The Effects Of Ownership And Regulation On Bank Earnings Quality: An Investigation Of The Conventional And Islamic Banks In Mena Region a Higher Institute of Management, University of Tunis, Tunisia. khawlabourkhis@hotmail.fr b Al Itqan Islamic Finance, Ariana, Tunisia. najarfinanceislamique@yahoo.fr Abstract— The study aims to examine the relationship between shareholding ownership structures, national institutional factors and earnings quality of banks across MENA countries (Middle East and North Africa). Using four distinct earnings quality measures that detect different aspects of earnings properties on a sample of 158 banks (44 Islamic banks and 114 conventional banks), the study finds that all four measures of earnings quality are higher for listed and widely held banks; and, that state-owned banks have less persistent, less predictable, and less managed earnings. Moreover, Islamic banks in MENA countries appear to have significantly higher quality of earnings than their conventional counterparts in terms of earnings persistence, cash- flows predictability and income smoothing using loan loss provisions. The study shows also that tighter supervision improves earnings reporting quality by reducing earnings management practices even in the presence of large shareholding. Keywords- ownership structure; bank regulation; earnings quality; earnings management; Islamic banks I. INTRODUCTION Financial reports constitute the primary information source for stakeholders of both banking and non-banking firms. Financial reports including balance sheet, income and cash flow statements provide information regarding bank performance, solvency and soundness. Hence, reporting financial accounting information need to be with high quality to insure reliability, accuracy and informativeness. Incomes represent the most important accounting information that investors, managers, directors and regulators rely on in their decision making process. Earnings are the primary information source for investors rather than any other performance indicators such as dividend and cash flows (Francis et al., 2004)[36]. Consequently, reporting quality of earnings is crucial for the well-functioning markets. Investors, analysts and policy makers require credible accounting information to assess the real firm’s economic performance and take subsequently optimal decisions. Penman and Zhang (2002) [57] consider that earnings with high quality if, before extraordinary items that are freely identified on the income statement, it is a good indicator of future earnings. Dechow et al. (2010) [26], further, indicate earnings with high quality if it provides more information about the features of a firm’s financial performance that are relevant to a specific decision made by a specific decision-maker. More broadly, earnings reporting quality refers to the ability of accounting earnings to signal future firm’s earnings and cash flows (Francis et al., 2004; Chen et al., 2007; Giao and Raposo, 2011; Demerjian et al., 2012) [36, 21, 37 & 28]. Mainly, earnings are considered as high quality if they are predictable and easy to forecast. Yet, fraudulent reporting is absolutely low quality. Kanagaretnam et al. (2014) [44] examine earnings quality from two different perspectives: an informative and an opportunistic earnings management perspective. The informative perspective indicates that earnings quality is improved when managers report less noisy earnings by taking reporting actions that reveal accurate and precise information about firm’s real performance. However, earnings quality decreases if managers behave opportunistically and intervene deliberately in the earnings reporting process by altering the firm economic performance to mislead outsiders and/ or increase their own welfare at expense of investors. Thereby, earnings management has a lot in common with earnings quality. Highly managed earnings are definitely low quality. Literature on financial reporting quality and banking industry shows that banking institutions often manage their accounting earnings through the use of loan provisions or the security gains and losses for several motives: income smoothing, regulatory incentives, signaling purposes and tax payment (Beaver and Engel, 1996; Ahmed et al., 1999; Anandarajan et al., 2007) [12, 5 & 9]. Prior researches demonstrate that bank earnings management is a worldwide phenomenon (Shen and Chih, 2005) [61] but its level depends on various internal and external factors. Banks risk level and governance mechanisms are the most widely analyzed (Cornett et al., 2009; Wan Mohammad et al., 2011; Abaoub et al., 2013) [24, 67 &1]. For instance, it is shown that bank executives’ compensation contracts increase managerial incentives to manage earnings (Uygur, 2013) [66]. However, the internal control systems increase accounting quality and limit discretionary behavior (Altamuro and Beatty, Khawla Bourkhis a and Mohamed Wajdi Najar b mailto:khawlabourkhis@hotmail.fr mailto:najarfinanceislamique@yahoo.fr EJIF – European Journal of Islamic Finance No 6, Feb (2017) http://www.ojs.unito.it/index.php/EJIF ISSN 2421-2172 2 2010) [7]. Further, Bouvatier et al. (2014) [16] indicate that European banks with higher level of ownership concentration display higher intensity of income smoothing through loan loss provisions (LLP). However, this negative association appears attenuated among countries with stronger supervisory regimes or higher external audit quality. Cross countries analyses such as Fonseca and Gonzalez (2008) [35] and Kanagaretnam et al. (2014) [44] have highlighted the role of institutional environment and supervisory regimes in enhancing bank earnings quality by reducing earnings management. Auditing quality, changes in accounting standards and bank listing status are also revealed as important determinants of earnings management in the banking industry. This study aims to examine whether differences in ownership status could explain variations in the quality of bank earnings across MENA countries. Our research question is therefore: Is there a relationship between MENA bank ownership structure and the content information of earnings? We suppose that MENA region constitutes a favorable field of research for three key reasons: Firstly, MENA banking institutions are well characterized by highly concentrated ownership structure with the predominance of government ownership (Kobeissi and Sun, 2010; Farazi et al., 2011). [47 & 34]. However the limited existing literature does not provide consistent findings on the implications of this distinctive ownership structure. There is no study in our knowledge examining MENA bank earnings reporting quality and its association with their ownership structures. Secondly and despite the noted higher ownership concentration that complicates the governance of banks, MENA legal institutions do not sufficiently protect minority shareholders’ interests against expropriation risk. MENA countries suffer from the weak enforcement of shareholders rights (Naciri, 2008) [56]. Nonetheless, several studies such as Caprio et al. (2007), Shehzad et al. (2010), Haw et al. (2010) and Busta et al. (2014) [19, 60, 39 & 18] prove that legal environment impacts significantly conflicts of interests between controlling shareholders and minority shareholders. Finally, banking system in MENA region is characterized by the co-existence of Islamic banks and conventional banks side-by-side. Islamic banking assets1 in MENA countries are about 1,197.9 billion dollars in 2014, it accounts for more than 20 percent of banking system assets in 10 countries. MENA region constitutes so an adequate sample for a comparative analysis between the two groups of banks. The principal objectives of the present research study are therefore described as follows: Firstly, the analysis aims to assess the quality of bank’s reported earnings and measure the extent of bank earnings management in the MENA region. Secondly, we intend to determine whether and to what extent shareholding ownership concentration matters in bank earnings quality. Thirdly, the study tries to explore whether certain types 1 Islamic Financial Services Industry Stability Report, May 2015. of majority shareholder are particularly conductive to lower quality of earnings of MENA banks. Fourthly, the study will determine to what extent Islamic religious values affect earnings reporting quality. Lastly, we will highlight the impact of regulation and banking supervision in emerging economies on bank earnings quality. We employ four distinct earnings quality measures that detect different aspects of earnings properties (earnings persistence, ability to predict cash flows, income smoothing through loan loss provisions and small positive net incomes on a sample of 158 banks (44 Islamic banks and 114 conventional banks) from 15 MENA countries observed over the period (2000-2013). To control the MENA institutional environment and evaluate its role in enhancing the earnings reporting quality, we use three indicators: Investor Protection Index, Official Supervisory Power index, and Private Monitoring Index. Our empirical findings suggest that all four measures of earnings quality are higher for listed and widely held banks; and, that state-owned banks have less persistent, less predictable, and less managed earnings. The study shows also that tighter supervision improves earnings reporting quality by reducing earnings management practices even in the presence of large shareholding. Moreover, Islamic banks in MENA countries appear to have significantly higher quality of earnings than their conventional counterparts in terms of earnings persistence, cash flows predictability and income smoothing using loan loss provisions. The rest of the paper is organized as follows: Section 2 provides a discussion on relevant literature and develops research hypotheses. Data and research design are explained in section 3. Section 4 presents empirical findings. Section 5 concludes the paper. II. LITERATURE REVIEW AND HYPOTHESES DEVELOPMENT A. Share ownership structure and bank earnings quality It is highly argued in the literature that corporate share ownership structure has a significant impact on the agency costs inherent in the stockholder-manager relationship, in the extent that it influences the efficiency of the implemented monitoring mechanisms. Both agency theory and empirical evidence suggest that owners’ ability and incentives to carry out monitoring effort and mitigate agency problems vary with the control ownership structure. Prior studies, for instance, show that single investors with small ownership stakes have little incentive to control the firms. However, blocked shareholders with large ownership stakes gain control power and have more incentive to monitor management activity (Schleifer and Vichny, 1986) [63]. Additionally, managerial ownership is deemed as relatively useful in aligning management and shareholders’ interests and reducing hence agency conflicts (Morck et al., 1988) [55]. Two competing views have been suggested by the literature regarding ownership concentration and earnings reporting quality association. On the one hand, http://www.ojs.unito.it/index.php/EJIF ownership concentration alleviates agency costs between managers and shareholders (Jensen and Meckling, 1976; Shleifer and Vishny, 1986) [41 & 63]. In fact, being large owner with a considerable number of shares creates strong incentives and gives more power to oversee firm management and evaluate executives in order to assure that shareholders’ interests are protected. In addition, Demsetz and Lehen (1985) [29] deem that majority ownership is an effective mechanism to mitigate managerial expropriation. One form of this managerial expropriation is represented in the lower financial and accounting reporting quality. Thereby, carrying on control over firm management definitely includes control over accounting information and reporting policies; and this is in order to reduce the scope of managerial opportunism. Large block holders therefore can force managers to adequately report financial information. However, managers of firms with widely dispersed ownership are assumed as in a better position to adopt discretionary accounting practices that serve their self-interests. In consonance with the monitoring hypothesis, many of the previous empirical studies reveal that the concentration of ownership leads to better quality of accounting numbers. For example, Warfield et al. (1995) and Alves (2012) [69 & 8] find that ownership concentration enhances the quality of reporting earnings by reducing the levels of earnings management. Similarly, Dechow et al. (1996); Yeo et al. (2002) and Jung and Kwon (2002) [27, 70 & 42] point out that block holders of share improves the credibility, reliability and informativeness of financial information. On the other hand, block holder ownership can generate unluckily an entrenchment behavior against minority shareholders (Jensen and Meckling, 1976; Shleifer and Vishny, 1986) [41 & 63]. When ownership control is high enough, largest shareholders have the incentive to expropriate small shareholders’ wealth. They might possibly put pressure on managers to engage in earnings management practices in order to expropriate firm resources at the expense of outside investors. In the East-Asian context, Fan and Wong (2002) [32] reveal that earnings informativeness, proxied by the accounting earnings-stock returns relationship, decreases with ownership concentration. The study shows that controlling owners report accounting information for self-interested purposes, causing hence the reported earnings to lose credibility to outside investors. In addition, Leuz et al. (2003) [50], in a sample of 8000 firms from 31 countries, indicate that managers and controlling shareholders can use their control over the firm to benefit themselves at the expense of other stakeholders. For NYSE firms, Zhong et al. (2007) [71] too reveal a positive association between block-holder ownership and discretionary accruals. Further, Kung et al. (2010) [48] find in the sample of listed Chinese companies a negative relationship between block ownership and earnings conservatism. Drawing on the above discussion and on the recent empirical evidence from the banking sector (Isenmila and Elijah, 2012; Tsai and Hsieh, 2013; and Bouvatier et al., 2014) [40, 65 & 16] showing that banks with higher ownership concentration conduct more earnings management and income smoothing that lower the quality of earnings than banks with low ownership concentration; we postulate that: H1: Ownership concentration lowers earnings quality of MENA banks. Religious ethical values are deemed as one of the monitoring mechanisms considered in the literature in limiting opportunistic and unethical corporate behavior (Dyreng et al., 2012) [31]. For instance, Kennedy and Lawton (1998) , Conroy and Emerson (2004) and Longenecker et al. (2004) [46, 23 & 53] have emphasized the role of religion in constraining unethical practices in the business organizations. Lewis (2001) [52] argues that religion plays an important and constructive role in guiding and controlling human behavior, it provides values of truthfulness, honesty, morality, justice and accountability. Religious beliefs and codes are in fact the source of ethical and moral behavior. In that way, religious identity of the organization is appeared to mitigate opportunistic behavior among managers and enhance as a result financial information reliability and the integrity of the financial reporting process (Abdul Rahman, 2012) [3]. Prior studies such as McGuire et al. (2012) [54] assert that religion-influenced firms are less likely to engage in financial reporting irregularities. They find a negative association between religiosity and abnormal accruals. Earlier literature on Islamic finance and banking suggests that Islamic banking firms are subject to an additional layer of governance in the form of Shariah governance. Abdel Karim (1990) [2] argues that the Shariah Supervisory Board could be viewed as similar to the independent company auditors in limiting discretionary behavior. Further, Quttainah et al. (2013) [58] find that Islamic banks employ less earnings management than their conventional peers. In contrary to the above conclusions, other researches such as (Zoubi and El Ghazali, 2007; Taktak et al., 2010; Ben Othman and Mersni, 2014) [72, 64 & 13] indicate that earning management is not much different between Islamic and conventional banking institutions. We postulate therefore that: H2: Islamic banks display higher earnings quality than their conventional counterparts. Empirical research examining the role of government shareholding in financial reporting quality is scant and, to the extent available, it is inconclusive. Starting from the role of government ownership in the economic and financial sectors (social, political and agency views) diverse relationships are expected. Under the development or called also the social view of state ownership, government intervene in the financial markets to promote macroeconomic growth and cure market financial failures. Government is assumed as powerful and has incentives enough to safeguard stakeholders’ interests and improve consequently the general welfare. Government is accountable to the public to monitor public firms since the latter are authorized to use public funds. Then, monitoring state- owned enterprises management and more specifically accounting information quality is deemed as crucial to avoid robbery and corruption. Thereby, state-owned companies have EJIF – European Journal of Islamic Finance No 6, Feb (2017) http://www.ojs.unito.it/index.php/EJIF ISSN 2421-2172 4 to exhibit good financial reporting quality. Consistently with the development hypothesis, Abdul Rahman (2012) [3] find that Malaysian firms with high concentrated government ownership have higher accounting conservatism and lower degree of discretionary accruals. In the Chinese context moreover, Wang and Yung (2011) [68] note that Chinese state-owned firms have better accounting earnings quality (better accruals quality and lower levels of abnormal accruals) than privately owned firms. Similarly, Bo and Wu (2009) [15] find that level of income- increasing earnings management is lower in state-owned firms than in privately-owned firms. Conversely, the political view of state ownership supposes that state-owned enterprises are created to satisfy personal and political objectives rather than to maximize social welfare. Shleifer (1998) [62] argues that state shareholding has detrimental effect on corporate performance. Government acquires ownership in highly sensitive sectors to ultimately serve its political agendas. Unlike the development theory of government shareholding, the political view assumes that politicians and bureaucrats are self-interested individuals and pursue their own personal objectives at the expense of the State (Sapienza, 2004) [59]. They often seek to stay in power and maintain the perquisites of their positions (Shleifer, 1998) [62]. Thereby, monitoring decisions or expropriating corporate resources for the benefit of supporters is mainly influenced by various political and individual concerns. For example, politicians mostly in developing countries abuse their power to transfer some corporate resources into the hands of their supporters (Kung et al, 2010) [48]. Furthermore and in order to gain public trust and support, executives in state-owned companies can manage financial reports to mask large losses. In China, Chen and Yuan (2004), Ding et al. (2007) and Aharony et al. (2010) [20, 30 & 4] find that government ownership is more associated with earnings management practices than private ownership. No consensus is provided in the banking literature about the impact of the largest owner identity in the financial reporting process. Bouvatier et al. (2014) [16] show that income smoothing via LLP is independent to whether the ultimate owner is an institution, a family or an industrial firm. Following Tsai and Hsieh (2013) [65] who suggest that large government shareholding in the bank may limit earnings increasing management, we postulate that: H3: State ownership in MENA banks is associated with higher earnings quality. B. Bank earnings quality and the regulatory environment Prior literature on banking firms displays international differences in earnings quality, suggesting, therefore that the latter could be affected by differences in the institutional environment and banking regulation. Investor protection, regulation, supervision and financial development have shown 2 We refer to the World Bank Group Doing Business Reports to identify MENA countries (Algeria, Bahrain, Egypt, Iran, Iraq, Jordan, Kuwait, Lebanon, Libya, Malta, Morocco, Oman, Qatar, Saudi Arabia, Palestinian Authority, Syria, Tunisia, United Arab Emirates great effect on earnings management and income smoothing behavior. Cross-countries studies such as Biurrun (2010) and Kanagaretnam et al. (2014) [14 & 44] indicate negative relationship between banking regulation and supervision, and earnings management. Fonseca and Gonzalez (2008) [35] who find significant influence of investor protection and legal enforcement in reducing incentives to smooth income, argue that banks with higher incentives to shift risk in the event if bank run, have higher incentive too to manage earnings to hide their risk-shifting. Since efficient bank regulation and supervision proves to be in limiting bank risk, then it will prove to be in limiting income smoothing. In that way, bank regulators and supervisors who have greater power to reduce bank risk- taking through monitoring and disciplining management will obviously reduce banks’ incentives to smooth earnings. Based on what precede, we suggest that banking regulation matters and works to constrain banks’ earnings management. We postulate therefore that: H4: Stronger bank regulation improves MENA bank earnings quality. III. DATA AND METHODOLOGY A. Data Description For the purpose of the current investigation, we build a multivariate database on banks’ individual ownership information, bank-specific variables and some country characteristics. We collect data of Islamic and conventional banks operating in 15 MENA countries2 namely (Bahrain, Egypt, Iran, Jordan, Kuwait, Lebanon, Malta, Morocco, Oman, Palestine, Qatar, Saudi Arabia, Tunisia, UAE and Yemen) during the period 2000 to 2013. We retrieve bank accounting data from BvD BankScope and macroeconomic data from the World Bank. For the time period covered by this study, we can collect full data for only 158 banks (see Table 1). The regression analyses are conducted by using STATA 11.1. B. Ownership Measures Bank ownership database is particularly difficult to construct. Initially, we rely on BankScope, Union of Arab Banks, Zonebourse, GulfBase and Zawya databases to define the ultimate bank owners. These sources provide information for only one year while we have to detect variation of ownership structure over the time period of this study (2000-2013). Hence, and to compile ownership data, we use earlier bank annual reports and/or national institutions publications such as central banks, stock exchange and ministries of finance. In order to fulfill the gaps in our database, we tape into the online archives of business magazines and pull up articles about previous bank merger and acquisition event3 that occurred in the MENA region during the period of the study. Collecting data on ownership constitutes a fundamental contribution of this and Yemen). Nonetheless, Algeria, Libya, Syria and Iraq are excluded from our sample due to missing data. 3 About 6 bank mergers took place in our sample of country-years. http://www.ojs.unito.it/index.php/EJIF research. We have hence four variables Widely: is a dummy variable that equals one if no legal entity owns 10 % or more of the voting rights, and zero otherwise. Control Right: equals the fraction of the direct and indirect bank’s voting rights, if any, owned by its controlling shareholder. Control Right equals zero if the bank is widely held. Cash Flow Right: equals the fraction of the bank's cash-flow rights owned directly and indirectly by its controlling shareholder. Cash Flow Right equals zero if the bank is widely held. State: is a dummy variable that equals one if the state (or a foreign state) is the controlling shareholder, and zero otherwise. TABLE 1: DISTRIBUTION OF BANKS IN THE SAMPLE Number of banks Number of listed banks Number of Islamic banks Bahrain 16 10 8 Egypt 19 15 2 Iran 8 0 8 Jordan 12 11 3 Kuwait 11 10 4 Lebanon 19 5 0 Malta 4 2 0 Morocco 6 5 0 Oman 7 5 0 Palestine 2 1 1 Qatar 9 8 3 Saudi Arabia 6 5 4 Tunisia 15 10 1 UAE 20 17 7 Yemen 4 0 3 Total 158 104 44 C. Regulatory Index To provide information on the legal environment quality, we use three indicators:  Strength of Investor Protection Index: obtained from the World Bank Doing Business database. It measures the strength of minority shareholder protections 4 The index is based on 9 answers to yes or no questions from the survey: 1. Are auditors required to communicate directly to the supervisory agency any presumed involvement of bank directors or senior managers in illicit activities, fraud, or insider abuse? 2. Does the banking supervisor have the right to meet with the external auditors and discuss their report without the approval of the bank? 3. In cases where the supervisor identifies that the bank has received an inadequate audit, does the supervisor have the powers to take actions against the external auditor? 4. Do banks disclose to the supervisor off-balance sheet items? 5. Does the supervisory agency require banks to constitute provisions to cover actual or potential losses? 6. Does the supervisory agency require banks to reduce or suspend dividends to shareholders? 7. Does the supervisory agency require banks to reduce or suspend bonuses and other remuneration to bank directors and managers? 8. Does the supervisory agency have the powers to perform the following problem bank resolution activities? a) Declare insolvency b) Supersede shareholders' rights c) Remove and replace bank senior management and directors. 9. Can the supervisory authority force a bank to change its internal organizational structure? Yes/No responses to the previous questions are coded as 1/0. Responses to questions 7, 8(b) and 8(c) are multiplied by 2. Index= 1+2+3+4+5+6+(7*2)+8(a)+(8(b)*2)+(8(c)*2)+9 against directors’ misuse of corporate assets for personal gain. IPI ranges from 0 to 10, with higher values indicating more investor protection.  Official Supervisory Power Index: drawn from the World Bank’s Regulation and Supervision databases Cihák et al. (2012) [22] (Survey IV). The index measures the degree of official supervisory power in a country. It indicates the extent to which supervisory authorities have the power to take prompt actions to prevent, correct problems, and restructure and reorganize troubled banks4. Higher values imply greater power.  Private Monitoring Index: drawn from Cihák et al. (2012) [22] (Survey IV), it covers audit requirements, the extent to which banks have to be rated by international and domestic rating agencies, and whether and to what degree depositors are protected by an explicit deposit insurance scheme5. Higher values imply more private oversight. D. Testing for Earnings Quality The principal objectives of the current study are to assess earnings quality of banks across MENA countries; and examine its relation between share ownership structures and national institutional factors. To successfully perform them, several traditional proxies of earnings quality are required. Dechow et al. (2010) [26] define three categories of earnings quality proxies, namely properties of earnings, investor responsiveness to earnings and external indicators of earnings misstatements. Consistently with the object aimed and because of limited data availability and accessibility, we decided hence to focus only on the earnings properties as an indicator of earnings quality. Ahrens (2010) [6] identifies three essential properties of earnings that should be verified to qualify earnings reporting as a good quality: current earnings should represent firm current 5 The index is based on 12 answers to yes or no questions from the survey: 1. Is an audit by a professional external auditor required for all commercial banks in your jurisdiction? 2. Does the external auditor have to obtain a professional certification or pass a specific exam? 3. How many of the top ten banks (in terms of total domestic assets) are rated by domestic credit rating agencies? 4. How many of the top ten banks (in terms of total domestic assets) are rated by international credit rating agencies (e.g., Moody's, Standard and Poor)? 5. Does accrued, though unpaid, interest/principal enter the income statement while the loan is non-performing? 6. Does accrued, though unpaid, interest/principal enter the income statement while the loan is still performing? 7. Are bank directors legally liable if information disclosed is erroneous or misleading? 8. Do banks disclose to the public off-balance sheet items? 9. Do banks disclose to the supervisors governance and risk management framework? 10. Are bank regulators/supervisors required to make public formal enforcement actions, which include cease and desist orders and written agreements between a bank regulatory/supervisory body and a banking organization? 11. Is subordinated debt allowed as part of Tier 2 capital? 12. Is there an explicit deposit insurance protection scheme? Yes/No responses to the previous questions are coded as 1/0. For questions 3 and 4: 100%=1; otherwise 0. Index=1+2+3+4+5+6+7+8+9+10+11+12 EJIF – European Journal of Islamic Finance No 6, Feb (2017) http://www.ojs.unito.it/index.php/EJIF ISSN 2421-2172 6 performance, current earnings should be an indicator for future earnings, and finally current earnings should be free from earnings management. To better assess these properties, we resort henceforth to four widely used measures by the reputed literature Altamuro and Beatty (2010), Kanagaretnam et al. (2011, 2014) and Fang et al. (2014) [7, 43, 44 &33] namely earnings persistence, cash flows predictability, income smoothing through loan loss provisions, and small positive net income. a) Earnings Persistence (EQ1): is defined as the coefficient on last period earnings before taxes in a regression of current earnings before taxes on last earnings before taxes. A positive and significant coefficient α1 indicates earnings persistence. Higher coefficient implies more persistent earnings stream. i,j, t = i,j,t-1 variables of interest + control variables i,j,t-1 + e (1) i,j,t = Earnings before taxes of bank i in country j during year t scaled by total assets at the beginning of the year (i.e. at year t-1). e: Error term b) Cash-Flow Predictability (EQ2): measures earnings’ ability to predict current cash flows as the coefficient from a regression of current earnings before taxes and loan loss provisions on last period net income before taxes. A positive and significant coefficient signifies that earnings are able to predict future cash-flows. Pi,j,t = i,j,t-1 variables of interest + control variables i,j,t-1 + e (2) EBTPi,j,t = Earnings before taxes and loan loss provisions of bank i in country j during year t scaled by total assets at the beginning of the year (i.e. at year t-1). e: Error term c) Income Smoothing through Loan Loss Provisions (EQ3): measures whether bank managers use the LLP to smooth bank earnings. Consistent with Kanagaretnam et al. (2004), Taktak et al. (2010) and Bouvatier et al. (2014) [45, 64 & 16], we use an equation that explicitly models the non- discretionary portion of LLP which is expected to cover credit losses and exhibits a cyclical pattern. We use total loan (Loans), non-performing loans (NPL) and the GDP growth rate (GDP). The expected signs of the coefficients on these variables are as follows: an increase in total loans is likely to result in an increase in the LLP due to doubtful loans. An increase in nonperforming loans is likely to result in an increase in the provision for loan losses. Both total loans outstanding and non- performing loans are proxies for the default risk of the bank. At macroeconomic level, negative relation of loan loss provisioning and GDP growth reflects higher riskiness of the credit portfolio when the business cycle goes down (Pro- cyclical behavior). In order to capture the discretionary component of LLP, we use earnings before taxes and loan loss provisions scaled by total assets EBTP to test if banks use loan loss provisions to smooth their income; a positive relationship between EBTP and LLP would be consistent with the income smoothing hypothesis. Bank managers play down (exaggerate) LLP when earnings are expected to be low (high). Moreover, banks can use the LLP for capital management objective. Even its occurrence is limited to Basel 1 and in some extent Basel 2, banks could manipulate the provisions accounts to keep their capital ratio adequate. Negative relationship between equity to total assets ratio (Equity) and LLP validates capital management hypothesis. LLPi,j,t = LLPi,j,t-1 + EBTP i,j,t + EBTP*variables of interest + Equity i,j,t + Loans i,j,t + NPL i,j,t + GDP j,t + e (3) d) Small Positive Net Income (EQ4): Burgstahler and Dichev (1997), Leventis and Dimitropoulos (2012) and Hamdi and Zarai (2012) [17, 51 & 38] use the frequency of the small positive net income as a metric of managing towards positive earnings. Managers in fact make every possible effort to report small positive net income rather than negative net income (loss avoidance). SPOS equals 1 if net income deflated by lagged total assets is between 0 and 0.25% for each given year and 0 otherwise. After estimating the dichotomous variable, we introduce it as the dependent variable in the following logit regression model: Logit (SPOS i,j,t) = 0 +1variables of interest + 2EBT i,j,t +3size i,j,t + 4Loans i,j,t + 5Equity i,j,t + 6Listed +  7Islamic +  8GDP j,t +e (4) IV. EMPIRICAL RESULTS A. Descriptive Statistics To start our analyses, we report first in Table 2 descriptive statistics of both bank characteristics and macroeconomic variables used for the different models of earnings quality and collected for 158 banks operating in 15 MENA countries during 2000-2013. http://www.ojs.unito.it/index.php/EJIF TABLE 2: DESCRIPTIVE STATISTICS OF BANK–SPECIFIC VARIABLES Num of obs Mean Std. Dev Min Max EBT 1752 0.0193 0.0323 -0.2530 0.4752 EBTP 1751 0.0222 0.1475 -6.0153 0.4752 Income 1756 0.0170 0.0311 -0.2530 0.4752 LLP 1755 0.0038 0.1157 -0.0564 0.3862 NPL 1231 0.2332 3.0225 0.000036 0.8455 Size 1917 7.9291 1.5533 3.6467 11.7104 Equity 1917 0.1541 0.1444 0.000047 0.9944 Loans 1915 0.5304 0.2815 0.0059 0.9934 Deposits 1879 0.6675 0.3776 0.0020 0.9426 GDP 1901 0.0501 0.0401 -0.1508 0.2617 Note: Definitions of variables are as follows: EBT is earnings before taxes of bank i during year t deflated by lagged total assets; EBTP is earnings before taxes and loan loss provisions of bank i during year t deflated by lagged total assets; Income is the net income of bank i during year t deflated by lagged total assets; LLP is the ratio of loan loss provisions to total assets of bank i during year t; NPL is the non-performing loans to total assets ratio of bank i during year t; Size is the log of total assets; Equity is the ratio of bank equity to total assets of bank i during year t; Loans is the total loan to total assets ratio of bank i during year t; Deposits is deposits to total assets ratio of bank i during year t; GDP is the GDP growth rate of country j during year t. Because of the unavailability of some bank-level variables for some countries such as (non-performing loans for Tunisia and Iran); and the use of one-year lagged values for other bank characteristics variables, sample distributions as noted in Tables 3 and 4 differ between the four measures of earnings quality. The samples for earnings persistence (EQ1) and cash-flow predictability (EQ2) tests are relatively smaller than the sample for small positive net income test (EQ4) because of the use of lagged values. The sample for income smoothing test (EQ3) is the smallest one due to the more data requirements. TABLE 3: DISTRIBUTION OF BANK-YEAR OBSERVATIONS BY YEAR EQ1 EQ2 EQ3 EQ4 2001 _ _ 60 110 2002 109 108 61 111 2003 110 110 66 119 2004 117 117 69 131 2005 128 128 71 132 2006 134 134 79 142 2007 142 142 84 147 2008 144 144 96 151 2009 149 149 108 156 2010 154 154 117 154 2011 151 151 117 154 2012 143 143 115 146 2013 85 85 71 103 Total 1566 1565 1114 1756 Regarding the regulatory environment, the Investor Protection Index (IPI) has a median of 4.3 and ranges from 2.7 (for Morocco during the period 2006-2010) to 6.7 (for Saudi Arabia during the period 2009-2013). Official Supervisory Power ranges for our sample from 6 (Morocco) to 14 (Jordan). Private Monitoring Index ranges from 6 (Yemen) to 10 (Bahrain). We note a little difference between MENA countries regarding the investors’ legal protection and banking regulation; both are high on average. TABLE 4: DISTRIBUTION OF BANK-YEAR OBSERVATIONS BY COUNTRY EQ1 EQ2 EQ3 EQ4 Bahrain 155 155 117 173 Egypt 215 214 63 234 Iran 77 77 8 89 Jordan 131 131 122 145 Kuwait 98 98 95 120 Lebanon 202 202 193 225 Malta 34 34 35 41 Morocco 53 53 30 59 Oman 70 70 77 83 Palestine 20 20 6 22 Qatar 92 92 100 101 Saudi Arabia 51 51 48 57 Tunisia 130 130 7 145 UAE 201 201 192 221 Yemen 37 37 21 41 Total 1566 1565 1114 1756 EJIF – European Journal of Islamic Finance No 6, Feb (2017) http://www.ojs.unito.it/index.php/EJIF ISSN 2421-2172 8 To conduct our research, we opt to split the full sample into two subgroups (high and low) based on the cross-country median values (above-median and below-median) of each variable (Investor Protection Index, Official Supervisory Power and Private Monitoring Index) TABLE 5: DESCRIPTIVE STATISTICS OF COUNTRY-LEVEL REGULATORY ENVIRONMENT B. Earnings Persistence And Cash Flow Predictability To obtain full insight on MENA banks’ earnings streams, we firstly employ two related but distinct measures of earnings quality: namely, persistence of earnings and cash flows predictability (ability of current earnings to predict future cash flows). Tables (6 and 7) report regressions results for those two measures. Following previous research on bank earnings quality Altamuro and Beatty (2010), Kanagaretnam et al. (2011, 2014) and Fang et al. (2014) [7, 43, 44 & 33], we use as control variables bank specific characteristics (size, leverage, deposits and loans) and country level variable (GDP growth). Time and country effects are also included in the models. Regarding the estimation method, we use OLS regression with clustered robust errors to account for serial and cross sectional correlations in the residuals Kanagaretnam et al. (2014) [44]. Our results show (from specification (a) to specification (j)) that current earnings before taxes is positively and significantly associated with past earnings (EBTt-1) at the 1% level (see Table 6). For the second measure of earnings quality, specifications summarized in Table 7 display that bank current cash flows are positively and significantly associated with past earnings (EBTt- 1) at 1% and 5% level. These results indicate that MENA banks are characterized by persistent earnings able to predict future cash flows. Further analyses focusing on potential dissimilarities of earnings properties exhibit significant variations between different groups of banks depending on listing status and bank category. In fact, T-values of the interaction term (Islamic*EBT) show that Islamic banks tend to have more persistent (significant at 1% level) and more predictable earnings (significant at 5% level) than their conventional peers (see columns (e) and (f) in each Table 6 and 7). Moreover, regression results show that (at 1% level of significance) listed banks maintain more persistent earnings stream than unlisted banks (see specifications c and d in Table 6). This finding is fully consistent with the usefulness of earnings to equity investors; more persistent earnings would yield to higher equity valuations (Dechow et al., 2010) [26]. Turning now to our main concern, all regression models that control for ownership concentration variables in Tables (6 and 7) show negative and significant coefficient on the interaction variable (control right*EBT). The estimations results reveal therefore that, at 1% level of significance, banks with higher ownership concentration exhibit lower quality of earnings (less persistent and less predictable). In order to compare between state-owned banks and private banks’ earnings quality, we include an interaction term (Control right*State*EBT) in the regression models. Columns (g) in Tables 6 and 7 show that state-owned banks have less persistent and less predictable earnings than their private counterparts (significant at 1% and 10% level respectively). In the aim to examine the effect of banking regulation and supervision as an external governance mechanism on earnings quality; we include additional interaction variables (high ipi*EBT; high official*EBT; and high private*EBT). Columns (h, i and j) in Tables (6 and 7) show that banks in countries with higher private monitoring have more persistent earnings than banks in countries with lower private monitoring index; but banks in countries where the investor protection index is high disclose less persistent earnings. C. Income Smoothing Through Loan Loss Provisions Referring to Dechow et al. (2010) [26], persistence of earnings is dependent to a large extent on firm’s fundamental performance and accounting measurement system. Although separating the role of each is not so obvious, we aim to study to what extent the perceived earnings persistence is achieved by engaging in earnings management practices (Fang et al., 2014) [33]. Our third measure of earnings quality will subsequently focus on the accounting process; and try to capture earnings management using the largest accounting accrual for banks (Loan Loss Provisions). Earnings management is accordingly a specific dimension of earnings quality. In order to examine the determinants of loan loss provisioning practices for MENA banks and test the income smoothing and capital management hypotheses, we use Bouvatier et al. (2014)’s [16] regression model that allows for dynamic adjustments of loan loss provisions. We apply the Generalized-Method-of-Moments Country Investor Protection Index’s Mean Official Supervisory Power Index Private Monitoring Index Bahrain 4,7 11 10 Egypt 3,55 11 7 Iran 3,08 n.a n.a Jordan 3 14 8 Kuwait 5 11 8 Lebanon 5 7 7 Malta 5,7 12 8 Morocco 3,24 6 9 Oman 5 13 7 Palestine 5,3 9 8 Qatar 4,3 9 7 Saudi Arabia 6,32 n.a n.a Tunisia 4,7 9 7 UAE 4 9 7 Yemen 4 12 6 http://www.ojs.unito.it/index.php/EJIF (GMM) estimators appropriate for dynamic models of panel data. The panel estimator controls for potential endogeneity using instruments applied to the lagged dependent variable. We present in Table 2 descriptive statistics of both bank characteristics and macroeconomic variables used for the income smoothing test. Across all sample countries, the mean of loan loss provisions is 0.38% and the mean of earnings before taxes and provisions is 2.22%. Regression results reported in Table 8 for specifications (a, b, c, e and f) show positive and significant (at 1 % level) coefficient on the earnings before taxes and loan loss provisions (EBTP). This finding indicates that banks in MENA countries use discretionary loan loss provisions to smooth their incomes, they play down (exaggerate) provisions when earnings are expected to be low (high). Further, equity to total assets ratio is found to have a negative and significant coefficient at 1% level in all specifications (Table 8), implying that MENA banks use loan loss provisions for capital management objective. These results are consistent with the majority of academic evidence pertaining that commercial banks in all over the world manipulate loan loss provisions for the purpose of income smoothing and capital management (Anandarajan et al., 2007; Taktak et al., 2010; Ben Othman and Mersni, 2014; Curcio and Hasan, 2015) [9, 64, 13 & 25]. Regarding the non-discretionary component of loan loss provisions, data reveals that the lagged dependent variable (LLPt-1) is positively significant which indicates that banks dynamically adjust their provisions according to the credit risk level and macroeconomic conditions to cover future potential losses. Therewith the coefficient of non-performing loans capturing the credit risk is positive and significant. GDP growth coefficient is negative and significant at 1% level. Concerning the listing status of banks, specification (b) in Table 8 shows a negative and significant coefficient of the interaction term (Listed*EBTP) implying that listed banks use less loan loss provisions for income smoothing. Consequently, we deem that unlisted banks engage more aggressively in earnings management than listed banks. This finding contradicts the common opinion that publicly traded banks have more incentives for income smoothing to signal their private information about future bank prospects (Beatty et al., 2002; Anandarajan et al., 2007) [11 & 9]. A plausible explanation is that relatively to unlisted banks, listed banks in MENA countries are more monitored by regulators and official supervisions authority. Extensive supervision and scrutiny reduce thereby incentives to manage earnings and improve TABLE 6: REGRESSION RESULTS FOR THE EARNINGS PERSISTENCE TEST (EBT T ) (A) (B) (C) (D) (E) (F) (G) (H) (I) (J) INTERCEPT -0.0113 (-2.27)** -0.01 (-1.95)* -0.0076 (-1.53) -0.0082 (-1.67)* -0.0118 (-2.41)** -0.0107 (-2.20)** -0.0117 (-2.37)** -0.0090 (-1.83)* -0.0095 (-1.94)* -0.0094 (-1.93)* EBT(T-1) 0.5028 (21.95)*** 0.6066 (23.49)*** 0.3185 (8.33)*** 0.5026 (8.96)*** 0.2612 (7.21)*** 0.3833 (7.65)*** 0.5496 (22.70)*** 0.7216 (11.87)*** 0.6039 (23.30)*** 0.4761 (10.76)*** CONTROL RIGHT*EBT -0.3733 (-7.79)*** -0.2846 (-4.57)*** -0.2158 (-3.82)*** -0.4005 (-8.07)*** -0.3920 (-7.75)*** -0.2859 (-5.34)*** CONTROL RIGHT * STATE*EBT -0.2609 (-5.05)*** LISTED 0.0004 (0.28) 0.001 (0.67) LISTED*EBT 0.2501 (5.72)*** 0.1075 (1.93)* ISLAMIC -0.006 (-3.44)*** -0.0038 (-2.12)** ISLAMIC*EBT 0.3625 (8.45)*** 0.2660 (5.14)*** HIGH IPI *EBT -0.1183 (-2.09)** HIGH OFFICIAL*EBT 0.0548 (1.16) HIGH PRIVATE*EBT 0.1533 (3.62)*** SIZE 0.00035 (0.71) 0.0004 (0.86) 0.00006 (0.13) 0.00023 (0.46) 0.00073 (1.50) 0.0007 (1.42) 0.00052 (1.05) 0.0003 (0.78) 0.0004 (0.88) 0.0003 (0.64) EQUITY 0.0387 (5.59)*** 0.0379 (5.60)*** 0.0400 (5.84)*** 0.0386 (5.69)*** 0.0382 (5.61)*** 0.0370 (5.47)*** 0.0409 (5.94)*** 0.0375 (5.54)*** 0.0373 (5.49)*** 0.0368 (5.45)*** DEPOSITS 0.0042 (2.38)** 0.0038 (2.23)** 0.0037 (2.11)** 0.0036 (2.10)** 0.0045 (2.63)*** 0.0042 (2.45)** 0.0035 (2.04)** 0.0037 (2.19)** 0.0038 (2.22)** 0.0044 (2.60)*** LOANS 0.0039 (1.67)* 0.003 (1.26) 0.0026 (1.10) 0.0022 (0.93) 0.0052 (2.23)** 0.004 (1.75)* 0.0035 (1.50) 0.0029 (1.25) 0.0028 (1.20) 0.0037 (1.61) GDP 0.0653 (3.82)*** 0.0537 (3.21)*** 0.0607 (3.59)*** 0.0534 (3.19)*** 0.0670 (4.00)*** 0.0589 (3.54)*** 0.0614 (3.63)*** 0.0573 (3.41)*** 0.0547 (3.26)*** 0.0642 (3.79)*** Num of banks 157 157 157 157 157 157 157 157 157 157 Observations 1550 1534 1550 1534 1550 1534 1534 1534 1534 1534 Adj.R² 0.3620 0.3855 0.3780 0.3872 0.3899 0.4034 0.3714 0.3868 0.3856 0.3904 * Significant at 10% level ** Significant at 5% level *** Significant at 1% level EJIF – European Journal of Islamic Finance No 6, Feb (2017) http://www.ojs.unito.it/index.php/EJIF ISSN 2421-2172 10 earnings reporting quality (Fonseca and Gonzalez, 2008) [35]. Additional analyses show that Islamic banks exhibit less income smoothing behavior through loan loss provisions than conventional banks. T-value of the interaction term (Islamic*EBTP) is negative and significant at 1% level (see specification (c) in Table 8). This finding is inconsistent with Zoubi and El Ghazali (2007), Ben Othman and Mersni (2014) and Ashraf et al. (2015) [72, 13 & 10] but it converges with Taktak et al. (2010) and Quttainah et al. (2013) [64 & 58]. Focusing now on our main concern which is whether differences in ownership concentration explain differences in the level of earnings management, estimation (d) show that banks with higher concentrated ownership display higher degrees of income smoothing through loan loss provisions (EBTP*control). Going deeper in our research, we introduce in the regression model three dummy variables reflecting three groups of banks with different level of ownership concentration. Widely includes banks with no controlling owner; group 1 includes banks in which the controlling shareholder holds between [10%, 25%] of the voting rights; group 2 includes banks in which the controlling owner holds between] 25%, 50%] of the voting right. We leave group 3, which includes banks in which the controlling shareholder holds more than 50% of the voting rights, to be the reference group. Similarly to Bouvatier et al. (2014)’s [16] findings, specification (f) in Table 8 shows (at 1% level of significance) that banks without majority shareholder (widely) and banks with low and medium level of ownership concentration (groups 1 and 2) behave differently from banks with higher ownership concentration in the way they use loan loss provisions to smooth their incomes. These banks display a lower level of income smoothing behavior than banks with higher ownership concentration (group3). In particular, banks in group 3 display the income smoothing behavior previously observed for the overall sample with a coefficient (0.79) that is significant at 1% level. Due to their large shareholdings (more than 50% of control rights), controlling owner have higher incentives to engage in income smoothing practices. These findings support our TABLE 7: REGRESSION RESULTS FOR THE CASH-FLOW PREDICTABILITY TEST (EBTP T ) (A) (B) (C) (D) (E) (F) (G) (H) (I) (J) INTERCEPT -0.0569 (-1.89)* -0.0553 (-1.01) -0.0533 (-0.98) -0.0544 (-0.99) -0.0576 (-1.04) -0.0566 (-1.03) -0.0575 (-1.05) -0.0543 (-1.00) -0.0553 (-1.01) -0.0552 (-1.00) EBT (T-1) 0.4109 (2.99)*** 0.5132 (9.57)*** 0.3045 (2.48)** 0.5289 (4.85)** 0.1907 (2.29)** 0.3277 (3.62)*** 0.4601 (7.09)*** 0.7130 (5.02)*** 0.5120 (9.31)*** 0.4110 (5.81)*** CONTROL RIGHT*EBT -0.3671 (-2.90)*** -0.3450 (-2.92)*** -0.2375 (-2.05)** -0.4143 (-3.06)*** -0.3753 (-3.31)*** -0.2986 (-2.71)*** CONTROL RIGHT* STATE*EBT -0.2751 (-1.95)* LISTED 0.0105 (1.08) 0.0115 (1.17) LISTED*EBT 0.1162 (0.64) -0.0615 (-0.37) ISLAMIC -0.0035 (-0.62) -0.0001 (-0.16) ISLAMIC*EBT 0.3272 (2.62)** 0.2174 (1.59) HIGH IPI *EBT -0.2056 (-1.29) HIGH OFFICIAL*EBT 0.0243 (0.19) HIGH PRIVATE*EBT 0.1201 (1.18) SIZE 0.0061 (2.03)** 0.0061 (1.04) 0.0052 (0.98) 0.0054 (1.01) 0.0064 (1.10) 0.0064 (1.09) 0.0062 (1.06) 0.0061 (1.03) 0.0061 (1.04) 0.0060 (1.02) EQUITY 0.0928 (2.23)** 0.0927 (2.17)** 0.0954 (2.21)** 0.0948 (2.17)** 0.0915 (2.22)** 0.0910 (2.19)** 0.0959 (2.21)** 0.0920 (2.16)** 0.0925 (2.17)** 0.0919 (2.12)** DEPOSITS 0.0017 (0.16) 0.0014 (0.46) 0.0007 (0.20) 0.0007 (0.20) 0.0021 (0.73) 0.0018 (0.62) 0.0011 (0.35) 0.0013 (0.41) 0.0014 (0.45) 0.0019 (0.59) LOANS 0.0233 (1.63) 0.0225 (1.28) 0.0192 (1.30) 0.0189 (1.27) 0.0243 (1.42) 0.0233 (1.36) 0.0230 (1.31) 0.0224 (1.28) 0.0224 (1.27) 0.0231 (1.32) GDP 0.0541 (0.53) 0.0432 (1.67)* 0.0469 (1.89)* 0.0396 (1.58) 0.0553 (2.17)** 0.0471 (1.87)* 0.0506 (1.91)* 0.0494 (1.74)* 0.0436 (1.70)* 0.0514 (2.03)** Num Of Banks 157 157 157 157 157 157 157 157 157 157 Observations 1549 1533 1549 1533 1549 1533 1533 1533 1533 1533 Adj.R² 0.0288 0.0293 0.0302 0.0302 0.03 0.03 0.0288 0.0295 0.0293 0.0295 * Significant at 10% level ** Significant at 5% level *** Significant at 1% level http://www.ojs.unito.it/index.php/EJIF predictions on the effects of share ownership structures on income smoothing; that is higher ownership concentration lowers earnings quality by increasing income smoothing through loan loss provisions. Nonetheless, this relationship is reversed when the controlling owner is a state. Estimation (e) in Table 8 indicates that banks with higher state ownership display lower degrees of income smoothing through loan loss provisions. The T-value of the interaction term (Control*State*EBTP) is negative and significant at 1% level. Regarding the role of the regulatory environment (see Table 9), regression analyses reveal (at 1% level of significance) that banks in countries with stronger supervisory regimes and higher private monitoring engage less in income smoothing practices through loan loss provisions. However, banks in countries where the investor protection index is high use more discretion in loan loss provisioning. These findings are in consistent with Fonseca and Gonzalez (2008), Biurrun (2010) and Bouvatier et al. (2014) [35, 14 & 16]. Tighter official supervision and greater private oversight improve earnings reporting quality by reducing earnings management practices. D. Small Positive Net Income Since our main concern is to investigate ownership concentration effects on bank earnings quality, we test by a logistic regression whether the occurrence of small positive profits is associated with the existence of blocked shareholders with large ownership stakes. To be consistent with the literature Burgstahler and Dichev (1997), Shen and Chih (2005), Kanagaretnam et al. (2011), Leventis and Dimitropoulos (2012), Hamdi and Zarai (2012), Quttainah et al. (2013) and Fang et al. (2014) [17, 61, 43, 51, 38, 58 & 33 ]; we include along with our variable of interest (ownership concentration) bank specific variables (size, equity, EBT, loans, listed) and a set of yearly and country dummy variables in the logistic model. Initially, logistic regression’s results summarized in Table 10 do not find evidence that banks with higher control concentration are more likely to report small positive earnings (column a). However results reported in columns (b, c and d) show significant coefficient at 5% and 1% level with a positive sign TABLE 8: REGRESSION RESULTS FOR THE INCOME SMOOTHING TEST LLP T (A) (B) (C) (D) (E) (F) LLP T-1 0.1327 (1.83)* 0.1624 (2.15)** 0.1627 (2.13)** 0.1682 (2.14)** 0.1227 (1.87)* 0.1465 (1.90)* EBTP 0.7925 (131.76)*** 0.7974 (126.76)*** 0.7934 (158.45)*** 0.1093 (1.11) 0.7942 (179.59)*** 0.795 (388.88)*** LISTED 0.0278 (8.75)*** LISTED*EBTP -0.7128 (-12.24)*** ISLAMIC*EBTP -0.6264 (-13.92)*** CONTROL*EBTP 1.1241 (6.98)*** CONTROL*STATE *EBTP -0.8431 (-5.33)*** WIDELY 0.0184 (2.50)** WIDELY*EBTP -0.7773 (-13.60)*** GROUP 1 0.0144 (2.02)** GROUP 1*EBTP -0.6466 (-10.12)*** GROUP 2 0.0119 (3.13)*** GROUP 2*EBTP -0.6169 (-7.34)*** EQUITY -0.1216 (-4.68)*** -0.0804 (-2.57)** -0.1055 (-3.87)*** -0.1042 (-3.52)*** -0.1045 (-4.27)*** -0.0984 (-3.38)*** LOANS -0.0015 (-1.24) -0.00005 (-0.08) -0.00089 (-0.79) 0.00024 (0.19) -0.0018 (-1.53) -0.00024 (-0.31) NPL 0.0001 (2.19)** 0.0001 (3.00)*** 0.0001 (2.41)** 0.00008 (2.07)** 0.0001 (3.12)*** 0.00012 (3.16)*** GDP -0.0373 (-3.51)*** -0.0227 (-3.01)*** -0.0231 (-2.82)*** -0.0251 (-2.77)*** -0.0328 (-3.33)*** -0.0254 (-3.11)*** CONSTANT 0.0053 (1.74)* -0.0076 (-1.63) 0.0047 (1.47) 0.006 (1.76)* 0.0072 (2.51)** 0.0022 (0.53) Number of Banks 130 130 130 130 130 130 Observations 1102 1102 1102 1093 1093 1099 R² (Within) 0.99 0.9956 0.9912 0.99377 0.9918 0.9942 * Significant at 10% level ** Significant at 5% level *** Significant at 1% level EJIF – European Journal of Islamic Finance No 6, Feb (2017) http://www.ojs.unito.it/index.php/EJIF ISSN 2421-2172 12 TABLE 9: INCOME SMOOTHING AND REGULATORY FACTORS INCOME SMOOTHING (EQ3) (G) (H) (I) LLP T-1 0.1519 (1.95)* 0.1396 (1.94)* 0.1553 (2.03)** EBTP 0.5228 (4.13)*** 0.7960 (362.23)*** 0.7959 (350.84)*** WIDELY 0.0138 (1.73)* 0.0120 (1.84)* 0.0186 (2.50)** WIDELY*EBTP -0.5693 (-5.66)*** -0.5422 (-7.46)*** -0.7649 (-12.19)*** GROUP 1 0.0151 (2.06)** 0.0101 (1.53) 0.0112 (1.58) GROUP 1*EBTP -0.6011 (-8.32)*** -0.4165 (-4.40)*** -0.5126 (-5.48)*** GROUP 2 0.0125 (3.18)*** 0.0112 (3.41)*** 0.0095 (2.55)** GROUP 2*EBTP -0.5865 (-6.72)*** -0.5548 (-6.27)*** -0.4896 (-5.66)*** HIGH IPI*EBTP 0.273 (2.16)** HIGH OFFICIAL*EBTP -0.3095 (-4.42)*** HIGH PRIVATE*EBTP -0.2737 (-2.85)*** EQUITY -0.1015 (-3.41)*** -0.0954 (-3.40)*** -0.0955 (-3.28)*** LOANS -0.0007 (-0.80) -0.0005 (-0.64) -0.0001 (-0.25) NPL 0.0001 (3.98)*** 0.0001 (3.07)*** 0.0001 (3.08)*** GDP -0.035 (-3.67)*** -0.0301 (-3.80)*** -0.0254 (-3.14)*** CONSTANT 0.0025 (0.61) 0.0041 (1.06) 0.0030 (0.76) Num Of Banks 130 130 130 Observations 1090 1099 1099 R² (Within) 0.9946 0.9944 0.9943 * Significant at 10% level ** Significant at 5% level *** Significant at 1% level for the “cash flow rights” variable, indicating that the propensity of reporting small losses increases as the cash flow rights held by the controlling shareholders increases too. This finding suggests that, in case of financial difficulties largest shareholder owning more cash flow rights has more incentives to aggressively maintain positive earnings; because in case of loss he will bear high fraction proportional to his cash flow stakes. Largest owner hence will put pressure on bank managers to make all possible efforts to boost earnings and avoid losses. The logistic regression reveals in addition a significant negative relationship between earnings before taxes (EBT) and loss avoidance (column a, b and d), but no significant association with (EBTP) earnings before taxes and provisions (column c). This implies that banks are much more likely to report small net income in the interval ]0; 0.0025] when their earnings before taxes is decreasing. In addition, the insignificance of earnings before taxes and provisions (EBTP) indicates that bank managers are engaged in loss-avoidance management of earnings using extra techniques other than managing loan loss provisions. With regard to bank-level controls, our data prove evidence that highly leveraged banks and banks with higher loans are more associated with loss avoidance. High economic growth is negatively associated with the likelihood of small positive earnings. Intuitively, bank profits increase as the economy improves. State-owned banks are not associated with loss avoidance earnings management (estimation d). The negative coefficient on the dummy variable “listed” indicates that listed banks report less frequently small positive income to avoid losses than unlisted banks (results significant at 1% level). This is consistent with our previous finding concerning income smoothing using loan loss provisions (EQ3). Listed banks exhibit higher quality of earnings as opposed to the unlisted banks. Further results indicate that Islamic dummy variable is positively and significantly associated at 1% level with loss avoidance. Islamic banks report more frequently small positive earnings to avoid losses disclosure than their conventional peers. This result converges with Hamdi and Zarai (2012)[38] study showing that, in term of loss avoidance metrics, Islamic banks are more engaged in earnings management practices than conventional banks. Finally, estimation (f) shows that banks in countries with stronger supervisory regimes exhibit less loss avoidance than banks in countries with weaker supervision. V. CONCLUSION This study aims to examine the relationship between shareholding ownership structures, national institutional factors and earnings quality of banks across MENA countries. We use a sample of 158 banks from 15 MENA countries observed over the period (2000-2013). We focus on the essential properties of earnings that should be verified to qualify earnings reporting as a good quality (earnings persistence, ability to predict cash flows, income smoothing through loan loss provisions and small positive net incomes). Those selected measures are supposed to be complementary and not repetitive. We conduct consequently four different regression models that control for the effect of ownership structure and regulatory regime on the banks’ earnings quality. Overall, our findings show that MENA banking institutions are characterized by high quality of earnings in term of persistence and ability to predict cash flows. However, they use loan loss provisioning to smooth incomes, and conduct loss avoidance earnings management. Concentrated ownership have large impacts on the content information of earnings. In effect, our empirical study reveals significant and negative impact of the majority ownership on banks’ earnings quality. Banks with controlling owner display less persistent and less predictable earnings than banks with widely held shares. http://www.ojs.unito.it/index.php/EJIF Furthermore,our data shows that closely held banks have more incentives to manage reported earnings in both forms (income smoothing and loss avoidance) than widely held banks. These findings are supporting the entrenchment hypothesis of ownership concentration. Controlling owners intervene in the earnings determination process in order to conceal from minority shareholders, debt-holders and regulators their abilities to extract private benefits of control. These results validate our first research hypothesis that is ownership concentration in the MENA banking sector lowers earnings quality. Concerning the state ownership, our analyses find evidence that state-owned banks have less persistent and less predictable earnings than their private counterparts. However, state-owned banks conduct less earnings management in term of loan loss provisions and loss avoidance. In comparison with conventional banks, Islamic banks have significantly higher earnings quality in term of earnings persistence, cash flows predictability and income smoothing using loan loss provisions. Nonetheless and consistently with the findings of Hamdi and Zarai (2012) [38], Islamic banks are more likely to manage earnings to avoid losses. These findings lead us to partially validate the third research hypothesis that is Islamic banks display higher earnings quality than their conventional counterparts. Lastly, our findings converge with those of Shen and Chih (2005), Fonseca and Gonzalez (2008) and Biurrun (2010) [61, 35 & 14] regarding the role of banking regulation and supervision in improving the quality of earnings even in the presence of large shareholding. This validates our fourth research hypothesis. However, higher protection of minority shareholders proxied by the investor protection index does not appear as an efficient mechanism to reduce dominant owner opportunism. This research has emphasized the relative effectiveness of the bank governance mechanisms in MENA countries. The study reveals that legal shareholder protection as measured by the Investor Protection Index is insufficient to protect small bank shareholders interests from the discretionary behavior of the controlling shareholders. As well, we do not find evidence supporting the investor TABLE 10: LOGIT REGRESSION OF SMALL POSITIVE INCOME LOGIT (SPOS) (A) (B) (C) (D) (E) (F) (G) CONTROL RIGHT 0.370 (0.87) CASH-FLOW RIGHT 0.999 (2.29)** 0.894 (2.12)** 1.452 (3.65)*** 1.487 (3.78)*** 1.518 (3.84)*** 1.496 (3.83)*** EBT -17.749 (-4.60)*** -18.020 (-4.66)*** -18.832 (-5.30)*** -19.010 (-5.36)** -17.816 (-4.91)*** -19.564 (-5.47)*** EBTP 0.0153 (0.03) STATE 0.1478 (0.65) HIGH IPI -0.070 (-0.25) HIGH OFFICIAL -0.753 (-2.49)** HIGH PRIVATE -0.370 (-1.25) SIZE -0.058 (-0.61) -0.073 (-0.76) -0.1431 (-1.49) -0.0367 (-0.48) -0.025 (-0.33) -0.016 (-0.21) 0.0163 (0.20) LOANS 0.758 (2.47)** 0.777 (2.46)** 0.697 (2.20)** 0.4675 (1.43) 0.484 (1.53) 0.507 (1.69)* 0.527 (1.72)* EQUITY -4.009 (-2.93)*** -4.138 (-3.07)*** -5.167 (-3.40)*** -5.206 (-4.10)*** -5.076 (-3.99)*** -5.313 (-3.97)*** -4.923 (-3.84)*** ISLAMIC 0.851 (2.60)*** 1.010 (2.97)*** 1.017 (3.09)*** 0.670 (2.81)*** 0.656 (2.68)*** 0.993 (3.73)*** 0.812 (3.10)*** LISTED -1.313 (-4.48)*** -1.161 (-4.06)*** -1.265 (-4.54)*** -0.818 (-3.17)*** -0.837 (-3.25)*** -0.932 (-3.59)*** -0.871 (-3.38)*** GDP -7.100 (-1.78)* -6.890 (-1.72)* -9.237 (-2.34)** -4.092 (-1.28) -4.026 (-1.24) -5.053 (-1.55) -3.834 (-1.19) INTERCEPT -4.320 (-2.96)** -4.676 (-3.22)*** -4.145 (-2.90)*** -3.130 (-3.85)*** -3.100 (-3.48)*** -3.086 (-3.76)*** -3.462 (-4.14)*** Num Of Banks 158 158 158 158 158 158 158 Observations 1719 1719 1718 1719 1719 1719 1719 Log Likelihood -329.847 -327.515 -336.498 -342.695 -342.875 -339.527 -342.100 Pseudo R² 0.1984 0.2040 0.1821 0.1672 0.1667 0.1748 0.1686 * Significant at 10% level ** Significant at 5% level *** Significant at 1% level EJIF – European Journal of Islamic Finance No 6, Feb (2017) http://www.ojs.unito.it/index.php/EJIF ISSN 2421-2172 14 protection role in curtailing opportunistic earnings management for banking firms. We argue therefore that shareholder protection laws do not enhance bank governance of MENA counties. Regulators have to concern more about an effective enforcement of shareholders rights that would mitigate diversion behavior of dominant owners and reduce incentives for earnings management. Stronger protection of minority shareholders would limit the negative influence of insider shareholding. REFERENCES [1] Abaoub, E., Homrani, K. and Gamra, S. B. (2013), The determinants of earnings management: empirical evidence in the Tunisian banking industry (1999-2010), Journal of Business Studies Quarterly, Vol.4, N°3, pp:62-72. [2] Abdel Karim, R. A. (1990), The Independence of Religious and External Auditors: the case of Islamic Banks, Accounting, Auditing & Accountability Journal, Vol. 3, N°3, pp: 34-44. [3] Abdul Rahman, R. (2012), Religious ethical values and earnings quality: some evidence from Malaysia, PhD Dissertations, Massey University, Albany, New Zealand. [4] Aharony, J., Wang, J. and Yua,n H. (2010), Tunneling as an incentive for earnings management during the IPO process in China, Journal of Accounting and Public Policy, Vol. 29, N°1, pp:1-26. [5] Ahmed, A.S., Takeda, C. and Thomas, S. (1999), Bank loan loss provisions: a reexamination of capital management, earnings management and signaling effects, Journal of Accounting and Economics, Vol. 28, N° 1, pp: 1-25. [6] Ahrens, B. (2010), Capital market implications of earnings quality, Lohmar-Koln: Josef EulVerlagGmbH. [7] Altamuro, J. and Beatty, A. (2010), How does internal control regulation affect financial reporting? , Journal of Accounting and Economics, Vol.49, N°1, pp: 58-74. [8] Alves, S. (2012), Ownership structure and earnings management: evidence from Portugal, Australasian Accounting, Business and Finance Journal, Vol.6, N°1, pp: 57- 74. [9] Anandarajan, A., Hasan, I. and McCarthy, C. (2007), Use of loan loss provisions for capital, earnings management and signaling by Australian banks, Accounting and Finance, Vol.47, pp: 357-379. [10] Ashraf, A., Hassan, M. K. and Basher, S. A. (2015), Loan Loss Provisioning in OIC Countries: Evidence from Conventional vs. Islamic Banks, Journal of King Abdulaziz University: Islamic Economics, Vol. 28, N°1, pp: 23-59. [11] Beatty, A. L., Ke, B. and Petroni, K. R. (2002), Earnings management to avoid earnings declines across publicly and privately held banks, The Accounting Review, Vol.77, N°3, pp: 547-570. [12] Beaver, W. and Engel, E. (1996), Discretionary behavior with respect to allowances for loan losses and the behavior of security prices, Journal of Accounting & Economics, Vol. 22, N° (1-3), pp: 177-206. [13] Ben Othman H. and Mersni H. (2014), The use of discretionary loan loss provisions by Islamic banks and conventional banks in the Middle East region: A comparative study, Studies in Economics and Finance, Vol.31, N°1, pp: 106- 128. [14] Biurrun, M. V. (2010), Bank Earnings Management: Costs, Drivers and Remedies, Doctoral dissertation, WHU. [15] Bo, X. H. and Wu, L. S. (2009), The governance roles of state- owned controlling and institutional investors: a perspective of earnings management, Economic Research Journal, Vol.2, pp: 81-91. [16] Bouvatier, V., Lepetit, L. and Strobel, F. (2014), Bank income smoothing, ownership concentration and the regulatory environment, Journal of Banking & Finance, Vol.41, pp: 253- 270. [17] Burgstahler, D. and Dichev, I. (1997), Earnings management to avoid earnings decreases and losses, Journal of Accounting and Economics, Vol.24, N°1, pp: 99-126. [18] Busta, I., Sinani, E. and Thomsen, S. (2014), Ownership concentration and market value of European banks, Journal of Management & Governance, Vol.18, pp: 159-183. [19] Caprio, G., Laeven, L. and Levine, R. (2007), Governance and bank valuation, Journal of Financial Intermediation, Vol.16, pp: 584-617. [20] Chen, C.W.K. and Yuan, H. (2004), Earnings management and resource allocation: evidence from China’s accounting- based regulation of rights issues, The Accounting Review, Vol. 79 N°3, pp: 645-665. [21] Chen, K.Y., Lin, K.L. and Zhou, J. (2007), Audit quality and earnings management for Taiwan IPO firms, Managerial Auditing Journal, Vol.20, N°1, pp: 86-104. [22] Cihák, M., Demirgüç-Kunt, A., Pería, M.S.M. and Mohseni A. (2012), Banking regulation and supervision around the world: a crisis update, World Bank Policy Research Working Paper 6286, December. [23] Conroy, S. J. and Emerson, T. L. N. (2004), Business ethics and religion: Religiosity as a predictor of ethical awareness among students, Journal of Business Ethics, Vol.55, pp: 373- 386. [24] Cornett, M. M., McNutt, J.J. and Tehranian, H. (2009), Corporate governance and earnings management at large US bank holding companies, Journal of Corporate Finance, Vol.15, N°4, pp:412-430. [25] Curcio, D. and Hasan, I. (2015), Earnings and capital management and signaling: the use of loan-loss provisions by European banks, The European Journal of Finance, Vol. 21, N°1, pp: 26-50. [26] Dechow, P., Ge, W. and Schrand, C. (2010), Understanding earnings quality: A review of the proxies, their determinants and their consequences, Journal of Accounting and Economics, Vol.50, N°(2-3), pp:344-401. [27] Dechow, P., Sloan, R. and Sweeney, A. (1996), Causes and consequences of earnings manipulations: an analysis of firms subject to enforcement actions by the SEC, Contemporary Accounting Research, Vol.13, N°1, pp: 1-36. [28] Demerjian, P. R., Lev, B., Lewis, M. F. and McVay S. E. (2012), Managerial ability and earnings quality, The Accounting Review, Vol.88,N°2, pp:463-498. [29] Demsetz, H. and Lehn, K. (1985), The structure of corporate ownership: causes and consequences, Journal of Political Economy, Vol. 93, pp: 1155-1177. [30] Ding, Y., Hope, O., Jeanjean, T. and Stolowy, H. (2007), Differences between domestic accounting standards and IAS: http://www.ojs.unito.it/index.php/EJIF measurement, determinants and implications, Journal of Accounting and Public Policy, Vol.26, N°1, pp:1-38. [31] Dyreng, S., Mayew, W.J. and Williams, C.D. (2012), Religious social norms and corporate financial reporting, Journal of Business Finance and Accounting, Vol.39, N° (7-8), pp: 845-875. [32] Fan, J. P.H. and Wong, T. J. (2002), Corporate ownership structure and the informativeness of accounting earnings in East Asia, Journal of Accounting and Economics, Vol.33, N°3, pp: 401-425. [33] Fang, Y., Hasan, I. and Li, L. (2014), Banking reform, risk- taking and earnings quality- evidence from transition countries, Bank of Finland BOFIT- Institute for Economies in Transition, Discussion paper 19/2014. [34] Farazi, S., Feyen, E. and Rocha, R. (2011), Bank ownership and performance in the Middle East and North Africa, Policy Research Working Paper 5620, Washington, DC, United States: World Bank. [35] Fonseca, A.R. and Gonzalez, F. (2008), Cross-country determinants of bank income smoothing by managing loan-loss provisions, Journal of Banking & Finance, Vol.32, pp:217-228. [36] Francis, J., LaFond, R., Olsson, P. M. and Schipper, K. (2004), Costs of equity and earnings attributes, The Accounting Review, Vol.79, N°4, pp: 967-1010. [37] Gaio, C. and Raposo, C. (2011), Earnings quality and firm valuation: international evidence, Accounting and Finance, Vol.51, pp: 467-499. [38] Hamdi, F. M. and Zarai, M. A. (2012), Earnings management to avoid earnings decreases and losses : empirical evidence from Islamic banking industry, Research Journal of Finance and Accounting, Vol.3, N°3, pp: 88-107. [39] Haw, I.M., Ho, S.S.M., Hu, B. and Wu, D. (2010), Concentrated control, institutions, and banking sector: an international study, Journal of Banking & Finance, Vol.34, N°3, pp: 485-497. [40] Isenmila, P. A. and Elijah, A. (2012), Earnings Management and Ownership Structure: Evidence from Nigeria, Research Journal of Finance and Accounting, Vol.3, N°7, pp: 24-36. [41] Jensen, M. and Meckling, W. (1976), Theory of the firm: managerial behavior, agency costs and ownership structure, Journal of Financial Economics, Vol.3, pp: 305-360. [42] Jung, K. and Kwon, S. Y. (2002), Ownership Structure and Earnings Informativeness Evidence from Korea, The International Journal of Accounting, pp: 301-325. [43] Kanagaretnam, K., Lim, C. Y., and Lobo, G. J. (2011), Effects of national culture on earnings quality of banks, Journal of International Business Studies, Vol.42, pp:853- 874. [44] Kanagaretnam, K., Lim, C. Y., and Lobo, G. J. (2014), Effects of international institutional factors on earnings quality of banks, Journal of Banking & Finance, Vol.39, pp:87- 106. [45] Kanagaretnam, K., Lobo, G. J. and Mathieu, R. (2004), earnings management to reduce earnings variability : evidence from bank loan loss provisions, Review of Accounting and Finance, Vol.3, N°1, pp:128-148. [46] Kennedy, E. J. and Lawton, L. (1998), Religiousness and business ethics, Journal of Business Ethics, Vol.17, pp: 163- 175. [47] Kobeissi, N. and Sun, X. (2010), Ownership structure and bank performance: evidence from the Middle East and North Africa region, Comparative Economic Studies, Vol.52, N°3, pp: 287-323. [48] Kung, F. H., Cheng, C. L. and James, K. (2010), Effects of corporate ownership structure on earnings conservatism, Asian Journal of Finance & Accounting, Vol.2, N°1, pp: 47- 67. [49] Kung, F. H., Cheng, C. L. and James, K. (2010), Effects of corporate ownership structure on earnings conservatism, Asian Journal of Finance & Accounting, Vol.2, N°1, pp: 47- 67. [50] Leuz, C., Dhananjay, N. and Wysocki, P.D. (2003), Earnings management and investor protection: an international comparison, Journal of Financial Economics, Vol.69, pp: 505- 527. [51] Leventis, S. and Dimitropoulos, P. (2012), The role of corporate governance in earnings management: experience from US banks, Journal of Applied Accounting Research, Vol.13, N°2, pp: 161-177. [52] Lewis, M. K. (2001), Islam and accounting, Accounting Forum, Vol.25, N°2, pp: 103-127. [53] Longenecker, J. G., McKinney, J. A. and Moore, C. W. (2004), Religious intensity, evangelical Christianity, and business ethics: An empirical study, Journal of Business Ethics, Vol.55, pp: 373-386. [54] McGuire, S., Omer, T.C. and Sharp, N.Y (2012), The impact of religion on financial reporting irregularities, The Accounting Review, Vol.87, N°2, pp: 645-673. [55] Morck, R., Shleifer, A. and Vishny, R. (1988), Management ownership and market valuation: an empirical analysis, Journal of Finance Economics, Vol.20, pp: 293-315. [56] Naciri, A. (2008), Comparative corporate governance: The MENA countries national systems of corporate governance, Routledge Studies in Corporate Governance, New York [57] Penman, S. and Zhang, X. (2002), Accounting conservatism, the quality of earnings, and stock returns, The Accounting Review, Vol.77, N° 2, pp: 237-264. [58] Quttainah, M. A., Song, L. and Wu, Q. (2013), Do Islamic banks employ less earnings management ? Journal of International Financial Management & Accounting, Vol. 24, N°3, pp: 203-233. [59] Sapienza, P. (2004), The effects of government ownership on bank lending, Journal of Financial Economics, Vol.72, pp: 357- 384. [60] Shehzad, C.T., De Hann, J. and Schltens, B. (2010), The impact of bank ownership concentration on impaired loans and capital adequacy, Journal of Banking & Finance, Vol.34, pp: 399-408. [61] Shen, C.H. and Chih, H.L. (2005), Investor protection, prospect theory, and earnings management: an international comparison of the banking industry, Journal of Banking & Finance, Vol.29, pp: 2675-2697. [62] Shleifer, A. (1998), State versus private ownership, Journal of Economic Perspectives, Vol.12, N°4, pp: 133-150. [63] Shleifer, A. and Vishny, R. (1986), Large shareholders and corporate control, Journal of Political Economy, Vol.94, pp: 461-488. [64] Taktak, N., Zouari, S. and Boudrigua, A. (2010), Do Islamic banks use loan loss provisions to smooth their results?, Journal of Islamic Accounting and Business Research, Vol.1, N°2, pp: 114-127. [65] Tsai, H. and Hsieh, Y.T. (2013), Who Monitor Banks? International Evidence from Effects of Large Shareholder EJIF – European Journal of Islamic Finance No 6, Feb (2017) http://www.ojs.unito.it/index.php/EJIF ISSN 2421-2172 16 Structure on Bank Earnings Management, International Research Journal of Applied Finance, Vol. 4, N°10, pp: 1310- 1325. [66] Uygur, O. (2013), Earnings Management and Executive Compensation: Evidence from Banking Industry, Banking & Finance Review, Vol.5, N°2, pp: 33-54. [67] Wan Mohammad, M., Wasiuzzaman, S and Zaini, R. M. (2011), Panel Data Analysis of the Relationship between Earnings Management, Bank Risks, Loan loss Provision and Dividend per Share, Journal of Business and Policy Research, Vol.6, N°1, pp: 46-56. [68] Wang, L. and Yung, K. (2011), Do state enterprises manage earnings more than privately owned firms? the case of China, Journal of Business Finance & Accounting, Vol.38, N° (7‐ 8), pp: 794-812. [69] Warfield, T.D., Wild, J.J. and Wild, K.L. (1995), Managerial ownership accounting choices, and informativeness of earnings, Journal of Accounting and Economics, Vol.20, pp: 61-91. [70] Yeo ,G. H. H., Tan, P., Ho, K. W. and Chen, S. S. (2002),Corporate ownership structure and the informativeness of earnings, Journal of Business, Finance and Accounting, Vol.29, pp: 1023-1046. [71] Zhong, K., Gribbin, D.G. and Zheng, X. (2007), The effect of monitoring by outside blockholders on earnings management, Quarterly Journal of Business and Economics, Vol. 46, N° 1, pp: 1-37. [72] Zoubi, T. A. and Al-Khazali, O. (2007), Empirical testing of the loss provisions of banks in the GCC region, Managerial Finance, Vol.33, N°7, pp: 500-511. http://www.ojs.unito.it/index.php/EJIF EJIF – European Journal of Islamic Finance Editorial Team Editor in Chief Prof. Paolo Pietro Biancone, University of Turin, Italy Editorial Board Prof. Dian Masyita, University of Padjadjaran, Indonesia Prof. Abdulazeem Abozaid, Qatar Faculty of Islamic Studies – Qatar Prof. Ahmad Aref Almazari, King Saud University, Saudi Arabia Prof. Nidal A. Alsayyed, Inayah Islamic Finance Research Institute, USA Prof. Roberta Aluffi, University of Turin - Italy Prof. Ghassen Bouslama, NEOMA Business School - Campus de Reims, France Prof. Nazam Dzolkarnaini, Salford University, UK Prof. Kabir Hassan, University of New Orleans, USA Prof. Khaled Hussainey, Portsmouth University, UK Prof. Rifki Ismal, University of Indonesia Prof. Tariqullah Khan, Hamad bin Khalifa University, Qatar Prof. Ali Khorshid, ICMA Centre Reading University - UK Prof. Amir Kia, Utah Valley University, USA Prof. Laurent Marliere, Université Paris-Dauphine France Prof. Federica Miglietta, University of Bari - Italy Prof. Hakim Ben Othman, University of Tunis - Tunisia Prof. Mohamed Ramady, King Fahd University of Petroleum and Minerals, Saudi Arabia Prof. Mamunur Rashid, Nottingham University, Malaysia Prof. Younes Soualhi, International Islamic University Malaysia Prof. Laurent Weill, University of Strasbourg, France editorial_board.pdf Editor in Chief Editorial Board