“CEO hubris and Islamic banks’ performance: Investigating the roles of Sharia board vigilance and CEO power”

The purpose of the study is to thoroughly outline how the hubris behavior of chief ex- ecutive officers (CEO) is detrimental to Islamic banks’ (IBs) performance. Specifically, this study attempts to examine the role of the Sharia supervisory board (SSB), board vigilance, and CEO power in the relationship between CEO hubris behavior and decreased IBs’ performance. This study observes IBs’ performance during the period from 2014 to 2020 and develops eight models to test their determinants. Empirical testing of all models shows that CEO hubris has a detrimental impact on IBs’ perfor- mance. The moderating impact test shows the following results: firstly, the presence of SSB, which is represented by the reputation of its members, reduces the detrimental impact of hubris behavior by CEOs on IBs’ performance, while that impact, which is represented by member expertise, does not have a moderating effect. Second, the size and independence of the BOC both weaken the negative relationship between CEO hubris and IBs’ performance. Third, CEO power as represented by tenure and owner- ship has no moderating effect.


INTRODUCTION
Asad and Sadler-Smith (2020), Petit and Bollaert (2012), and Jiang et al. (2011) studied business, management, and finance issues and documented that the hubris behavior carried out by CEOs has an unfavorable impact and the failure of the company's business management. Unfavorable impacts are generally related to important decisions and detrimental outcomes for companies. These can be risk-taking ( (Tang et al., 2015) and the impact of pollution by companies (Zhang et al., 2020). Broadly, CEO hubris was studied in the non-financial industry; some studies have begun to be carried out in the financial/banking industry (Brennan & Conroy, 2013;Ferretti & Gonnella, 2021;Wu et al., 2021).
Investigations proved that the hubris behavior of CEOs in IFIs (IBs) was hindered by the view that, apart from being against the Islamic work ethic, there is an oversight carried out by the Sharia supervisory boards (SSB) to ensure the behavior of both individuals and organizations is sharia-compliant. The Islamic work ethic asserts that work is a virtue in fulfilling one's needs and the importance of establishing balance in individual and social life (Ali, 2008). In contrast, hubris-based behavior is destructive to both a person himself and others (Solihin et al., 2020). Therefore, hubris behavior is an act that is prohibited because it is not under the character of moral and Islamic work ethics (Laldin & Furqani, 2013). Meanwhile, the SSB, which consists of sharia professionals, aims to ensure implementation and compliance with sharia principles without compromise, even at high costs (Aribi et al., 2019;Mihajat, 2019;Muneeza, 2014).
The investigation of hubris behavior by CEOs with SSB vigilance associated with weakened IBs' performance is a novelty. The relationship between corporate governance (CG) as a determinant of IBs' performance is still being studied. The CG components studied in general are not much different from non-IBs, which include the board of commissioners (BOC), board of directors (BOD), committees, and top management. However, there is the main characteristic of IBs, namely the presence of SSB. The study of SSB and IBs' performance is partially related to size, qualification, composition, reputation, and cross membership. The results of the study on these variables are still very diverse in determining IBs' performance.

LITERATURE REVIEW
The framework of the agency theory has been extensively accustomed to illustrate the relationship between CG practices and financial performance (Aslam & Haron, 2020;Bank Indonesia, 2009. Corporate governance, according to the agency theory, exists as a mechanism of internal supervision that can help a company improve its quality (Mollah & Zaman, 2015) and credibility (Mollah et al., 2017). For example, the presence of independent commissioners should mitigate agency conflicts that arise as a result of the manager's objectives and interests being distinct from those of shareholders. Businesses with significant agency costs are more likely to reduce them through the use of corporate governance structures (Fama & Jensen, 1983;Jensen & Mecling, 1976 The study on the existence of SSB as a CG feature was developed according to its characteristics in influencing company performance. The latest study tries to investigate the competence of SSB as measured by expertise and reputation in influencing company performance (Baklouti, 2020). However, Baklouti (2020) researched 42 IBs spread across 12 countries for the period from 2011 to 2018 and failed to find empirical evidence that SSB competence as measured by expertise and reputation affects company performance. Previously, Almutairi and Quttainah This study is focused on the moderating impact of SSB on the relationship between CEO hubris and IBs' performance. The moderating of the SSB variable is used to encourage the corporate governance mechanism (Neifar et  Both studies conclude that SSB has a significant impact on strengthening the relationship between CG mechanisms and the performance of IBs. Indonesian IBs are established in the form of corporations based on the law. As stated in the corporate law, Indonesian companies adhere to a twotier body system in their organizational structure, namely BOC and BOD (The Republic of Indonesia, 2007). The BOC is tasked with representing shareholders to advise and monitor the company's management. Therefore, as a non-executive BOC consists of representatives of stockholders and/ or representatives from independent companies. Furthermore, the BOD is responsible for advising the BOC in managing the company. All members of the two bodies are selected by Bank Indonesia through a fit and proper test mechanism. Bank Indonesia requires at least one of the BOC members to come from outside the company (independent commissioners).
Several previous studies have demonstrated that CEO power plays a role in moderating the relationship between the CEO's hubris behavior and firm performance associated with its antecedent factors. Brahmana et al. (2021) explain that the moderating impact of CEO power occurs when it interacts with antecedent firm performance, namely the retrenchment strategy. Empirical evidence from 319 Malaysian companies was found that a retrenchment strategy would improve company performance when CEO power is high. Velte (2020) reviewed the theme of improving environmental, social, and governance (ESG) performance through the role of CEO power. The results show that the increase in ESG performance is influenced by higher financial performance. This relationship is getting stronger with CEO power (Velte, 2020). Park et al. (2018) and Rizka and Handoko (2020) looked at CEO hubris as an antecedent that interacts with CEO power in influencing financial performance in two different countries: Indonesia and Korea. The results of 100 Indonesian companies concluded that the length of office and the amount of ownership by the CEO did not make the CEO more powerful (Rizka & Handoko, 2020). As a result, the influence of CEO hubris on financial performance did not change (Rizka & Handoko, 2020). While the results of 164 Korean companies concluded that the problems caused by the behavior of CEO hubris would be worse when the CEO was more powerful (Park et al., 2018).
In general, this study aims to obtain empirical evidence that the CEO's hubris behavior is detrimental to IB's performance. Specifically, this study aims to examine whether the presence of SSB, board vigilance, and CEO power further strengthens or weakens the relationship between CEO hubris behavior and decreased performance of IBs. An illustration of the conceptual linkage among the focal variables in this study is imaged in Figure 1.
Given the literature review and the objectives outlined, the current study develops hypotheses that are used to guide empirical findings: H1: CEO hubris has a detrimental impact on the IBs' performance.
H2a: The expertise of SSB will reduce the detrimental impact of CEO hubris on IBs' performance.
H2b: The reputation of SSB will reduce the detrimental impact of CEO hubris on IBs' performance.
H3a: The larger the BOC size, the less detrimental the impact of CEO hubris on IBs' performance.
H3b: The higher the proportion of BOC independence, the less detrimental the impact of CEO hubris on IBs' performance.
H4a: The detrimental impact of CEO hubris on IBs' performance will be stronger in the case of the length of CEO tenure.
H4b: The detrimental impact of CEO hubris on IBs' performance will be stronger in the case of a high level of CEO ownership. This study relies on secondary data of CEO hubris in the form of an "inconspicuous" index. The aim is that the measurement is not biased. Two hubris indicators are used, namely appreciation and overconfidence in their abilities. The study analyzes the CEO's letter in the annual report using the DICTION application to obtain an overconfidence indicator.

METHODS
The paper uses panel data samples for the period 2014-2020. To avoid dealing with cross-section-al heteroskedasticity, the generalized least squares (GLS) procedure is used, which did not take into consideration within-unit serial correlation (Berrington et al., 2006). Hypothesis testing uses random-impacts models because some of the variables remained stable across time for the sample firms in the study population (Berrington et al., 2006).  2014  12  21  33  2015  13  21  34  2016  13  21  34  2017  13  21  34  2018  14  20  34  2019  14  20  34  2020  14  20  34  Total  93 144 237 where j indexes the two SSB variables; k indexes the two BOC variables; m indexes the two CEO power variables.

EMPIRICAL RESULTS
Descriptive statistics of Islamic banks' performance along with the explanatory variables are shown in Table 3. The average performance as measured by ROA for SCBs is higher than for SUBs, namely 6.7 and 2.5%, respectively. Analysis of CEO's letters and awards to measure CEO hu-bris on SCBs is lower than that of SUBs, which are 20.36 and 25.48, respectively. Visually, these figures explain that CEO' hubris behavior is more often carried out by CEOs on SUBs so that their performance is lower than SUBs.
The correlation matrix between independent variables is shown in Table 4. The correlation matrix is used to determine whether independent variables are multicollinear (Gujarati, 2004). Correlations between variables should not exceed 0.80, and the variance inflation factor should not exceed 2 (Gujarati, 2004). It is obvious that the correlation between the independent variables in Table 4 is less than 0.80, and the VIF value for each variable is no more than 2. As a result, it can be concluded that the model does not exhibit multicollinearity.
This study examines two sample groups of IBs: SCBs and SUBs as shown in Table 5. The results provide empirical evidence for the proposed hypotheses. Model I was constructed to examine the negative impact of CEO hubris on IBs' perfor-   Note: *, **, and *** respectively indicate a significant level at 10, 5, and 1%. This study analyses all the research models developed. The first is to analyze the baseline model, which aims to test the significance and direction of the CEO hubris coefficient in model I. The results show the CEO hubris coefficient has a negative and significant direction as a predictor in the two performance groups: SCBs (β = -0.452; ρ < 0.05) and SUBs (β = -0.341; ρ < 0.05). These results are stable in other models with different specifications from models II to VIII, thus providing strong support for concluding H1. Second, the presence of SSB expertise in models III and VIII does not affect the negative impact of CEO hubris on the performance of both SCBs and SUBs. Therefore, it fails to support the statement of H2a. However, model III reveals that the presence of SSB experience weakened the negative impact of CEO hubris on both the performance of SCBs (β = -0.068; ρ < 0.05) and SUBs (β = -0.054; ρ < 0.1).
The role of SSB reputable as a moderator predicted weakening of the negative impact of CEO hubris on IBs' performance continued significantly in model VIII, strongly supporting H2b. Third, this study found that both size and independent commissioners who represent board vigilance minimize the negative impact of CEO hubris on IBs' performance. Model IV reveals the significance and negative direction of the coefficient of CEO hubris*BOC size affecting SCBs performance (β = -0.021; ρ < 0.05) and SUBs performance (β = -0.016; ρ < 0.05). Meanwhile, model V reveals that the interaction coefficient of the CEO hubris*BOC independence has a negative and significant relationship with SCBs performance (β = -0.292; ρ < 0.01) and SUBs performance (β = -0.172; ρ < 0.01).
Through models IV, V and VIII, the results consistently strengthen the statements of H3a and H3b, indicating that board vigilance can reduce the negative impact of CEO hubris on IBs' performance. Finally, this study fails to prove the role of CEO tenure and CEO ownership as moderating impacts that strengthen the negative influence of CEO hubris on IBs' performance. Models VI, VII, and VIII show that the coefficients of both CEO tenure and CEO ownership are not significant in both SCBs and SUBs groups. These results fail to conclude the statements of H4a and H4b.
This study performs additional regressions to ensure the validity of the reported results and to determine their robustness. In empirical studies, a "robustness check" is a common exercise in which the paper examines how certain "core" regression coefficient estimates behave when the regression specification is changed by adding or removing

DISCUSSION
All models with different specifications from models I to VIII provide strong support for the same conclusion that CEO hubris has a negative impact on IBs' performance. H1 proved to be stable be- The results of this study are not in line with Rizka and Handoko (2020) who, using a sample of 100 public companies in Indonesia, observed the behavior of CEOs with various hubris characters.
The results show that CEO hubris is positively correlated with firm performance (Rizka & Handoko, 2020). They argue that the high power distance and collectivist culture in Indonesia limit CEOs' ability to be overconfident and inclined to prove their abilities (Rizka & Handoko, 2020). This study agrees with this argument even though hubris behavior is an act that is prohibited according to sharia. The holy book of the Qur'an documented that early humanity felt the damage caused by hubris behavior (Solihin et al., 2020). Therefore, this study confirms that in addition to the power distance and collectivist culture, there is a belief in their religion related to hubris behavior that has a detrimental impact on both themselves and the organization.
This study does not fully support that the presence of SSB is predicted to attenuate the negative im- The reputation of SSB members shows that they received a recommendation from the MUI to be selected because of their knowledge and experience in dealing with Islamic financial institution problems, including the hubris behavior carried out by the CEO. They will continue to improve their reputation to continue to receive MUI recommendations and remain in the position of members of the sharia national board (Nugraheni, 2018).
Findings related to board vigilance support Park et al. (2018), and Rizka and Handoko (2020). The size of the board and the representation of outside commissioners are two variables that influence its vigilance. Given that independence of the board's structure is a prerequisite for improving board vigilance, it is assumed that these two variables will promote board independence in ways that will reduce managerial entrenchment problems in the organization. The results of this study showed that when the size of the board and when a sufficient number of outside commissioners are assigned to ensure board independence, potential entrenchment problems become less of a source of concern. Corporate governance scholars have recommended that the board should be filled with a high percentage of outside commissioners and that board chairs should not be CEOs to increase board independence (Hayward & Hambrick, 1997; O'Sullivan, 2009). These findings are in accordance with these recommendations.
The empirical evidence of this study fails to conclude that CEO power determined by tenure and ownership will strengthen the negative influence of CEO hubris on IBs' performance. CEO

CONCLUSION
This study investigates the moderating impact of SSB, board vigilance, and CEO power on the relationship between CEO hubris and IBs' performance. Using a sample of IBs divided into two groups (SCBs and SUBs), during the period from 2014 to 2020, this study developed 8 models to analyze the proposed hypotheses (H1-H4b). The analysis uses the random impacts analysis of GLS regression for all panel data.
This study empirically proves that CEO hubris has a negative impact on IBs' performance. This finding is present in all models developed (1 to 8). Further empirical evidence explains that the negative impact of CEO hubris is getting weaker on IBs' performance under the following conditions: a) the interaction of SSB reputation with CEO hubris is getting stronger, b) board vigilance represented by the size and independent commissioners are increasing on CEO hubris. While the investigation of CEO power found no empirical evidence that both tenure and ownership as determinants of CEO power can strengthen the negative influence of CEO hubris on IBs' performance.