The Impact of Corporate Governance on Earnings Management in Palestine: The Moderating Effects of Political Instability

Introduction

At the end of the previous century and the beginning of the twenty-first century, the world has been awoken to a series of global financial scandals with numerous large companies in the United States of America, European Union, and some developing countries (Hamdan et al., 2017). The companies involved in the financial scandals were such as Parmalat, Enron, HealthSouth, WorldCom, Xerox, Adelphia, Tyco, Qwest, and AIG (Alareeni, 2018; Idris et al., 2018; Lopes, 2018; Zgarni & Fedhila, 2019). Thus, the CG issues received eminent attention due to the financial scandals, and also recently, due to an increasing number of bankruptcies emerged by fraud actions in financial reports elements (Buallay et al., 2017). The reason of these scandals is the weaknesses in the CG systems and issues related to EM activities (Bajra & Cadez, 2017; Buallay et al., 2017; Uwuigbe et al., 2014; Linda, et al., 2018; El-Halaby, et al., 2018; Nassar, 2018‎). An effective CG can mitigate the opportunistic manipulation of reported earnings. Conversely, weaknesses of CG mechanisms can stimulate corporate fraud and corruption, and thus further motivate the management to practice EM (Leventis & Dimitropoulos, 2012). According to Alzoubi (2016), the best practice of CG might mitigate the negative sequences of EM activities.

In respect of Palestine (West Bank and Gaza Strip), some scandals have occurred in firms listed on PEX, such as Globalcom., Arab Real-estate company, and Arab Concrete Products Company. PEX announced the suspension of trading the shares of Globalcom Communications until its conditions were corrected. The Palestinian Economic Portal (2016) reported that Globalcom. had engaged in manipulation, including the registration of non-existent assets of about USD 5 million, and also the company was evaluated at an amount greater than its size (PEP, 2016). In addition, PEX stopped trading the shares of the Arab Real Estate Corporation due to the company's non-compliance with the market rules and instructions, and the lack of quality of disclosure and CG as well, which constituted violations of shareholders' rights (WAFA, 2016). Therefore, this study focuses on investigating the quality of BOD and the quality of AC in reducing the level of EM in the non-financial corporations listed in PEX. In addition, this study investigated the role of political unrest situation in Palestine as a moderator variable between the relationship of the BOD and AC quality and EM levels to enable a deeper understanding of the Palestinian context. Moreover, there are very few studies that have been conducted on CG and EM in Palestine (Abdelkarim & Zuriqi, 2020).

The Palestinian Capital Market Authority (PCMA) had issued the CG code in 2009 to prevent the manipulation actions by companies listed on PEX, to prevent the occurrence of more crises in the future. As such, the role of PCMA is to monitor the stock market, adopt good CG regulations, and monitor the compliance of all listed Palestinian companies through sound governance rules, in line with the principles of the Organization for Economic Cooperation and Development (OECD, 2004). Similarly, PEX feels that it has a commitment to enhance the level of application of the Palestinian code of CG by listed firms, of which this in turn increases the transparency that leads to protecting the interests of all stakeholders including the shareholders (Aljadba et al., 2019; Hasan Al-Naser, 2019; Mokadem, and Muwafak, 2020‎). However, according to Hassan et al. (2016), the Palestinian’s experience in CG is still in its infancy phase as a developing country undergoing high political instability. Furthermore, the code of CG in Palestine remains without any reforms since it was issued, which condition reduces the effectiveness of CG.

Therefore, the present study helps in bridging the gap through the findings that can be implemented in the environment of Palestine, a country that is considered as a developing country with the most emerging Bursa and characterized by a high degree of political instability (Aljadba et al., 2019). In line with this implication, the present research was conducted based on two basic dimensions. First, due to the lack of studies on EM in developing countries, especially in Palestine, this study provides evidence on EM and CG effectiveness. Second, this study investigated whether the political environment weakens or strengthens the association between CG effectiveness (BOD and AC quality) and EM activities in Palestine. Hence, this study extends the literature concerning EM by examining political instability as a moderator for the relationship between BOD, AC quality, and EM. Besides contributing to the agency theory, this study contributes to prior literature within the context of political theory. The results of this study suggest that Palestinian users of financial statements, stakeholders, investors, policymakers, PEX, and practitioners should consider the surrounding political environment in Palestine when assessing the extent to which companies practice EM.

This paper is organized in several sections. Section 2 that follows provides an overview of PEX, while Section 3 provides a critical study of the previous literature, based on which the hypotheses were developed. Section 4 presents the research methodology and design, whereas Section 5 provides the analysis of data, followed by a discussion of the findings. The final section presents the conclusion of this study and the recommendations for future research.

An Overview of Palestine Exchange

PEX is considered one of the youngest and emerging exchanges in MENA countries (Hassan et al., 2016). PEX is a private firm that was founded in 1995. Less than two years later, the first session of trading was held in February 1997 to promote investments in the state of Palestine, which had recently emerged. The year of 2010 was a decisive year for the PEX, as it became the second public Arab Bourse, owned entirely by the private sector. PEX focuses basically on magnetizing potential foreign and domestic investors in the districts of diaspora. PEX is registered as a shareholding company licensed by the Palestinian Ministry of National Economy and is subject to its control. Moreover, PEX operates under the supervision of the Palestinian Capital Market Authority to monitor the implementation of sound CG mechanisms to protect current and potential investors and other stakeholders. The electronic website of PEX in 2018 shows that 48 firms were listed on PEX as of May 31, 2018, with a capitalization of about USD 3,789 million distributed among five sectors. These include the banking sector that has nine commercial and Islamic banks, the insurance sector with seven shareholding insurance firms, the investments sector, which includes nine real estate companies, the industries sector that contains thirteen corporations, and finally, the service sector that has eleven services and real estate companies.

The sectors in PEX are divided into two main sections, namely financial listed companies, and non-financial listed companies. The financial sector represents both bank and insurance companies with a total of sixteen companies out of the forty-eight companies listed. The non-financial listed companies include industries, services, and investments sector, with thirty-two companies listed (PEX, 2018). The sample of this study was the Palestinian non-financial companies listed on PEX for several reasons. First, the non-financial companies comprise 70% of the total companies listed on PEX, while the rest of this percentage belongs to financial companies listed. Second, the Palestinian Code of Corporate Governance, issued in 2009, is applied to all listed companies on PEX, while there is another strict code of CG that applies only to listed financial companies. PEX works in a unique environment, which is characterized by high levels of unstable economic and political situation, an environment that is unique to study, especially in terms of the relationship between CG and EM. Accordingly, the objective of this current study was to explore the association between CG effectiveness (BOD and AC quality) and EM in the Palestinian non-financial companies listed on the PEX.

Literature Review and Hypothesis Development

Board of Directors Quality

The main body of CG is the BOD because it plays a monitoring role and deals with agency problems and information asymmetry to mitigate them to the lowest level (Amrah, Hashim, & Ariff, 2015). According to the agency theory, the BOD is one of the essential mechanisms to monitor the acts of managers to maximize the stockholders' interest and wealth (Bathala & Rao, 1995; Fama & Jensen, 1983; Jensen & Meckling, 1976; Miller, 2002). In addition, a high quality of BOD plays a significant role in protecting the interests of stakeholders against management’s self-interests (AbuSiam et al., 2015; Mrabure & Abhulimhen-Iyoha, 2020). Furthermore, the effectiveness of the BOD contributes to the increase in the credibility of financial reports, and reducing EM (Klein, 2002).

Numerous previous studies have investigated the characteristics of CG individually. Chobpichien, Ward, Brown, and Rodriguez (2009) as well as AlQadasi and Abidin (2018) reported that it is essential to deal with CG mechanisms as a group to maintain stockholders’ interest because these mechanisms may be a supplementary to each other. Hence, some prior studies have addressed the BOD characteristics as a group to establish an aggregate measure of BOD effectiveness (e.g., Amrah et al., 2015; Gillan, 2011; AbuSiam, 2015; Hashim & Amrah, 2016). These researchers claim that using a comprehensive measure by combining the BOD characteristics, such a method contributes to avoiding the limitation in using them individually, which might provide mixed and contradictory results. Dhaliwal, Naiker, and Navissi (2006) measured the strength of a firm’s BOD employing four commonly used BOD characteristics; size, independence, ownership, and CEO-duality. They found that the higher effectiveness of BOD results in better accruals quality. Similarly, AbuSiam (2015) conducted a study in Jordan and found that BOD effectiveness is negatively related to EM. In contrary, Hoitash, Hoitash, and Bedard (2009) examined the association between BOD effectiveness with material weaknesses disclosures and they found that there is a positive relationship between these variables. Thus, the current paper presents a study on the effectiveness of BOD, by using six characteristics of BOD, namely BOD size, non-executive directors, meetings, CEO duality, nationality diversity, and the existence of governance committee. The term BOD quality refers to the effectiveness of BOD as an aggregate measure. Therefore, the following hypothesis was developed:

H01:      There is a negative relationship between BOD quality and EM in Palestinian non-financial companies listed on the PEX.

Audit Committee Quality

AC as an internal mechanism of CG is considered a vital tool in any business because of its special responsibilities, duties, and monitoring roles, which assist the BOD in confirming whether financial reports are free of manipulated items, notably reported earnings (Mohamed & Ragab, 2014). The multiple characteristics of an AC, such as the presence of financial and accounting experts and non-executive directors, should improve its quality (AbuSiam et al., 2015; DeZoort et al., 2002; Shawtari et al., 2015). DeZoort et al. (2002) suggest that it is better to explore the characteristics of AC as a composite score by combining them to measure the quality of AC. Zgarni, Hlioui, and Zehri (2016) established a score to measure the quality of AC based on five characteristics; AC existence, seize, independence, meetings of AC, and financial expertise. Empirically, they found that higher quality of AC mitigates EM practices in Tunisia. In relation to the discussion of the findings of these past studies, the current study constructed a score to represent the AC effectiveness by using three characteristics of AC, namely AC size, non-executive directors, meetings, and AC financial expertise. The term AC quality refers to the effectiveness of AC as an aggregate measure. Therefore, the following hypothesis was developed:

H02:      There is a negative relationship between AC quality and EM in Palestinian non-financial companies listed on PEX.

The Moderating Effects of Political Instability on CG and EM

Palestine has been occupied by Israel since 1984 and the Israeli occupation has damaged the Palestinian’s economy due to its dependence on the Israeli’s economy, beginning from 1948 until today (Abdelkarim, 2017). In fact, World Bank (2014) classifies the economy of Palestine as one of the most emerging economies characterized by low productivity, investment, and limited competition because of the high degree of political instability and ongoing wars as well as closures due to the Israeli’s procedures. In addition, the political instability in Palestine creates obstacles to the economic and investment climate, which in turn, reduces the economic growth rates and the productivity of the private sector in Palestine (World Bank, 2015). Hence, since that year, the Palestinian’s economy was down warding, which in effect increases the economic instability and negatively affects investors’ confidence (Hassan et al., 2016). In terms of Arab Spring revolution in 2011, which occurred in some Arab countries, especially Libya, Egypt, Jordan, Algeria, Bahrain, Yemen, and Syria, these revolutions had led to a change in the regimes in some countries (Ghosh, 2015). Palestine shares the Arab Spring states’ unstable economic and political conditions due to the continuous Israeli’s practices against the Palestinians and the Palestinian economy (Ghanem, 2013). In 2012 and 2014, several political events occurred, such as the wars launched on Gaza Strip, ongoing closures, restricting movements of goods, services, and also people as well as the continuous division between Gaza Strip and West Bank. World Bank (2015) reported that the 2014 war on Gaza Strip has left a disastrous impact on the economy of Palestine, which led to a negative growth rate and it also shaved about $460 million off Gaza’s economy, leading to a 15% contraction of its GDP.

Accordingly, several prior studies addressed the economic and political matters and their effect on EM and income smoothing (e.g., Anon 2008; Attia et al., 2016; Filip & Raffournier, 2014; Habib et al., 2013; Harymawan & Nowland, 2016; Hsiao et al., 2016; Obaidat 2017; Wang, 2011;). Hsiao et al. (2016) found that the US petroleum firms have engaged in EM practices during the Arab Spring. Abu Jamia (2013) states that the foreign investments had fled from the countries affected by the Arab Spring due to political unrest and this situation affected the economy of Palestine negatively. Also, Obaidat (2017) reported that Jordanian non-financial listed firms are more likely to engage in income smoothing during the periods of political crises. In contrast, Vladu (2013) found that income smoothing behavior decreased during the crisis periods in Spanish listed companies.

According to Harymawan and Nowland (2016), more political stability would lead to decreased earnings quality. Therefore, they suggested that future studies related to political stability and earnings quality to be conducted in order to motivate firms to improve the quality of financial statements. Ahmad-Zaluki, Campbell, and Goodacre (2011) found the Malaysian IPOs practice the income-increasing EM during the East Asian crisis. In another study, Mangena et al. (2012) found that firms’ performance is positively related to BOD size and ownership concentration in the post-presidential election period, but not in the pre-presidential election period. Overall, they suggested that a large BOD and ownership concentration are more critical in an environment with a severe political and economic crisis. Finally, these researchers recommended that more studies should examine disclosure and EM in an unstable political environment. In a different study, Attia et al. (2016) found that the magnitude of discretionary accruals (DAC) increased in the period after the revolution, which was a highly unstable period in Tunisia. Accordingly, the political and economic turmoil in Palestine is one of the motives for conducting the current study. Based on the above arguments, the following hypotheses were developed:

H03:      Political instability influences the EM practice positively in Palestine.

H04:      Political instability moderates the relationship between BOD quality and EM in Palestinian non-financial firms listed on the PEX.

H05:      Political instability moderates the relationship between AC quality and EM in Palestinian non-financial companies listed on the PEX.

Research Methodology and Design

Data Collection and Sample Selection

The samples of this study were the Palestinian non-financial companies listed on PEX and the data collected cover eight sequential years from 2011 to 2018. The governance data were manually collected from online annual reports that were publicly available on the PEX website from 2011 to 2018 (www.pex.ps). The Thomson Reuters One website was used to collect the financial data related to EM and the control variables. The financial companies were exempted due to different items in their financial reports and different models used in detecting EM. The period was chosen for the purpose of studying political effects as a moderating variable. The period from 2011 to 2018 had many critical events such as the devastating wars launched by Israel on Gaza Strip in 2012 and 2014. Table 1 shows that the total number of listed companies on PEX in 2018 was 48 companies. The total of non-financial listed companies was 33 companies, which includes the industries, services, and investment sections. Fifteen financial companies (banks and insurance companies) were excluded, and four of the companies in the non-financial sector were eliminated due to the missing annual reports during the study period. The final sample comprises of 29 firms for the eight-year period from 2011 to 2018 (232 firm-year observations).

Table 1. Population of the Study

Research Model Specification

The quantitative analysis was the method used in this study for the data over a period of eight years. Ordinary least squares (OLS) regression was carried out to examine the relationship between CG effectiveness and EM. The following is the mathematical representation of the model that includes interaction terms to investigate the moderation effects of POLINS. The formula of the regression model was drawn as follows:

where i = Firm, t = Time period, ABSDAC = The absolute value of DAC computed by the model of Kothari et al. (2005), BODQUALI = BOD quality score ranged from 0 to 6, ACQUALI = AC quality score ranged from 0 to 3, POLINS = Political instability as a moderator variable, Growth = Company’s growth, CASHFLOW = Cash flow from operating activities, FIRMAGE = Firm’s age since the company being established and ɛ = Error term.

Variables Definitions and Measurements

The dependent variable was total accruals. Several prior studies have used cash flow statement approach to calculate the total accruals (e.g. AbdulRahman & Mohamed Ali, 2006; Almomani & Ayedh, 2017; Gounopoulos & Pham, 2018; Klein, 2002; Latif & Abdullah, 2015; Xie et al., 2003). Under this approach, total accruals are computed as the equation below:

where

TACit = Total accruals for firm i in year t

N.Iit = Net income or Earnings before extraordinary items for firm i in year t.

CFOit = Cash flows from operating activities for firm i in year t.

Based on most prior studies, the cashflow from operational activities approach is better than the balance sheet approach because the cashflow approach eliminates any expected errors when computing the total accruals (Almomani & Ayedh, 2017; Gounopoulos & Pham, 2018; Latif & Abdullah, 2015). Therefore, this study uses the cashflow from operating activities to measure the total accruals.

The current study employed the magnitude DAC as a proxy for detecting EM practices, which is the most widely proxy employed by EM literature (Dechow, Ge, & Schrand, 2010; Jackson, 2017). Alareeni and Aljuaidi (2014) reported that the modified Jones model (1995) is not effective in detecting EM in Palestine. Moreover, Kothari et al. (2005) state that the modified Jones model (1995) might lead to severe errors in measuring the DAC because this model does not use the previous return on assets (ROA). Thereby, Kothari et al. (2005) developed a performance-matched DAC approach to overcome the problems of heteroscedasticity and misspecification matters that emerged with the previous models (Chang & Sun, 2009; Habbash et al., 2013; Kothari et al., 2005). Therefore, the current study employed the model of Kothari et al. (2005) to detect the practice of EM in Palestine. The model is as follows:

where:

 = Total accruals for company i in the year t,

= Total assets for company i at end in the year t-1,

= Change in sales revenues for company i in year t from year t-1,

 = Change in accounts receivables for company i in year t from year t-1,

= Gross property, plant, and equipment for company i at end in the year t,

= Lag return of assets for company i from year t-1,

 = Estimated parameters,

 = The residual.

The first independent variable was BOD quality score. This study addressed six characteristics of BOD to construct a composite score as proxy of the quality of BOD. These characteristics were BOD size, frequency of meetings, independence, BOD nationality diversity, CEO duality, the existence of a remuneration, and governance committee. The composite score ranges from zero to six, with each individual metric was scored either 0 or 1. A larger score means that the BOD is more effective, and a lower score means that it is less effective. The current study followed the same procedures as the previous literature to measure overall BOD effectiveness as a composite score (AbuSiam, 2015; Aldamen & Duncan, 2012; AlQadasi & Abidin, 2018; Hoitash et al., 2009; Ishak & Al-Ebel, 2013; Johl, Johl, Subramaniam, & Cooper, 2013; O’Sullivan, Percy, & Stewart, 2008). Table 4 shows how each metric of the composite score was calculated. It is worth noting that this study used the basic characteristics of the BOD to form the index of its quality. These characteristics are stipulated in the Palestinian Code of Governance. As for the diversity in the nationality of the members of the BOD, Palestinian companies are required to disclose the nationality of the members of the BOD, and this feature has been addressed because of its importance, as the PEX seeks to attract foreign investors, which enhances the confidence of investors, whether foreign or local, in investing in companies listed on the PEX. It increases the efficiency of the stock market and the level of competition, as this market operates under an exceptional circumstance, which is political instability.

Table 2. Constructing the BOD quality score

The second independent variable was AC quality score. This study addressed three characteristics of AC based on prior literature, including AC size, expertise, and independence. The composite measure used a three-point scale ranging from zero to three. A larger score represents higher effectiveness of the AC, and a lower score represents a lower effectiveness of the AC. The current study followed the same procedures as previous research in order to measure the AC score (AbuSiam, 2015; Dhaliwal et al., 2007; Goh, 2009; Kent et al., 2010; O’Sullivan et al., 2008). Table 3 shows how the AC Quality Score was calculated.

Table 3. Constructing the AC quality score

The third independent variable was political instability. The current study moderated the political instability based on several critical events that occurred from 2011 to 2014. Based on the prior literature, this study used a dummy variable to detect the effects of political instability in Palestine on the relationship between CG mechanisms and EM. One was assigned for the period from 2011 to 2014 that had wars in 2012 and 2014, and 0 was assigned to the period from 2015 to 2018. This measurement was used as a proxy of political instability by prior literature (Hsiao et al., 2016; Mangena et al., 2012; Obaidat, 2017). Furthermore, the political instability was used as an independent variable in the prior literature to explore its impact on firm performance (Mangena et al., 2012), income smoothing (Obaidat, 2017) and the magnitude of DAC (Attia et al., 2016). By reviewing the prior literature (e.g., Mangena et al., 2012; Obaidat, 2017; Attia et al., 2016), political instability affects the effectiveness of CG and influence the level EM practice. Thereby, this study for the first time uses the political instability variable as a moderator to explore whether it moderates the relationship between BOD quality and AC quality with EM in Palestine.

The current study addressed three control variables, firm age (FIRMAGE), firm growth (Growth), and (CASHFLOW). Several prior studies controlled the growth rate and measured this variable by the change in amount of total assets scaled by the previous amount of total assets (e.g., AbuSiam et al., 2018; Johl et al., 2013; Moslemany & Nathan, 2019). In this study, cash flow was measured by scaling the net operating cash flows by total assets (Can, 2019; Diri et al., 2020; Johari et al., 2009). The last control variable addressed in this study was firm age. Firm age is widely measured by the number of years since a firm was established (Black et al., 2018; Fang, 2016; Moslemany & Nathan, 2019; Shan, 2015; Zaid et al., 2020).

Table 4. Measurements of the variables

Empirical Results and Discussion

Descriptive Statistics

Table 5 shows that the mean of absolute value of DAC (ABSDAC) is 0.0617 with a minimum value of zero and a maximum value of 0.7639. The average of ABSDAC in this study is consistent with the findings of Idris et al. (2018), Abu Siam (2018), and Abbadi et al. (2016) who conducted their studies in Jordan. The findings of this study document that the data of DAC in Palestine were approximately like the neighboring countries. Concerning firm growth, the mean, maximum, and minimum values of firm growth are 0.0326; 1.2468; -0.5814, respectively. The result on firm growth in this study is consistent with the evidence obtained in the Indonesian context as documented by Harymawan and Nowland (2016). These researchers reported the mean, maximum, and minimum values of firm growth of 0.040, 1.225, -0.411, respectively. In the current study, the mean of firm age is approximately 24 years. This finding is closely similar to the study by Bassiouny (2016) in the Egyptian context, of which they found that the mean of firm age was 27 years. In respect of cash flow from operating activities, it is clear from the mean that there is a shortage of cash flow in the Palestinian non-financial firms listed. In comparison, Alsultan (2017) found a mean cash flow of 0.1067 in 85 Saudi non-financial listed companies, while the mean of cash flow in the Palestinian context was 0.0386. These results reflect that Saudi corporations are more effective in generating cash flow in comparison to the corporations in Palestine. Accordingly, the possible explanation of this difference in cash flow is the stable business environment in Saudi Arabia, while the opposite case exists in Palestine. Table 5 shows that the mean of the quality of BOD (BODQUALI) from 2011 to 2018 is 0.460, while the minimum and the maximum scores are 0.833 and 0.000, respectively. The quality of AC (ACQUALI) is 0.416, and the minimum and maximum scores are 0 and 1, respectively. These results are quite low, which shows a relative weakness in the effectiveness of the AC and BOD in the listed Palestinian non-financial companies on PEX.

Table 5. Descriptive statistics of the study variables

Correlation Analysis

Table 6 shows the results of the Pearson correlation tests of the predictors, control variables, and EM (ABSDAC). This Pearson correlation test evaluates the null hypothesis that no linear relation occurs among the variables involved. The significance of the test is indicated by a 0.01 significance level or a 0.05 significance level, and the values recorded are the Pearson correlation coefficients. These linear relationships show that multicollinearity does not exist between the independent variables. However, multicollinearity was tested before running the regression model. Table 6 shows that political instability is positively correlated with EM, which finding is consistent with the hypothesis prediction, while the correlation between BODQUALI and EM is inconsistent with the expectation of the hypothesis due to the positive correlation. On the other hand, ACQUALI correlates negatively with EM, which result is in alignment with the hypothesis.

Table 6. Correlation coefficients between variables of the study

Multiple Regression Results

The panel data regression models were divided into fixed effect method and random effect model (Hamdan et al., 2017). By using the panel data approach in this study, the relationship between EM and CG was captured using these models. The Hausman test (1978) was implemented to compare the fixed effects model, from one side, and the random effect model, from the other side (McKnight & Weir, 2009). The test found that the random effects model was the optimal one for capturing the relationship under study. In terms of the multicollinearity issues in the study’s models, they were investigated by calculating the variance inflation factor-VIF for each variable in each model. According to Gujarati (2009), a VIF higher than 10 indicates a serious multicollinearity problem. In this study, all VIFs are less than 10, which finding indicates that there is no multicollinearity. Table 5.11 shows the VIF values for each variable in each model. The Engle (ARCH) test was used to test for panel data heteroscedasticity (Greene, 2007). Heteroscedasticity was tested using panel groupwise heteroskedasticity tests, which include the Lagrange Multiplier LM Test, Likelihood Ratio LR Test, and Wald Test. These tests indicate that the model has a heteroskedasticity problem. To fix this problem, the cluster-robust standard errors technique was used, as recommended by Wooldridge (2012). Wooldridge (2002) discussed the test for serial correlation within the idiosyncratic errors of a linear panel-data model. Therefore, the Wooldridge test was used to check whether the study’s model suffered from an autocorrelation. The test aimed at testing the null hypothesis, which states that there is no serial correlation. Hence, it can be concluded that there is no serial correlation in the study’s models.

Table 7. Results of Panel Regression of Moderation Models

The panel data regression models were divided into fixed effect method and random effect model (Hamdan et al., 2017). By using the panel data approach in this study, the relationship between EM and CG was captured using these models. The Hausman test (1978) was implemented to compare the fixed effects model, from one side, and the random effect model, from the other side (McKnight & Weir, 2009). The test found that the random effects model was the optimal one for capturing the relationship under study. In terms of the multicollinearity issues in the study’s models, they were investigated by calculating the variance inflation factor-VIF for each variable in each model. According to Gujarati (2009), a VIF higher than 10 indicates a serious multicollinearity problem. In this study, all VIFs are less than 10, which finding indicates that there is no multicollinearity. Table 5.11 shows the VIF values for each variable in each model. The Engle (ARCH) test was used to test for panel data heteroscedasticity (Greene, 2007). Heteroscedasticity was tested using panel groupwise heteroskedasticity tests, which include the Lagrange Multiplier LM Test, Likelihood Ratio LR Test, and Wald Test. These tests indicate that the model has a heteroskedasticity problem. To fix this problem, the cluster-robust standard errors technique was used, as recommended by Wooldridge (2012). Wooldridge (2002) discussed the test for serial correlation within the idiosyncratic errors of a linear panel-data model. Therefore, the Wooldridge test was used to check whether the study’s model suffered from an autocorrelation. The test aimed at testing the null hypothesis, which states that there is no serial correlation. Hence, it can be concluded that there is no serial correlation in the study’s models.

Based on the global chi-square test here, the model is significant, of which the model explains 37.8% of the information in the data. Political Instability (POLINS) was used as the moderation variable to investigate the moderation effect on the relationship between EM and BOD and AC Quality. The interaction terms between POLINS and BOD Quality (BODQUAL_POLINS) and POLINS and AC Quality (ACQUALI_POLINS) were used to investigate the effect of moderation. Based on the model estimates, a significant moderation on the relationship exists between the EM and BODQUALI and ACQUALI at a 0.10 significance level. However, ADQUALI had a significant and negative relationship with EM, while BODQUALI had a significant and positive association with EM.

Table 7 shows a positive relationship between BOD quality and EM practices at p-value <5%. This result was unique, unusual, and not as the hypothesis expected. Hence, H01 is unsupported, and thus rejected. The estimated coefficient implies that an increase in BOD quality by 1% will lead to an increase in EM by 0.063%. This result is inconsistent with the agency theory and some previous studies (AbuSiam, 2015; Dhaliwal et al., 2007; Ishak & Al-Ebel, 2013; O’Sullivan et al., 2008). A possible explanation of this result is that there is still a weak level of compliance by the Palestinian listed companies in applying the rules of the Palestinian code of CG. Additionally, the result of BOD quality is consistent with Hassan (2016), who found that financial performance is negatively associated with CG effectiveness and that CG in Palestine is still in a primitive and in undeveloped stage.

In term of AC quality, Table 5 shows that AC quality is associated negatively and significantly with EM in Palestine at p <5%. Hence, hypothesis 02 is accepted. The estimated coefficient illustrates that an increase in AC quality by 1% will lead to reduced EM level by approximately 0.301%. This output is consistent with the agency theory assumptions that more AC quality and effectiveness lead to the improvement of the financial reporting quality and credibility (Mohamed & Ragab, 2014).

Table 7 reveals that there is a positive and significant association between EM and the political instability at p-value <10%. Thus, managers in Palestine are more likely to engage in EM during the times of severe unstable environment due to the political situation. The estimated coefficient elaborates that when political instability increases by one point, the EM level increases by 0.023 point as well. Such finding is consistent with several previous studies that have addressed financial crises and political environment and their influences on EM (e.g., Attia, et al., 2016; Hsiao et al., 2016; Habib et al., 2013; Harymawan & Nowland, 2016; Obaidat, 2017). In Palestine, the performance of firms listed on PEX has been negatively impacted by the political instability because of the wars in the Gaza strip launched by the Israeli occupation in 2012, and 2014 (PEX, 2018). The Arab Spring revolutions, which occurred in 2011, have left negative influences on the economics of Arab Spring countries and Palestine, which in turn leads to increased information asymmetry and agency problems.

Regarding to the interaction between POLINS and the BODQULAI (BODQUALI*POLINS), Table 7 shows that the beta coefficient for the interaction between political instability and the quality of BOD is negative and significant at p-value <5%. Therefore, H 04 is supported. So, when political instability is quite high, BODs of Palestinian shareholding companies are stronger than during periods of light political instability. A possible explanation of the occurrence of this output that CG might provide a substitute for regulations during a severe political crisis to protect businesses from any opportunistic behaviors. This result is consistent with the arguments of the political theory, which is associated with Gourevitch (2003), Pagano and Volpin (2005) as well as Roe (2003). Roe (2003) argues that stockholders might reform the structure of CG systems to mitigate the negative consequences of threats generating by political and economic instability.

Table 7 demonstrates that the interaction between POLINS and ACQULAI and EM (POLINS*ACQULAI) is positively insignificant at p-value >10%. The outcome supports the conclusion that the effectiveness of AC on EM is weak when moderating political instability. Therefore, hypothesis 05 is supported because the political instability plays a moderating role between the relationship AC quality and EM in Palestine. The weak influence of AC quality during political stability is in line with some previous studies (Ahmad-Zaluki et al., 2011; Hsiao et al., 2016; Obaidat, 2017). Possibly, the BODs in Palestine substituted the AC role throughout the crisis from 2011 to 2014. Another possible interpretation that the voluntary formation of AC in non-financial listed firms on PEX may weaken the role of AC in a situation of unstable political conditions.

In terms of control variables, Table 5 shows that company’s growth (AGROWTH) and EM are positively associated at a p-value <1%. This result reveals that managers may use their discretion regarding EM tools to avoid expected reporting negative growth rates that might affect their bonuses. This outcome is consistent with the study of Moslemany and Nathan (2019) that shows firms with better growth rates have good opportunities to engage in EM. The current study finds a negative relationship between firm age (FIRMAGE) and the magnitude of DAC. Black et al. (2018) and Bassiouny (2016) both concluded that as time passes, companies behave more wisely and gain more experience. The result of CASHFLOW reveals that there is an adverse relationship between CASHFLOW and EM practice in Palestine. This outcome is consistent with many previous studies (e.g., Habbash, 2010; Johari, Saleh, Jaffar, & Hassan, 2009; Masud, Anees, & Ahmed, 2017).

Conclusions

This article aims to explore the influence of BOD and AC quality on EM by using the data from the published annual reports of non-financial companies listed on PEX during 2011-2018. In the study presented in this paper, we addressed two independent variables representing CG effectiveness, namely, BOD and AC quality. The perspectives within the agency theory were considered, which are the most reasonable perspectives in the theory to elaborate the influence of CG effectiveness on EM. Furthermore, political instability was included in the model to investigate the impact of BOD quality and AC quality on EM during severe political instability. The random effects regression results show that BOD quality is positively and significantly associated with EM. This finding suggests that there is a problem in the CG structure in Palestine, which needs to be solved. Hence, PEX must make numerous reforms in the CG Code and motivate Palestinian shareholding companies to be more compliant with its rules. On the other hand, AC quality is related adversely to EM. Interestingly, the results of regression model indicate that political instability is positively related to EM at the significant level and moderates the relationship between BOD and AC quality and EM. Therefore, the interaction between BOD quality and political instability makes the BOD quality more effective in reducing EM under a severe unstable situation. The interaction between political instability and AC quality weakens its relationship with EM, which means AC is not effective in mitigating EM during severe political instability in Palestine since it was positive before the interaction and negative after the interaction.

The practical implication of this study is that it explored the moderating impact of political instability on the relationship between the quality of BOD and the quality of AC with EM. The random effects regression results reveal that political instability, which is a proxy that acts as a dummy variable, is associated with EM positively. The significant result of BOD quality provides support that an increase (decrease) in the quality of the BOD drives a decrease (increase) in EM. The significant result of AC quality provides evidence that an increase (decrease) in the quality of the AC drives an increase (decrease) in EM practices. The unique findings of political instability reveal that the situation of the political environment is essential to understand the association between the quality of the BODs and the ACs with EM in Palestine. Based on this argument, the results of this study illustrate that there is a problem in the CG structure in Palestine. Thereby, PEX should be required to make several reforms in the CG codes and motivate Palestinian corporations to be more compliant with the codes. Relatedly, the study elevates the sensibility of the need to raise the level of the BOD’s effectiveness to reduce EM practices in Palestine where the environment is unstable. In terms of the quality of AC, this study found a negative relationship between the quality of the committee and the level of practicing EM in Palestine. Hence, policymakers and the Palestine Exchange must strengthen the role of ACs in listed Palestinian companies. Additionally, this study demonstrates that the quality of the AC has an effective role in limiting the EM level. Therefore, this study recommends that the role of an AC should be expanded, and all listed non-financial companies are subjected to form an AC.

This study recommends that the policymakers and government bodies should update regulations, laws, and governance codes to provide sufficient protection to the shareholders, which action would attract further local and foreign investors to infuse their investments into the Palestinian economy. This might mitigate the negative consequences of political instability in Palestinian corporations. Good CG and effective BODs should be a pivotal tool to deter EM in an unstable political environment, such as in Palestine. Palestinian government, regulatory bodies, and PCMA should encourage private institutions and listed companies to improve their CG practices to enhance the role of BODs in Palestinian companies.

Empirical studies often have their limitations. However, it is worth doing such studies to get over ancestral limitations. This study also has its own limitations. The sample size was the most critical one facing this empirical work because the companies listed on PEX are very few compared to other stock markets. The reason of few listed companies on PEX is because Palestine, as a very small country is characterized by unstable political and economic situation. The scarcity of previous studies concerning the CG topic and EM issues in Palestine is considered an important limitation, which this study attempted to overcome by doing this research in Palestine. This study excluded the ownership structure as a CG characteristic, which exclusion opens the room for further studies in the future. In addition, future studies should pay attention to the ownership structure by using the aggregate measure as a proxy for the optimal ownership structure and its relationship with EM practices.