45.7% 56.7% 39.4%
*, **, and *** denote signicance at 1%, 5%, and 10%, respectively (two-tail). The control variables have been included in this model but are not shown for reasons of
brevity. The variable denitions are provided in Table 1. Robust t-statistics are shown in parentheses.
negative relationship between political connections and market risk disclosure, given that the signs of all of the control var-
iables remain unchanged.23
5.2. Additional tests
5.2.1. Alternative measures of political connection

We conducted several additional tests of our main analyses.24 First, we used alternative measures of political connection that
comprised the natural log of the number of ruling family directors on the board (RoyalNumDir) and the existence of a chairman of
the board who was a ruling family member (RoyalChairman), scored as 1 (and 0 otherwise). The existence of a board member
with close family ties to ruling families (RoyalRank) was scored as 1 if present and 0 otherwise. Previous studies investigating po-
litical connections on disclosure have tended to rely on a single measure (Chaney et al., 2011). We were able to rely on several
measures of ruling family members in this study, as the annual reports of nancial rms disclosed full family names and tribal
afliations of their directors. The results in Table 7 show that each of these additional measures of political connection are con-
sistently, signicantly, and negatively associated with the extent and quality of market risk disclosure.
5.2.2. Government and family board directorships

Second, we subdivided our sample into two groups based on the presence of family and government representation on the
board. Results provided in Table 8 show that our inference with respect to the association between political connections
(RoyalF_Dir) and the extent or quality of market risk disclosure are inuenced by the existence of family or government represen-
tatives on boards. Both MRDFACTOR and EMRDs are reduced where royal family or government ofcials are represented on the
board.
5.2.3. Discretionary accruals

Third, we re-ran the main regression model, controlling for discretionary accruals (Dechow & Dichev, 2002). Specically, we
estimated the OLS model by country and by industry, following Chen, Hope, Li, and Wang (2011). The unreported results show
that the relationship between risk disclosure and political connections does not change after controlling for discretionary accruals.
5.2.4. Government and royal family board directorships

We also assess the impact of both government and royal family representation in the board on risk disclosure patterns. We
include dummy variables for government representation and ruling family representation on the boards, and an interaction
term of these two variables in the regression model after controlling for heteroscedasticity. Government representation
(GovDir) is scored as 1 if a member of the board is a government ofcial, otherwise 0. Using EMRD as the dependent variable,
we found evidence (shown in Table 9) that the interaction between RoyalF_Dir and GovDir further suppresses the extent of
risk disclosures.25 Government representation appears to further suppress the extent of market risk disclosures and is consistent
23
In support of the theoretical connection with the use of IVs, we conducted an under-identication test, which consisted of combining Hansen's test of over-
identifying restrictions with Hausman's endogeneity test (1978). Based on work by Cameron and Trivedi (2010), Table 6 presents the test statistics for the relevance
and validity of the 2SLS estimates. The results show that the excluded IVs are correlated with the endogenous regressor, because the Cragg-Donald-Wald F-statistic
is greater than the Stock and Yogo (2005) critical value. The results from Hansen's test of over-identifying restrictions do not reject the null hypothesis (p N 0.10), which
suggests that the IVs are uncorrelated with the error term but are correctly excluded in the second-stage regression, which reects the validity of the IVs used for 2SLS.
Finally, we conducted the Hausman test to ascertain whether an endogeneity problem existed. The Hausman test strongly rejected (p b 0.10) the exogeneity of
RoyalF_Dir. Hence, the 2SLS estimates are shown to be preferable to the OLS estimate.
24
For each of the additional tests, we included the suite of control variables used in our base model (see Table 5), but these results are not included in the tables for the
sake of brevity.
25
We also ran this model using institutional ownership. In the unreported results, we nd that ruling family directors have no power to reduce the extent or quality of
risk reporting in the presence of high institutional ownership even when the number of ruling family members on the board increases.
Table 9
Association between government ownership and ruling family directorships and disclosure types (EMRD and QMRD).
Panel A: Government ownership directors Panel B: Government ownership and royal family directors
EMRD QMRD EMRD QMRD

Intercept 0.621*** 8.084*** Intercept 1.5547* 0.1320*
(8.52) (8.60) (1.67) (1.91)
Gov_Ownership 0.001*** 0.016*** Gov_Ownership 0.013 0.001
(2.95) (2.77) (1.25) (1.36)
RoyalF_Dir 0.567** 0.058***
(2.04) (2.85)
RoyalF_Dir * Gov_Ownerhsip 0.161** 0.005
Country FE Yes Yes (2.44) (1.21)
Year FE Yes Yes Yes Yes
Industry FE Yes Yes Yes Yes
All controls Yes Yes Yes Yes
N 667 667 N 667 667
Adj. R-sq 74.0% 69.9% Adj. R-sq 74.2% 70.3%
We rst regress proxies of market risk disclosure and government ownership concentration:
EMRDi,t/QMRDi,t = 0 + 1Gov_Ownerhshipi,t + 2CG_Indexi,t + 3LNMVALi,t + 4Leveragei,t + 5ROEi,t + 6BVMBi,t + 7AuditQi,t + 8Volatility_ROE_3yeari,t +
Country Dummies + Year Dummies + IND Dummies + i,t (Eq. 7).
The equation below shows the interaction term between government ownership concentration (Gov_Ownership) and ruling family director (RoyalF_Dir) and its
association with EMRD and QMRD:
EMRD/QMRDi,t = 0 + Gov_Ownerhsip%i,t + 2RoyalF_Dir i,t + 3 Gov_Ownerhsip%i,t * RoyalF_Diri,t + 4 CG_Indexi,t + 5LNMVALi,t + 6Leveragei,t + 7ROEi,t +
8BVMBi,t + 9AuditQ i,t + 10Volatility_ROE_3yeari,t + Country Dummies + Year Dummies + IND Dummies + i,t (Eq. 8).
The control variables have been included in this model but are not shown for reasons of brevity. The variable denitions are in Table 1. Robust t-statistics are in
parentheses.
with the notion that government representation on the board is likely to be an important determinant of business decisions by
rms in the GCC, including decisions related to the nature and degree of risk reporting (e.g., Polsiri & Jiraporn, 2012). The probable
reason for this nding is that the governments in the GCC, unlike those in Western economies, may comprise royal family direc-
tors, who may lack the incentive or the necessary expertise to consider adequate risk reporting (Al-Shammari et al., 2008; Hertog,
2012; Polsiri & Jiraporn, 2012).
5.2.5. Mandatory versus discretionary disclosures

Fifth, we subdivided our risk disclosure index into mandatory market risk disclosure and voluntary market risk disclosure
groups to assess whether the negative association between risk disclosures and ruling family membership held for both categories
of disclosure. Table 10 shows a negative and signicant association between both mandatory and voluntary disclosures and po-
litically connected rms.
5.2.6. Market risk, nancial distress, and ruling family directorships

Sixth, we assessed the association between the existence of ruling family directors and a rm's risk disclosures when the
rm was in a state of high risk and separately when the rm was in a state of nancial distress. Prior studies that have exam-
ined the association between rm risk and disclosure levels have produced mixed results (see, e.g., Elshandidy, Fraser, &
Hussainey, 2013). In terms of risk disclosures, Linsley and Shrives (2006) argue that the association between rms' risk levels
(as measured by beta) and risk disclosure levels could potentially be positive or negative. Abraham and Cox (2007) nd a pos-
itive association between risk levels, as measured by the variance of stock returns, and the extent of risk disclosures. Lin,
Owens, and Owers (2010) show that mandatory disclosure of market risk information is positively associated with a rm's
total, systematic, and unsystematic risk. Vandemaele, Vergauwen, and Michiels (2009) also report a positive association be-
tween systematic risk and aggregated risk disclosure levels. Nevertheless, Marshall and Weetman (2002) indicate that high-
risk rms are likely to disclose less than those exposed to marginally lower risk.26 In the GCC context, politically connected
26
In Table 11, we have negative coefcients for beta and the Altman-Z score, whereas in Tables 5 and 6, we have positive coefcients for Leverage and
Earning_Volatility. Increased leverage is expected to increase disclosures because debt covenants may require a certain degree of information transparency
as a measure to control potential information asymmetry issues or agency-related problems (Deumes & Knechel, 2008; Elshandidy et al., 2013). We also nd
that earnings volatility is positively associated with risk disclosures consistent with the ndings of Elshandidy et al. (2013) and Jayaraman (2008). Increased
disclosures in the face of earnings volatility may reduce investor uncertainty and may control agency-related problems. The negative coefcient for beta is ex-
pected, as rms tend to reduce risk disclosures when exposed to less market risk. This is consistent with Elshandidy and Neri (2015), who nd a negative as-
sociation between beta and risk disclosure for Italy- and U.K.-listed rms. We also nd that distressed rms disclosed less risk information, which is consistent
with Koch (2002).
Table 10
Association between mandatory and voluntary disclosure types and ruling family directorships.
Mandatory Index+ Voluntary Index#
Intercept 0.153*** 0.095

(2.81) (1.34)
RoyalF_Dir 0.055*** 0.087***
(2.64) (4.41)
Country FE Yes Yes
Year FE Yes Yes
Industry FE Yes Yes
All controls Yes Yes
N 667 667
adj. R-sq 31.9% 33.8%
+
Mandatory_Item_Indexi,t = 0 + 1RoyalDiri,t + 2CG_Indexi,t + 3LNMVALi,t + 4Leveragei,t + 5ROEi,t +
6BVMBi,t + 7AuditQi,t + 8GovDiri,t + 9FamilyDiri,t + 10Ownership_Coni,t + 11Volatility_ROE_3yeari,t +
#
Voluntary_Item_Indexi,t = 0 + 1RoyalDiri,t + 2CG_Indexi,t + 3LNMVALi,t + 4 Leveragei,t + 5ROEi,t +
6BVMBi,t + 7AuditQi,t + 8GovDiri,t + 9FamilyDiri,t + 10Ownership_Coni,t + 11Volatility_ROE_3yeari,t +
The control variables have been included in this model but are not shown for reasons of brevity. The variable
denitions are in Table 1. Robust t-statistics are in parentheses.
rms could have much lower market risk due to the protection they are likely get from royal patronage in times of economic
shocks or nancial distress.
We followed Linsley and Shrives (2006) and Elshandidy et al. (2013) in using beta as a measure of market risk. We
calculated the Altman Z-score to proxy for nancial distress. We then re-ran the models separately, including beta and
the Altman Z-score, and interacted these measures with ruling family membership on the board. Table 11 (Panel
A) shows that the interaction term between beta and ruling family directors is negative and statistically signicant
for both the extent and the quality of market risk disclosures. We also nd that the coefcient of the interaction
term between ruling family directors and our measure of nancial distress (Altman Z-score) is negatively and signi-
cantly associated with both the extent and the quality of market risk disclosures. This nding suggests that disclosures
are made opportunistically under these conditions. Ruling family directors do not consider the disclosure of additional
risk information as necessary or relevant when distressed or when subject to higher levels of risk. One possible reason
for this nding is that ruling family directors do not want to attract unnecessary attention through greater disclosures
in difcult nancial times. However, a more likely reason is that ruling family directors consider that their connections
with key stakeholders, such as other royal family groups or government ofcials, will protect the rm in difcult times.
Thus, politically connected rms could have lower disclosure levels, even in the face of higher risk levels or of nancial
distress, due to the protection that their connections may offer to the rm. Protection through ruling family member-
ship could be attained in such areas as markets, access to capital, business promotion, or regulatory and government
support.
6. Conclusions
Using a sample comprising 667 rm-year observations, we document that the extent and quality of market risk
disclosures provided in the annual reports of publicly listed GCC nancial rms are signicantly lower for rms
that are politically connected. These results are consistent when using several alternative proxy measures of political
connection, including the presence of ruling family members on the board of directors, rms chaired by a ruling fam-
ily member, the presence of a board representative directly connected with a ruling family member, and the number
of ruling family members on the board. Furthermore, politically connected rms appear to have lower disclosure
levels even in the face of higher risk levels or in times of nancial distress, possibly due to the protection that
their connections may offer to their rms. Our results are supported by agency theory tenets, which suggest that po-
litically connected rms seize the private benets of shareholders in an opportunistic manner. Our regression results
are robust after controlling for endogeneity.
This study makes the following contributions. First, although a few studies have examined the association between
political connections and disclosure in emerging economies (e.g., Chaney et al., 2011; Guedhami et al., 2014; Leuz &
Oberholzer-Gee, 2006), to the best of our knowledge, this is the rst study to explicitly examine the association be-
tween political connections and the extent or quality of market risk disclosures of GCC-listed nancial rms.
Table 11
Impact of risk, nancial distress, and political connections on risk disclosures.
Panel A: Association between market risk disclosures (EMRD and QMRD), market risk (Beta), and political connections (RoyalF_Dir)
EMRD QMRD
Model 1 Model 2
Intercept 1.1361 0.0951

(1.37) (1.52)
RoyalF_Dir 0.2219 0.0063
(0.73) (0.27)
Beta 1.0947** 0.0930***
(2.43) (2.80)
RoyalF_Dir * Beta 0.0127*** 0.0009***
(11.20) (11.23)
Country FE Yes Yes
Year FE Yes Yes
Industry FE Yes Yes
N 535 535
Adj. R-sq 0.3025 0.3048
Panel B: Association between market risk disclosures (EMRD and QMRD), nancial distress (Altman Z-score), and political connections (RoyalF_Dir)
EMRD QMRD
Model 1 Model 2
Intercept 4.0773*** 0.3268***

(3.37) (3.63)
RoyalF_Dir 0.3162 0.0158
(0.97) (0.68)
Altman Z-Score 0.0266*** 0.0019***
(5.00) (4.28)
RoyalF_Dir * Z-Score 0.1476** 0.0117**
(2.13) (2.34)
Country FE Yes Yes
Year FE Yes Yes
Industry FE Yes Yes
N 457 457
Adj. R-sq 41.57% 45.24%
+
beta is measured as a rm's beta for at least 12 months (Al-Hadi et al., 2015).
EMRD/QMRDi,t = 0 + 1RoyalF_Diri,t + 2Betai,t + 3Betai,t * RoyalF_Diri,t + 4CG_Indexi,t + 5LNMVALi,t + 6Leveragei,t + 7ROEi,t + 8BVMBi,t + 9AuditQi,t +
10Volatility_ROE_3yeari,t + Country Dummies + Year Dummies + IND Dummies + i,t (Eq. 11).
parentheses.
EMRD/QMRDi,t = 0 + 1RoyalF_Diri,t + 2Z-Scorei,t + 3Z-Scorei,t * RoyalF_Diri,t + 4CG_Indexi,t + 5LNMVALi,t + 6Leveragei,t + 7ROEi,t + 8BVMBi,t +
9AuditQi,t + 10Volatility_ROE_3yeari,t + Country Dummies + Year Dummies + IND Dummies + i,t (Eq. 12).
brackets.
Second, this study's empirical ndings show that political connections play an important function in dictating risk dis-
closure practices for GCC rms. We provide consistent empirical evidence to indicate that the extent and quality of market
risk disclosures are signicantly and negatively associated with the existence of political connections.
Third, this study provides important insights into agency problems that are likely to affect the risk disclosure prac-
tices of GCC rms. These agency problems could lead to a more opaque reporting environment in which risk informa-
tion is selectively omitted or sufciently aggregated, such that specic risk exposures are not completely disclosed. In
particular, we nd that control by ruling family board members or connected directors has effects that are more sa-
lient than the newly established regulations around corporate governance or the efforts to comply with the reporting
requirements of IAS/IFRS.
Our ndings show a negative relationship between risk disclosure and political connections. To further explore this relation-
ship, future research could investigate the effects of royal family board membership on investment efciency and on the effective-
ness of various board committees. Finally, the results of this study should be informative for various capital market participants
such as shareholders, regulators, government administrators, auditors, and nancial analysts, especially concerning questions of
why rms use discretion in disclosing risk information.
Appendix A. Market Risk Disclosures Index

Value at Risk (VaR) Index: Qualitative Sources Sensitivity Analysis (Sen): Qualitative Sources
Items Items
a. Effect of VaR on cash ow/fair IFRS 7: Paragraphs 1840a, FRR. a. Effect of Sen on cash ow/fair IFRS 7: Paragraphs 1840a, FRR. 48:
value/earning (M) 48:305(a): 1F value/earning (M) 305(a): 1F
b. Two or more VaR effects on cash IFRS 7: Paragraphs 1840a, FRR. b. Two or more Sen effects on cash IFRS 7: Paragraphs 1840a, FRR. 48:
ow/fair value/earning (V) 48:305(a): 1F ow/fair value/earning (V) 305(a): 1F
c. Objective of risk management (M) IFRS 7: Paragraphs 1840a c. Objective of risk management (M) IFRS 7: Paragraphs 1840a
d. Policies of risk management (M) IFRS 7: Paragraph 33, FRR 0.48305 d. Policies of risk management (M) IFRS 7: Paragraph 33, FRR 0.48305
(a)(1) (a)(1)
e. Limitations of risk management (M) IFRS 7: 41C e. Limitations of risk management (M) IFRS 7: 41C
f. Other risk exposure except for interest f. Other risk exposure except for
rate, currency, and price risk (V) interest rate, currency, and price risk
(V)
g. Disclosure of gain from VaR (V) IFRS 7: Paragraph 25: B20 g. Disclosure of gain from Sen (V) IFRS 7: Paragraph 25: B20
h. Non-trading market risk (V) IFRS 7: Paragraph 20 h. Non-trading market risk (V) IFRS 7: Paragraph 20
i. Immaterial market risk exposure (V) IFRS 7: Paragraphs 1740a i. Immaterial market risk exposure IFRS 7: Paragraphs 1740a
(V)
j. Risk target of the rm (V) IFRS 7: Paragraph 40 j. Risk target of the rm (V) IFRS 7: Paragraph 40
k. Other stress testing (V) Basel II k. Other stress testing (V) Basel II
l. Qualitative description of the stress IFRS 7: Paragraphs 1920 & 19-a, FRR. l. Qualitative description of the stress IFRS 7: Paragraphs 1920 & 19-a,
test (V) 48: 305(a): 1 AD & Basel II test (V) FRR. 48: 305(a): 1 AD & Basel II
m. Stress test result (V) IFRS B19 B m. Stress test result (V) IFRS B19 B
n. Near-term risk exposure (M) FRR. 48,305(a) 4.a and IFRS B19 B n. Near-term risk exposure (M) FRR. 48,305(a) 4.a and IFRS B19 B
Qualitative Score: 14 points Qualitative Score: 14 points

Quantitative Index for Each Interest Rate, Foreign Exchange, Equity Price, and Quantitative Index for Each Interest Rate, Foreign Exchange, Equity Price,
Other Commodity Price and Other Commodity Price
VaR Characteristics Sen Characteristics
a. Holding period under VaR (M) IFRS: B20, Basel & FRR. a. Potential loss 10% (M) IFRS 7: B19a, B18b & FRR. 48,305(a)
48,305(a)(1)(iii): C (1)(ii)
b. Condence level (e.g., 99%, 95%) IFRS: B20, Basel & FRR. b. Potential loss b10% (V)
(M) 48,305(a)(1)(iii): C
c. Type of VaR model (M) IFRS: B20 c. Economic justication for b10% (V)
d. Data time frame (M) IFRS: B20, Basel & FRR. 48,305 A (i), d. Multiple scenarios (10% and 20%)
(i) a or (100, 200 points) (V)
VaR Statistics: Summary Sen Statistics: Summary

a. Annual average VaR (V) FRR. 48,305 a 1(iii) (A) a. Annual ave. Sen over the year (V)
b. Minimum VaR over the year (V) FRR. 48,305 a 1(iii) (A) b. Minimum Sen over the year (V) IFRS 7 B19a + B19b
C. maximum VaR over the year (V) FRR. 48,305 a 1(iii) (A) c. Maximum Sen over the year (V)
d. Year-end VaR (M) IFRS: B20, FRR. 48,305 a 1(iii) (A) d. Period indication for ave., max., and
min. Sen (M)
e. Diversication effect (V) FRR. 48: 305(a): 1 E e. Individual exposure in the risk type
(e.g., all currencies) (V)
VaR Inter-temporal Comparison Sen Inter-temporal Comparison

a. Summary information about the IFRS: B20, FRR. 48: 305 a 3 a (1) i a. Summary information about the
previous years' VaR (M) previous years' Sen (M)
VaR Back-Testing b & c. Change in Sen from % to point IFRS 7 C9: 36A & Holder 2002
or point to %** (V)
a. Number of exceptions (V) Basel II d. Justication for the change from %
to point or vice versa (V)
b. Explanation of exceptions (V) Basel II
Graphical Presentation of Daily VaR

a & b. Histogram of daily VaRs and/or Basel II
plot of daily VaRs* (V)
Quantitative Score: 14 points for each market risk exposure Quantitative Score: 13 points for each market risk exposure
Total VaR Score: 28 Points Total Sen Score: 27 points
M and V denote mandatory (based on IFRS 7) and voluntary disclosure, respectively.* Score of 1 if the histogram of daily VaR is disclosed or score of 2 if both
histogram and plot are disclosed.
** A score of 1 is applied if the rm changes the Sen values from percentage to a point system or vice versa, or a score of 2 is applied if both the percentage and
point changes are disclosed in the annual report (Al-Hadi et al., 2016).
References
Abraham, S., & Cox, P. (2007). Analysing the determinants of narrative risk information in UK FTSE 100 annual reports. The British Accounting Review, 39(3), 227248.
Abu-Nassar, M., & Rutherford, B. A. (1996). External users of financial reports in less developed countries: The case of Jordan. The British Accounting Review, 28(1),
7387.
Admati, A. R., & Pfleiderer, P. (2000). Forcing firms to talk: Financial disclosure regulation and externalities. The Review of Financial Studies, 13(3), 479519.
Aivazian, V. A., Ge, Y., & Qiu, J. (2005). The impact of leverage on firm investment: Canadian evidence. Journal of Corporate Finance, 11(1), 277291.
Al-Hadi, A., Hasan, M. M., & Habib, A. (2016). Risk committee, firm life cycle, and market risk disclosures. Corporate Governance: An International Review, 24(2),
145170.
Al-Hadi, A., Taylor, G., & Hossain, M. (2015). Disaggregation, auditor conservatism and implied cost of equity capital: An international evidence from the GCC. Journal of
Multinational Financial Management, 29, 6698.
Al-Hassan, A., Oulidi, N., & Khamis, M. (2010). The GCC banking sector: Topography and analysis. IMF working papers (pp. 145).
Ali, A., Chen, T. -Y., & Radhakrishnan, S. (2007). Corporate disclosures by family firms. Journal of Accounting and Economics, 44(1), 238286.
Aljifri, K., & Hussainey, K. (2007). The determinants of forward-looking information in annual reports of UAE companies. Managerial Auditing Journal, 22(9), 881894.
Al-Rumaihi, M. (1986). Beyond oil: Unity and development in the Gulf. London: Al-Saqi Books.
Al-Shammari, B., Brown, P., & Tarca, A. (2008). An investigation of compliance with international accounting standards by listed companies in the Gulf Co-Operation
Council member states. The International Journal of Accounting, 43(4), 425447.
Al-Yahyaee, K. H., Pham, T. M., & Walter, T. S. (2011). The information content of cash dividend announcements in a unique environment. Journal of Banking & Finance,
35(3), 606612.
Amenta, E. (2000). Bold relief: Institutional politics and the origins of modern American social policy. Princeton, NJ: Princeton University Press.
Amore, M. D., & Bennedsen, M. (2013). The value of local political connections in a low-corruption environment. Journal of Financial Economics, 110(2), 387402.
Anderson, R. C., & Reeb, D. M. (2003). Founding-family ownership and firm performance: Evidence from the S&P 500. The Journal of Finance, 58(3), 13011327.
Andres, P. d., & Vallelado, E. (2008). Corporate governance in banking: The role of the board of directors. Journal of Banking & Finance, 32(12), 25702580.
Atiyyah, H. S. (1992). Research note: Research in Arab countries, published in Arabic. Organization Studies, 13(1), 105110.
Barclay, M. J., & Holderness, C. G. (1989). Private benefits from control of public corporations. Journal of Financial Economics, 25(2), 371395.
Beattie, V., McInnes, B., & Fearnley, S. (2004). A methodology for analysing and evaluating narratives in annual reports: A comprehensive descriptive profile and met-
rics for disclosure quality attributes. Accounting Forum, 28(3), 205236.
Boot, A. W. A., & Thakor, A. V. (2001). The many faces of information disclosure. The Review of Financial Studies, 14(4), 10211057.
Boubakri, N., Cosset, J. C., & Saffar, W. (2012a). The impact of political connections on firm's operating performance and financing decisions. Journal of Financial
Research, 35(3), 397423.
Boubakri, N., Guedhami, O., Mishra, D., & Saffar, W. (2012b). Political connections and the cost of equity capital. Journal of Corporate Finance, 18(3), 541559.
Bradbury, M. E. (1990). The incentives for voluntary audit committee formation. Journal of Accounting and Public Policy, 9(1), 1936.
Bunkanwanicha, P., & Wiwattanakantang, Y. (2009). Big business owners in politics. Review of Financial Studies, 22(6), 21332168.
Cameron, A. C., & Trivedi, P. K. (2010). Microeconometrics using Stata. College Station, TX: Stata Press.
Certo, S. T. (2003). Influencing initial public offering investors with prestige: Signaling with board structures. Academy of Management Review, 28(3), 432446.
CFA (2013). User perspective on financial instrument risk disclosure underinternational financial reporting standards derivatives and hedging activities disclosures.
Charlottesville, VA: CFA Institute.
Chaney, P. K., Faccio, M., & Parsley, D. (2011). The quality of accounting information in politically connected firms. Journal of Accounting and Economics, 51(1), 5876.
Chen, F., Hope, O., Li, O., & Wang, X. (2011). Financial reporting quality and investment efficiency of private firms in emerging markets. The Accounting Review, 86(4),
12551288.
Cheng, E., & Courtenay, S. M. (2006). Board composition, regulatory regime and voluntary disclosure. The International Journal of Accounting, 41(3), 262289.
Claessens, S., Feijen, E., & Laeven, L. (2008). Political connections and preferential access to finance: The role of campaign contributions. Journal of Financial Economics,
88(3), 554580.
Crystal, J. (1995). Oil and politics in the Gulf: Rulers and merchants in Kuwait and Qatar. Cambridge, U.K.: Cambridge University Press.
Dechow, P. M., & Dichev, I. D. (2002). The quality of accruals and earnings: The role of accrual estimation errors. The Accounting Review, 77(s-1), 3559.
Deumes, R., & Knechel, W. R. (2008). Economic incentives for voluntary reporting on internal risk management and control systems. Auditing: A Journal of Practice &
Theory, 2(1), 3566.
Ding, S., Jia, C., Wu, Z., & Zhang, X. (2014). Executive political connections and firm performance: Comparative evidence from privately-controlled and state-owned
enterprises. International Review of Financial Analysis, 36, 153167.
Dobler, M. (2008). Incentives for risk reporting: A discretionary disclosure and cheap talk approach. The International Journal of Accounting, 43(2), 184206.
Einhorn, E. (2005). The nature of the interaction between mandatory and voluntary disclosures. Journal of Accounting Research, 43(4), 593621.
Elshandidy, T., & Neri, L. (2015). Corporate governance, risk disclosure practices, and market liquidity: Comparative evidence from the UK and Italy. Corporate
Governance: An International Review, 23(4), 331356.
Elshandidy, T., Fraser, I., & Hussainey, K. (2013). Aggregated, voluntary, and mandatory risk disclosure incentives: Evidence from UK FTSE all-share companies.
International Review of Financial Analysis, 30, 320333.
Emanuel, D., Wong, J., & Wong, N. (2003). Efficient contracting and accounting. Accounting and Finance, 43, 149166.
Faccio, M. (2006). Politically connected firms. The American Economic Review, 96(1), 369386.
Faccio, M., Masulis, R. W., & McConnell, J. (2006). Political connections and corporate bailouts. The Journal of Finance, 61(6), 25972635.
Fan, J. P., Wong, T. J., & Zhang, T. (2007). Politically connected CEOs, corporate governance, and post-IPO performance of China's newly partially privatized firms. Journal
of Financial Economics, 84(2), 330357.
Francis, J. R., Khurana, I. K., & Pereira, R. (2005). Disclosure incentives and effects on cost of capital around the world. The Accounting Review, 80(4), 11251162.
Gilson, R. J., & Gordon, J. N. (2003). Controlling controlling shareholders. University of Pennsylvania Law Review, 152, 785843.
Guedhami, O., Pittman, J. A., & Saffar, W. (2014). Auditor choice in politically connected firms. Journal of Accounting Research, 52(1), 107162.
Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1995). Multivariate data analysis with readings. New Jersey: Englewood.
Halawi, A., & Davidson, B. (2008). Power matters: A survey of GCC boards (TNI Market Insight). Retrieved from the Hawkamah Institute for Corporate Governance
website: https://hawkamah.org/wp-content/uploads/2014/10/Powermatters.pdf.
Hammoudeh, S., & Choi, K. (2006). Behavior of GCC stock markets and impacts of US oil and financial markets. Research in International Business and Finance, 20(1),
2244.
Hausman, J. A. (1978). Specification tests in econometrics. Econometrica: Journal of the Econometric Society, 46(6), 12511271.
Healy, P. M., & Palepu, K. G. (2001). Information asymmetry, corporate disclosure, and the capital markets: A review of the empirical disclosure literature. Journal of
Accounting and Economics, 31(1), 405440.
Hertog, S. (2012). How the GCC did it: Formal and informal governance of successful public enterprise in the Gulf Co-operation Council countries. Towards new ar-
rangements for state ownership in the Middle East and North Africa. Paris: OECD Publishing.
Hirtle, B. (2007). Public disclosure, risk, and performance at bank holding companies. Federal Reerve Bank of New York staff report 293 (New York, NY. Available at https://
www.newyorkfed.org/medialibrary/media/research/staff_reports/sr293.pdf).
Hodder, L., Koonce, L., & McAnally, M. L. (2001). SEC market risk disclosures: Implications for judgment and decision making. Accounting Horizons, 15(1), 4970.
Hooks, J., & Staden, C. J. (2011). Evaluating environmental disclosures: The relationship between quality and extent measures. The British Accounting Review, 43(3),
200213.
IFC & Hawkamah (2008). A corporate governance survey of listed companies and banks across Middle East and North Africa. Pennsylvania: International Finance Corpo-
ration and Hawkamah.
Jaggi, B., Leung, S., & Gul, F. (2009). Family control, board independence, and earnings management: Evidence based on Hong Kong firms. Journal of Accounting and
Public Policy, 28(4), 281300.
Jayaraman, S. (2008). Earnings volatility, cash flow volatility, and informed trading. Journal of Accounting Research, 46(4), 809851.
Jensen, M., & Meckling, W. (1976). Theory of the firm: Managerial behavior, agency costs, and ownership structure. Journal of Financial Economics, 3(4), 305360.
John, K., Litov, L., & Yeung, B. (2008). Corporate governance and risk-taking. The Journal of Finance, 63(4), 16791728.
Jorgensen, B. N., & Kirschenheiter, M. T. (2003). Discretionary risk disclosures. The Accounting Review, 78(2), 449469.
Jorion, P. (2002). How informative are value-at-risk disclosures? The Accounting Review, 77(4), 911931.
Joshi, P. L., & Wakil, A. (2004). A study of the audit committees' functioning in Bahrain: Empirical findings. Managerial Auditing Journal, 19(7), 832858.
Kaufmann, D., Kraay, A., & Mastruzzi, M. (2009). Governance matters VIII: Aggregate and individual governance indicators, 19962008. World bank policy research
working paper no.4978 (Available at http://www.wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2009/06/29/000158349_20090629095443/
Rendered/PDF/WPS4978.pdf).
Khanna, T., & Palepu, K. (2000). Is group affiliation profitable in emerging markets? An analysis of diversified Indian business groups. Journal of Finance, 55, 867891.
Khanna, T., & Rivkin, J. (2001). Estimating the performance effects of business groups in emerging markets. Strategic Management Journal, 22, 4574.
Koch, A. S. (2002). Financial distress and the credibility of management earnings forecasts. Available at http://marriottschool.net/emp/tc259/accframework/files/
2002%20Koch.pdf
Kshetri, N., & Ajami, R. (2008). Institutional reforms in the Gulf Cooperation Council economies: A conceptual framework. Journal of International Management, 14(3),
300318.
Lagoarde-Segot, T., & Lucey, B. M. (2007). International portfolio diversification: Is there a role for the Middle East and North Africa? Journal of Multinational Financial
Management, 17(5), 401416.
Larcker, D. F., & Rusticus, T. O. (2010). On the use of instrumental variables in accounting research. Journal of Accounting and Economics, 49(3), 186205.
Leuz, C., & Oberholzer-Gee, F. (2006). Political relationships, global financing, and corporate transparency: Evidence from Indonesia. Journal of Financial Economics,
81(2), 411439.
Li, S. (2010). Does mandatory adoption of international financial reporting standards in the European Union reduce the cost of equity capital? The Accounting Review,
85(2), 607636.
Lin, C. M., Owens, W. L., & Owers, J. E. (2010). The association between market risk disclosure reporting and firm risk: The impact of SEC FRR no. 48. Journal of Applied
Business Research, 26(4).
Linsley, P. M., & Shrives, P. J. (2006). Risk reporting: A study of risk disclosures in the annual reports of UK companies. The British Accounting Review, 38(4), 387404.
Marshall, A. P., & Weetman, P. (2002). Information asymmetry in disclosure of foreign exchange risk management: Can regulation be effective? Journal of Economics
and Business, 54(1), 3153.
Mazaheri, N. (2013). The Saudi monarchy and economic familism in an era of business environment reforms. Business & Politics, 15(3), 295321.
Miihkinen, A. (2012). What drives quality of firm risk disclosure? The impact of a national disclosure standard and reporting incentives under IFRS. The International
Journal of Accounting, 47(4), 437468.
Mina, W. (2007). The location determinants of FDI in the GCC countries. Journal of Multinational Financial Management, 17(4), 336348.
Morck, R. K., Stangeland, D. A., & Yeung, B. (1998). Inherited wealth, corporate control, and economic growth: The Canadian disease? NBER working paper no. 6814.
Cambridge, MA: National Bureau of Economic Research.
Myers, S., & Majluf, N. (1984). Corporate financing and investment decisions when firms have information that investors do not have. Journal of Financial Economics,
13(2), 187221.
Nier, E., & Baumann, U. (2006). Market discipline, disclosure and moral hazard in banking. Journal of Financial Intermediation, 15(3), 332361.
Perignon, C., & Smith, D. R. (2010a). The level and quality of value-at-risk disclosure by commercial banks. Journal of Banking & Finance, 34(2), 362377.
Prignon, C., & Smith, D. R. (2010b). Diversification and value-at-risk. Journal of Banking & Finance, 34(1), 5566.
Piotroski, J. D., Wong, T. J., & Zhang, T. (2010). Political incentives to suppress negative financial information: Evidence from state-controlled Chinese firms. (Unpub-
lished working paper). Stanford University, Chinese University of Hong Kong, and City University of Hong Kong.
Polsiri, P., & Jiraporn, P. (2012). Political connections, ownership structure, and financial institution failure. Journal of Multinational Financial Management, 22(1), 3953.
Rajgopal, S. (1999). Early evidence on the informativeness of the SEC's market risk disclosures: The case of commodity price risk exposure of oil and gas producers. The
Accounting Review, 74(3), 251.
Schipper, K. (1989). Commentary on earnings management. Accounting Horizons, 3(4), 91102.
Sharar, Z. A. (2011). Corporate governance in Quatar: A comparative analysis. Corporate Goveranance eJournal. Qatar University (Available at: http://epublications.
bond.edu.au/cgi/viewcontent.cgi?article=1022&context=cgej).
Shleifer, A., & Vishny, R. W. (1997). A survey of corporate governance. The Journal of Finance, 52(2), 737783.
Sidani, Y., & Al Ariss, A. (2013). Institutional and corporate drivers of global talent management: Evidence from the Arab Gulf region. Journal of World Business, 49(2),
215224.
Stock, J. H., & Yogo, M. (2005). Testing for weak instruments in linear IV regression. Identification and inference for econometric models: Essays in honor of Thomas
Rothenberg (Avaiable at : http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1734933).
Stulz, R. M. (2005). The limits of financial globalization. The Journal of Finance, 60(4), 15951638.
(Power matters: A survey of GCC boards (TNI Market Insight). Retrieved from the Hawkamah Institute for Corporate Governance website: https://hawkamah.org/wp-
content/uploads/2014/10/Powermatters.pdf) Vandemaele, S., Vergauwen, P., & Michiels, A. (2009). Management risk reporting practices and their determinants.
Avaiable at: https://uhdspace.uhasselt.be/dspace/handle/1942/9392.
Wan-Hussin, W. N. (2009). The impact of family-firm structure and board composition on corporate transparency: Evidence based on segment disclosures in Malaysia.
The International Journal of Accounting, 44(4), 313333.
Xu, N., Xu, X., & Yuan, Q. (2013). Political connections, financing friction, and corporate investment: Evidence from Chinese listed family firms. European Financial Man-
agement, 19(4), 675702.
Xu, N., Yuan, Q., Jiang, X., & Chan, K. C. (2015). Founder's political connections, second generation involvement, and family firm performance: Evidence from China.
Journal of Corporate Finance, 33, 243259.

Alhadi 2016

Diunggah oleh

Informasi Dokumen

Judul Asli

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

Alhadi 2016

Diunggah oleh

Hak Cipta:

Format Tersedia

ACCOUN-00763; No of Pages 21

The International Journal of Accounting xxx (2016) xxxxxx

Contents lists available at ScienceDirect

The International Journal of Accounting

2. Theory and hypothesis development

2.1. Market risk disclosures

2.2. Political connections of ruling family members on the board

2.3. Political connections and market risk disclosures

3.1. Sample and data

3.2. Dependent variables

Variables Denition and measurement

Dependent variable: disclosure proxies

Political connection proxies

3.2.1. Component factor analysis

MRDFACTOR Composite Highest Score of Eigenvalue 1

Panel A: Sample selection

Panel B: Sample distribution

Bank Financial Insurance Investment

3.2.2. Extent of market risk disclosures

3.2.3. Quality of market risk disclosures

3.3. Independent variables

3.4. Control variables

3.5. Empirical model

3.5.1. Ordinary least squares (OLS)

MRDFACTOR=EMRD=QMRDi;t 0 1 RoyalF Dir i;t 2 CG Indexi;t 3 LNMVALi;t 4 Leveragei;t a5 ROEi;t

4. Empirical results and discussion

4.1. Descriptive statistics

4.2. Univariate t-test

Panel A: Political connection based on country

Country Mean Sum

Panel B: Descriptive statistics

Variable N Mean S.D. Min Mdn Max

MRDFACTOR 667 0.760 0.27 0.170 0.730 3.800

Bank 0.689 0.576 0.112 1.50

RoyalF_Dir Mean Difference t-statistic

Bank 0.712 0.781 0.068 2.37**

4.3. Correlation analysis

4.4. Regression analysis

Political connections and market risk disclosures.

A. Al-Hadi et al. / The International Journal of Accounting xxx (2016) xxxxxx

MRDFACTOR (OLS) EMRD (OLS) QMRD (OLS)

Intercept 1.134*** 0.204*** 2.509***

5. Robustness and additional tests

5.1. Robustness test: two-stage least squares (2SLS)

MRDFACTOR (2SLS-2nd stage) EMRD (2SLS-2nd stage) QMRD (2SLS-2nd stage)

Intercept 1.075*** 0.159* 1.924*

MRDFACTOR EMRD QMRD

FamilyDir =1 FamilyDir =0 GovDir = 1 GovDir = 0 FamilyDir =1 FamilyDir =0 GovDir =1 GovDir =0

Intercept 1.124*** 0.951*** 1.759*** 0.806*** 0.066 0.368*** 0.287* 0.297***

5.2. Additional tests

5.2.1. Alternative measures of political connection

5.2.2. Government and family board directorships

5.2.3. Discretionary accruals

5.2.4. Government and royal family board directorships

EMRD QMRD EMRD QMRD

5.2.5. Mandatory versus discretionary disclosures

5.2.6. Market risk, nancial distress, and ruling family directorships

Mandatory Index+ Voluntary Index#

Intercept 0.153*** 0.095

Intercept 1.1361 0.0951

Intercept 4.0773*** 0.3268***

Appendix A. Market Risk Disclosures Index

Qualitative Score: 14 points Qualitative Score: 14 points

Intercept 1.134* 0.204* 2.509***

Intercept 1.124* 0.951* 1.759* 0.806* 0.066 0.368*** 0.287* 0.297***

Intercept 4.0773* 0.3268*