This study investigates the level of compliance and disclosure of corporate governance mechanisms in Middle East and North Africa countries. The study uses a panel data of 250 companies from MENA countries between 2009 and 2016. The ordinary least square multiple regression analysis technique is used to examine the relationships. Additionally, to alleviate the concern of potential endogeneity, we use fixed effect regression, two-stage least squares using instrumental variables. The results show that the level of voluntary compliance with and disclosure of corporate governance mechanisms among MENA countries varies substantially across countries and is low. The result is consistent with the neo-institutional theory. Future research could investigate more sets of firm-level internal CG mechanisms and country-level variables and use of weighted index. This study extends, as well as contributes to the extant CG literature by offering new evidence on the effect of corporate governance mechanisms among listed firms in ten different MENA countries within a neo-institutional theoretical perspective. The findings will help regulators and policy makers in their countries to pursue reforms to improve national governance quality.
Enhancing corporate compliance and disclosure to reflect sound CG practices is a global phenomenon    . MENA countries, seeking to encourage domestic savings and attract foreign investment to their capital markets have also engaged in continuous economic and financial reforms  . One question that may be asked is―what underlying theory or framework can best explain the affinity of public corporations to comply with enhanced CG compliance and disclosure practices. Several studies have investigated at this question from the perspective of agency, legitimacy, resource dependence and stakeholder theories  . Notwithstanding, solid argument can be made that neo-institutional theory may best explain the momentum for continuous improvements in CG codes of practice .
According to neo-institutional theorists, institutional factors such as economic, social and political forces, motivate corporate entities to adhere to evolving business norms and practices . Corporations and nation state regulatory bodies are influenced by institutional factors that focus efforts towards economic efficiency and increased social legitimacy through both substantive and symbolic management . At the national level, neo-institutional theory has been used to illustrate how institutional forces and either propel or hinder both the evolution and adoption of corporate practices.
As a result, one of the aims of this study is to contribute to the literature on CG and voluntary disclosure applying neo-institutional theory to corporate efficiency and legitimating motives operating within an institutional environment. There are several motives that help explain why corporations do this. Firstly, a legitimating perspective holds that corporations want to improve their legitimacy and social acceptance and so adhere to regulative institutional pressures in order that they are seen as conforming to expected social behaviour and complying with international standards . Secondly, from an efficiency perspective, neo-institutional theory posits that if corporations adhere to coercive, imitative and normative institutional forces, they can improve their corporate image, gain goodwill and reduce political costs  . This is seen as promoting the capacity of corporations to access important and critical resources such as capital, social relations and business contracts which enhance corporate performance and promote the general interests of shareholders  .
While past studies have investigated the nature and extent and the motives and antecedents of corporate voluntary disclosure  it can be argued that these studies have limitations. In some studies, the number of CG provisions used was too small   , had limited observations and/or referenced too short a period . Some studies have employed a survey to measure compliance with CG best practices   . Arguably, these weaknesses limit the generalisability of their findings. Additionally, emerging markets have shown observable interest in developing CG practices evidenced by the stream of continuing reforms have been introduced over the last decade . There is however an acute scarcity of studies that investigate CG practices in developing countries . This arguably impairs the reliability and validity of application of the findings from developed countries to developing countries, such as those in the MENA region. Lastly, while neo-institutional theoretical perspective has been used to show how institutional forces drive the introduction of CG practices at the firm level  there is a scarcity of studies investigating the diffusion of CG practices at national and corporate levels. This places limits on understanding the rationale underlying world-wide diffusion of CG practices.
The motivation for carrying out this study with MENA countries is their common customs and traditions inclusive of language (Arabic) and religion (Islam) which affect economic, information sharing and corporate practices . This enables the study of harmonisation and convergence of CG codes at corporate and national levels . Additionally MENA countries are increasingly viewed as emerging markets and desirous of developing their stock markets. They are likely to pursue continued economic and financial reforms to attract and capture foreign direct investment . Therefore, the evolution of their GC codes can be seen as an important antecedent to their success . Other motivations are the strong Islamic beliefs in these countries.
This study is expected to expand existing knowledge making contribution to the current literature. The current study extends existing knowledge by offering new cross-country evidence with respect to degree of compliance with, and disclosure of good CG practices in MENA countries. This study has adapted its index from the good CG practices index promoted in the United Nations Conference on Trade and Development . This study will also contribute new knowledge on the effect of antecedents such Islamic values at the levels of firms and countries with respect to voluntary CG compliance and disclosure practices and add to existing literature with respect to how board characteristics and ownership structure influence voluntary CG compliance and disclosure practices. The application of the neo-institutional theoretical perspective will show how good CG practices are introduced at corporate and national level in MENA countries.
2. A Neo-Institutional Framework for Good Corporate Governance Practices
According to institutional theory, organisations become structured and behave in ways that are socially accepted, are influenced by social norms and embrace symbols  . Neo-institutional theory examines the interaction between organisations and their environments. Its aim is to ascertain how the stability and survival of organisations are dependent on the incorporation of institutionalised norms and rules . System-based, institutional theory suggests reciprocity of influence between the organisation and society . Three structural levels of analysis are associated with institutional perception―the social institutional level, the governance structures level and at the level of actors in institutional settings. First, social institutions often global in nature, potentially shape the institutional environment regarding what is socially acceptable and informally diffused in society . Second, at the governance level, the perception is of division into organisations and into organisational sectors or industries. The third level of analysis looks at the behavior of individuals and groups.
Neo-institutional perspective also posits three types of institutional pressure: coercive/regulative, cognitive/mimetic and normative. Neo-institutional theorists argue that organisations under coercive pressure, must adhere to governmental rules or other regulations such as in capital markets. Under cognitive/ mimetic pressure, organisations follow normative practices in their field. In balance with normative pressure, organisations often adhere voluntarily to conventional practices and norms . Consequently, institutional theorists predict that organisational practices over time should become isomorphic because of these pressures .
This study aims to apply the generalised neo-institutional theory, considering efficiency and legitimation as the primary motives underlying the economic variables in the institutional environment . The study will also examine differences between MENA countries and among corporate entities regarding the adoption of CG voluntary disclosure practices.
3. Literature Review and Development of Hypotheses
Several studies have examined antecedents of CG and noted differences in how voluntary disclosures promote good CG practices . This study will expand on the current literature, illustrating connections between voluntary disclosure and good CG practices. Applying neo-institutional theory, it examines the correlation between firm-level CG factors such as Islamic values, board characteristics, and ownership structure mechanisms; country-level factors such as religion and quality of national governance; and the degree to which MENA listed corporate entities engage in voluntary CG compliance and disclosure practices.
3.1. Islamic Values Disclosure
Religious values play an important role in Islamic business operations. Sharia institutions offering Sharia-compliant products do not charge interest (riba) in keeping with the practice that goods and services should only carry a price valuation  . With no service charge of interest, these firms realise uncertain returns based on the borrowing of the realised profits of the firm . Sharia law values also promote voluntary disclosure  .
Neo-institutional theory, from its efficiency perspective, is also represented in the relationship between Islamic banks and their borrowers. This is based on risk sharing or profit and loss sharing. For example, a mudarba contract is based on a predetermined ratio for profit-sharing or limited liability provisions for the borrower or entrepreneur, with banks sharing the losses .
From a legitimisation perspective, neo-institutional theory can be applied to Islamic business ethics, which encourages transparency in business activities and promotes voluntary disclosure . On the basis of these arguments, the first hypothesis is:
H1. There is a positive association between Islamic Values Disclosure Index and the level of voluntary corporate governance disclosure.
3.2. Corporate Board Characteristics Variables
Board directors are considered to be the ultimate decision makers in corporations, monitoring the behaviour of management and protecting the interests of shareholders  carrying out the roles of surveillance, control, advising managers and linking the firm to its external environment  . Corporate board structural characteristics have been shown to have an important impact on the degree of corporate disclosure.
3.3. Board Size
From the efficiency perspective, neo-institutional theory holds that large boards are more efficient in monitoring management and promoting shareholders’ interests  , as large boards are less likely than small boards to be influenced by a strong CEO . Boards also reduce information asymmetry between managers and different stakeholders with respect to the quality of financial reporting .
Neo-institutional theory, from the legitimisation perspective holds that larger boards are more effective in providing better and more expert counselling because some directors on the board may have firm-specific knowledge and managerial expertise  .
Notwithstanding, from an efficiency perspective, other studies have shown that large boards can lack coordination, clutter channels of communication between Board members and give rise to “free-rider” problems  . Given inconclusive findings in theoretical and empirical literature, the second hypothesis is:
H2. There is an association between board size and the level of voluntary corporate governance disclosure.
3.4. Board Diversity
Board diversity is seen as improving the effectiveness of the board   evidenced where board members are drawn from different backgrounds and possess different attributes and characteristics . Some of the attributes are demographic including gender, age, race and ethnicity and others are cognitive or unobservable characteristics including education, and occupation.
Neo-institutional theory, from the efficiency perspective, holds that board efficiency is more likely improved when members are of different genders and cultural backgrounds enabling exploration of strategic issues and options from multiple perspectives that would not be raised on less diverse boards  . From the legitimation perspective, more diversified boards enable better macro-environmental links for companies and with influential stakeholders  enhancing company legitimacy and the board’s trustworthiness .
Several studies have shown that board diversity have had a positive impact on voluntary CG disclosure   , but only a few studies have been carried out in MENA countries . The current research study appears to be the first across MENA countries investigating the impact of board diversity on CG disclosure. Based on these arguments, the third hypothesis is:
H3. There is a positive association between board diversity on the basis of gender and the level of voluntary corporate governance disclosure.
3.5. Board Independence
Outside directors, independent from management, are referred to as non-executive directors (NEDs) . Neo-institutional theory, from the efficiency perspective, holds that boards of directors comprised of more NEDs tend to be more effective in protecting shareholders’ interests because they do not have any special ties to the firm or management  . Outside directors also bring much specialised expertise in terms of technology, law, and capital markets . Neo-institutional theory, from a legitimating perspective, holds that independent NEDS increase greater representation from outside stakeholders, lessening legitimacy concerns based on separation of ownership and control, and encouraging more voluntary disclosure .
Studies have borne mixed results. Studies showing positive results between NEDs and voluntary disclosure include  while other studies have shown negative impacts . Similarly, there are mixed results in the MENA countries with positive results  and negative results in others . On the basis of the inconclusive current data, the fourth hypothesis is:
H4. There is a positive association between the proportion of NEDs and the level of voluntary corporate governance disclosure.
3.6. Government Ownership
Ownership structure mechanisms have been cited as having an impact on voluntary disclosure . Level of ownership concentration and the degree of control exerted by majority shareholders can also affect compliance and disclosure with respect of CG practices.
Neo-institutional theory holds that the highest level of institution in society is government. The three-level model supports this because the state has coercive power to regulate and control the actions of other institutions, namely, firms and organisations . Governments have the power to force compliance to international codes of CG and to voluntary practices such as IFRS which is supported by transnational institutions like EU, OECD and the World Bank  . This theory also holds that corporations with a high percentage of government ownership often have good CG practices . Government support lends legitimacy to corporate operations  and in terms of efficiency assists corporations in obtaining subsidies, tax exemptions and contracts. Based on these arguments, the sixth hypothesis is:
H5. There is an association between government ownership and the level of voluntary corporate governance disclosure.
3.7. Director Ownership
Director ownership is thought to influences decisions with respect to voluntary CG disclosure practices  , but the results are inconclusive . Neo-institutional theory from the efficiency perspective holds that higher director ownership lessens agency conflicts between directors and shareholders through aligning their interests. As a result, boards do not have to apply pressure to increase voluntary CG disclosure . From the legitimisation perspective, where there is lower director ownership, firm are more likely to invest more in CG practices and voluntary CG disclosure, thereby improving the legitimacy of the firm and confidence of stakeholders in the board . Most studies have shown a negative relationship between director ownership and voluntary disclosure of CG practices .  showed an insignificant impact. Based on these arguments, the seventh hypothesis is:
H6. There is a negative association between director ownership and the level of voluntary corporate governance disclosure.
3.8. Block Ownership
According to  , agency conflict is mitigated by concentration of ownership, which decreases information asymmetry, and in the process improves firm value . When there is diffused ownership, firms are more likely to increase corporate disclosure to compensate for the greater monitoring . Neo-institutional theory from the efficiency perspective holds that block ownership or concentrated ownership reduces agency conflict because block ownership serves to monitor management. Therefore, there is less need for increased voluntary disclosure. From the legitimation perspective, there would appear to be less pressure on firms with concentrated ownership to be carrying out more transparent disclosure nor calls for greater public accountability . Block ownership is seen as having negative impact on CG disclosure . Based on these arguments, the eighth hypothesis is:
H7. There is a negative association between block ownership and the level of voluntary corporate governance disclosure.
4. Research Design
Our sample is consisting of 823 non-financial and non-utility firms listed in the stock exchanges of ten different countries. Qatar, Kuwait, United Arab Emirates (UAE), Saudi Arabia, Oman, Jordan, Egypt, Bahrain, Morocco and Tunisia with complete data over the period 2009 to 2016 resulting in 2000 firm-year observations from 250 companies. Ten industries, basic materials, consumer goods, consumer services, financials, health care, industrials, oil and gas, technology, telecommunications, and utilities, were represented among these firms.
Financial and utility firms were excluded, as these industries are subject to different regulations and have different capital structure, which is likely to impact their governance structures differently than firms in other industries . The sampling period starts from 2009 after the global financial crisis 2007/2008 as this increased the debates of the effectiveness of governance mechanisms. Therefore, this study uses time series and cross sectional data.
Data of corporate governance are collected from annual reports manually. Annual reports are the main source of information for this study and the assumption is made that the variables related to corporate governance is accurate and reliable as it is provided by management to the shareholders. The annual reports were obtained from the companies’ websites, capital markets websites and Perfect Information Database. Other accounting and financial variables are collected from DataStream.
Finally, the firm-level data includes firm size, sales growth and leverage. Country-level data includes GDP, inflation and corruption index. This data was collected from the website of the World Bank, other global sources of financial information related to each country, International Monetary Fund website and World Federation of Exchanges.
The variables are categorised into three different groups. The first group of variables are independent variables that contain: Islamic value index (IVDI); total number of directors on the board (BRDS); board diversity (BDIV); non-executive directors (NED); government ownership (GOWN); director ownership (DOWN) and block ownership (BOWN).
The second is the independent variable that represent corporate governance mechanisms index (MCGI) (Table 1, Appendix). To ensure reliability, validity and consistency, the content analysis technique of the coding narratives was employed.  . The index is based on five sections used to construct the sub-indices: ownership structure (OS); financial transparency (FT); auditing (AD); corporate responsibility (CR); and board structure (BS). The index keeps track of a checklist designed in the United Nations Conference on Trade and Development by Experts on International Standards of Accounting and Reporting (ISAR) . The (MCGI) is constructed by giving a value of one if the provision in the corporate governance disclosed and zero otherwise.
The third group are the control variables. These are: (i) firm-level variables, namely firm size (LNFS); sales growth (SGR); leverage (LVG) and (ii) country-level variables, namely gross domestic product growth (GDP); inflation (INF); corruption index (CORR IDX).
After validating all the assumption of multivariate regression, the following multivariate ordinary least square (OLS) regression model is used
5. Empirical Results
Table 2 illustrates summary descriptive analysis of the variables over the eight years investigated (2009-2016). Panel (A) shows descriptive statistics for the independent variables. For example, firm Islamic value index (IVDI), while the average (median) firm complied with 16.33% (0%) of the Islamic index. These findings are supported by other research studies that show that in a low percentage of the sampled firms comply with Islamic values  . The board size (BRDS) the minimum of all firms is (5), while the largest is (22). These are the same statistics for the overall period; the boards studied remained the same throughout. The average size for the overall period was (9.12). Previous studies reveal that boards can be too large or too small. With regard to board diversity (BD), the overall sample period has an average of (8.65), with a standard deviation of (13.22). The minimum for this period is (0) and the maximum is (64.45), which suggests that on average the boards are dominated by males on the listed firms. Further, board independence (NED) demonstrates a minimum of (36%) and maximum of (100%), with average of (84.01%). The overall average for government ownership (GOWN) is 15.73% and the highest average is 32.5% in
Table 1. Summary of variables and measures.
Table 2. Summary descriptive statistics of the dependent, independent and control variables for all sampled firms.
Qatar. Director ownership (DOWN) is an important corporate governance mechanism because directors are considered necessary for monitoring the actions of management. The results show that Saudi listed firms have the lowest average director ownership. The overall period has a mean of (46.23%) and a standard deviation of (28.13%). The minimum for this period is (0) and the maximum is (98.73%). On the other hand Omani firms have the highest average bock ownership (BOWN) and the overall average for all countries is (58.76%).
Panel (B) shows the descriptive statistics for the overall corporate governance compliance and disclosure index (MCGI). The dependent variable shows variability in its distribution. It ranges from the minimum of (25.26%) to a maximum of (89.42%). Noticeably, these findings are consistent with corporate governance literature in Middle East countries .
Table 3 illustrates the correlation matrix (including both Person’s parametric and Spearman’s non-parametric) for the variables to test multicollinearities among variables. Direction and magnitude of coefficients are fairly similar, indicating that any non-normality may not pose a statistical problem. Additionally,
Table 3. Pearson’s and spearman’s correlation matrices of the variables.
Notes: the upper right half of the table shows Pearson’s parametric correlation coefficients, whereas the bottom left half of the table contains Spearman’s non-parametric correlation coefficients. **, and * denote correlation is significant at the 1%, and 5% level, respectively (two-tailed tests). Variables are defined as follows: MENA Countries Corporate Governance Index (MCGI), Islamic Values Disclosure Index (IVDI), Board Size (BRDS), Board Diversity (BDIV), Non-Executive Directors on Board (NED), Government Ownership (GOWN), Directors Ownership (DOWN), Shares Held by Shareholders With at Least 5%. (BOWN), Firm Size. (LNFS), Growth Opportunity (SGR), Leverage (LVG), Gross Domestic Product Growth (GDP), Inflation (INF).
the bivariate correlations among variables used are relatively weak, indicating non-existence of serious multicollinearity problems.
Table 4 presents the findings relating the antecedents of CG compliance and disclosure practices. It reports the findings of the OLS analysis of board characteristics and ownership structure variables with the control variables. With respect to Islamic Values Disclosure Index (IVDI), the model shows a positive and significant relationship with Corporate Governance Compliance and Disclosure (MCGI) (0.034**), suggesting that H1 is empirically supported. This result is consistent with the neo-institutional theory framework prediction. The evidence supports the perspective that corporate entities with an Islamic values orientation are more likely to embrace best practices in CG disclosure. This is seen to enhance their ability to attract and retain investment from Islamic sources. The findings reinforce the neo-institutional legitimisation perspective that corporate entities following Islamic values are most likely to pursue good CG practices to sustain and improve reputation, image and standing. Empirically, the research results are consistent with results of previous studies. They are in line with the  study supporting the perspective that Islamic banks are more inclined to engage with corporate entities that have more transparent CG disclosure practices, the  study that suggests that Islamic values orientation drives voluntary compliance and disclosure practices in Middle Eastern countries and the  study that suggests that Islamic banking institutions adhering to Islamic CG appear to provide more voluntary CG disclosure than their more conventional
Table 4. OLS regression results of corporate governance compliance and disclosure (MCGI).
Notes: coefficients are in front of parenthesis. ***, ** and * denote p-value is significant at the 1%, 5% and 10% level, respectively. Also, year 2013 are excluded from the regression analyses. It is used as base year, respectively, for purposes of comparison.
Second, the findings indicate that board size (BRDS) is negatively and significantly linked to Corporate Governance Compliance and Disclosure (MCGI) (-0.020*),These research findings are also aligned to the   studies that support the neo-institutional legitimisation perspective suggesting that larger boards enhance stakeholder representation, increase the need for voluntary disclosure to attract resources from powerful stakeholders and are often linked to increased monitoring of management activities which can positively impact voluntary CG disclosure practices.
Third, board gender diversity (BDIV) is significantly and positively associated with the MCGI (0.237***), and thus H3 is supported. Empirically, this finding aligns with   and  that board diversity is closely and positively related to CG disclosure practice. The findings reinforce the neo-institutional perspective that diverse boards positively influence adherence to good governance practices as a means to ensure corporate legitimacy, attract resources and enhance the capability of the board and its’ committees to efficiently and effectively monitor management activities, all of which impacts positively on voluntary CG practices. Fourth, the proportion of non-executive directors (NED) is significantly and positively associated with the MCGI (0.072***); therefore H4 is empirically supported. This suggests that boards composed of more independent and outside directors are more likely to favor enhanced disclosure CG practice. Again, our research findings are consistent with prior studies including  and  which also found a positive and significant correlation between the proportion of independent outside directors and voluntary disclosure practices.
Table 4 illustrates findings on ownership structure variables. The findings suggest ownership structure variables have a mixed influence on explaining differences in voluntary CG compliance and disclosure practices. Specifically, government ownership (GOWN) and block ownership (BOWN) have a statically insignificant effect on the level of compliance with and disclosure of CG practice recommendations. Therefore H5 and H7 are not empirically supported. Further, the results run contrary to neo-institutional theory that suggests increased block ownership is correlated with diminished information asymmetry and agency conflict, both of which are expected to impact negatively on disclosure practices .
In terms of control variables, the coefficients on them in Table 4 generally significant, for example, firm size (LNFS), gross domestic product growth (GDP) and inflation (INF) are significantly and positively associated with voluntary CG compliance and disclosure practices, providing empirical support to the results of . Discernibly, sales growth (SG) has insignificant relationship with voluntary CG compliance and disclosure practices. The findings are consistent with the results of prior studies, which find no association between this variable and voluntary CG disclosure .
6. Robustness Check
To determine the robustness of the study’s findings, five additional tests were undertaken. First, we created an alternative index. This index incorporated five sub-indices with equal weight (20%). We sought to determine if the results would hold regardless of the sub-indices weighting. On balance, the results of the weighted and non-weighted indices (MCGI) were similar. The results are consistent with the one obtained using the non-weighted corporate governance compliance and disclosure index (MCGI). Second, we re-estimated the first equation using alternative measure of explanatory variables, namely the logarithm of GDP (LGDP) and the average of the firm Islamic values index (AVIVDI). Third, to test for existence of any possible endogeneity, fixed effect model was estimated to address potential unobserved firm-level heterogeneity that OLS regression model may fail to control for  . Therefore, the model to be
Table 5. Sensitivity of corporate governance mechanisms on corporate governance compliance and disclosure (MCGI).
assessed is identified as:
The results for this modeling are presented in Table 5. Clearly, the results are consistent to those identified in Table 4, leading us to conclude that the findings are robust to any endogeneity arising from omitted factors.
Finally, to be consistent with prior study by  , we utilised their two-stage least squares and lagged-effect tests to correlate errors that might have occurred in OLS regression and to the panel data model. Again, clearly the results are similar reinforcing our belief that our results are robust to endogeneity concerns.
MENA countries have introduced business reforms, including issuing CG codes, to attract more private and foreign investments, but there are few studies looking at level of compliance, and disclosure of, CG practices. This study sets out to study the level of compliance among firms in these countries, and examines how firm-level and country-level factors explain compliance and disclosure related to CG practices and disclosure. Results show similarity in level of compliance among countries. The efficiency and legitimating inferences of neo-institutional theory support these findings, for firms that embrace Islamic values voluntarily, firms with higher board size, diversity and independent boards, government ownership and director ownership, have higher CG scores than other firms. The analysis of the level of compliance indicates that corporate governance practices among listed companies are varying considerably and low.
This study contributes the literature on MENA countries by being one of the largest and most extensive study using 250 MENA listed firms from 2009 to 2016, and including 2000 firm-year observations, which also promotes generalisability of the study. It provides empirical evidence that there are differences in voluntary CG disclosures in terms of both firm-level and country-level factors by using a variety of these factors. This study also contributes to the literature in this area because it uses panel data instead of relying on time series or cross section data, and so reduces the effect of multicollinearity, increases degree of freedom, controls for heterogeneity that is unobserved.
There are implications here for corporations, regulators and policy makers in MENA countries as well as in other developing countries for adopting CG reforms. Firms in MENA countries that follow strict Islamic principles are seen to have greater compliance and disclosure practices, and so firms can aspire to incorporate these principles into their operations for better overall performance.
One recommendation is that researchers, policy makers and others in MENA countries use a constructed index, based on inclusion of best CG practices. Other recommendations are also made with respect to future studies. For example, future studies could also use a larger representative sample than what was used, could examine different CG mechanisms, including external characteristics, such as government and other institutions, and country level factors, such as religion, level of corruption, and how these operate in affecting voluntary disclosure. Methodology could also be improved by using a multi-theoretical framework. Future studies could use other sources of information besides annual reports and carry out both quantitative and qualitative analysis for interpreting results.
Some limitations of this study were the use of limited set of firm-level internal CG mechanisms and country-level variables, and the use of un-weighted CG indices, which though generally robust, could have been better with a weighted index. The coding scheme was conducted by researchers, so that subjectivity may have affected the coding of the index.
Appendix. List of UNCTAD ISAR CGM on CG Compliance and Disclosure (MCGI)