5 CHAPTER 5: Conclusions, Limitations and Future Research

Opportunities

As a result of the increasing interests from the decision makers in MENA countries, in

developing and adopting effective corporate governance mechanisms, standards, and practices in

light of the international best practices regarding this issue, we explored in this research the

effect of some corporate governance issues on risk-taking behavior and financial and economic

performance of banks. Some of these corporate governance variables has been corporate

governance at both country and bank-level, Islamic v/s conventional banks, property rights

protection/investor protection, deposit insurance, the non-duality of the CEO0, and share of

government, family and institutional investor ownership, among others. The results showed the

influence on some of these issues on bank's risk-taking and performance, supporting the

argument of that corporate governance plays a crucial role in understanding bank risk-taking

behavior, especially in MENA countries, where building a good corporate governance

environment is still an ongoing process.

In this research, we considered a sample of 165 banks in MENA countries over the period

2005-2012. In accomplishing our research objectives, we used data taken from a range of

relevant public databases. At the firm-level we obtain the data from two sources, the first one is

BankScope International Bank Database maintained by Fitch/Bureau Van Dijk from publicly

available data, which provides information for financial institutions worldwide, the second was

from the annual reports for banks from the public websites whereas all banks in the sample are

publicly traded banks, these annual reports were reviewed to hand collect corporate governance

data and variables on banks level, in order to complete the absence of relevant information in

BankScope database.

At the country-level we obtain the data from different sources, the Access Database of the

World Bank, Heritage Foundation's Index of Economic Freedom, World Governance Indicators

compiled by the World Bank, Worldwide database on deposit insurance (Demirguc-Kunt et al.,

CHAPTER 5

159

2005), and the IFC Doing Business Database, The International Financial Statistics provided

online by IMF.

We apply several selection criteria. First, whenever a bank reports both consolidated and
unconsolidated bank accounts, we drop the latter to avoid double counting. Second, we link the

bank-specific data to various country-level databases that contain information on the

macroeconomic environment. More specifically, where databases offered data for multiple years,

we selected the time period of 2005- 2012 for the study, reflecting data availability across the

majority of variables and countries.

5.1 Relating the effect of corporate governance on banks' risk-taking in MENA countries

The first objective of our work concerned the identification of the determinants of banks

risk-taking in MENA countries. More specificaly, it was intended to analyze how corporate

governance affects the level of banks risk-taking behavior. Additionally, we analyzed the effect

of the bank size, net loans, the growth of gross loans, and cost to income on risk-taking, as well

as the effect of a set of macroeconomic factors. Particularly, we analyzed the relationship

between the risk-taking and of Gross Domestic Product (GDP) and inflation. Following Boyd

and Graham (1986), Čihák and Hesse (2010), Beltratti and Stulz (2009), Lel (2012), Davydov

(2015) and Hasan et al. (2011), we applied the Generalized Method of Moments (GMM) as an

estimation method.

Our results corroborated some empirical findings from other studies (Hassan & Mollah,

2014), proving the effect of corporate governance on the banks' risk-taking behavior. The

findings showed significant remarks. First of all, regarding the whole sample, we found that

corporate governance has a significant and positive impact on banks' risk-taking. Secondly, the

positive relationship between investor protection and banks' risk-taking behavior proved the role

of the investor protection in driving the risk-taking incentives, consistently with Gropp and

Köhler (2010) and Paligorova (2010) findings. Thirdly, the presence of Islamic banks induces

high risk-taking behavior, similarly to Čihák and Hesse (2010).

Fourthly, our empirical results showed that banks with big family ownership have a

significant and negative impact on the risk-taking. This finding is consistent with Anderson et al.

(2003) when mentioned that family banks have incentives to take less risky projects. In contrast,

160

CHAPTER 5

we found that neither important government ownership nor important institutional investor

ownership affect banks' risk-taking in MENA countries. Finally, concerning the deposit
insurance coverage level, the empirical result did not support our hypothesis which expected a

positive relationship between deposit insurance and risk-taking.

A main conclusion in this chapter is related with the fact that the elements considered as

positive from the point of view of corporate governance and protection rights could not be

positive for stability. This good governance acting in the interests of shareholders could lead to

excessive risk-taking. In this sense a conflict of interest between the objectives of the monetary

authority, interested in the solvency of the financial system, and shareholders, trying to maximize

their benefit may occur.

5.2 Relating the effect of corporate governance on banks financial and economic

performance in MENA countries

The second objetive of our work is to study the effect of corporate governance, property

and type of bank on banks' performance in MENA countries. More specifically, in this research

we used three measures to gauge the performance of the banks: ROAA, ROAE and NIM/TA. As

independent variables we included again those related to corporate governance issues: corporate

governance indices at both country and bank-level. Additionally, we consider Islamic banks,

property rights protection/investor protection, deposit insurance, the non-duality of the CE0, and

the share of government, family and institutional investor's ownership. We also controlled by a

set of banks ratios and macroeconomic indicators.

The results allow us to observe how certain variables that imply an increased risk also

negatively impact on the level of performance. In this sense, Islamic banks, despite having a

higher level of risk, do not obtain an improvement in the results in any of the measures

considered. This means that lower performance may be acting as a factor of financial instability.

By contrast, the variables related to corporate governance showed that the indicator at the

company level is significant and the sign indicates a positive relationship with the bank's

performance measured by financial profitability, confirming that greater exposure to risk allows

them to obtain greater profitability. The same is true for corporate governance at the country-

level, where there is a positive impact on all the performance variables. The protection of

CHAPTER5

161

investor rights also has a positive impact, suggesting that better protection of shareholders is
empirically associated with a higher performance. Additionally, there is a positive impact of

family ownership on banks'performance, which in turn achieve greater financial stability, unlike

those in which the government participates in a significant way. Finally, the presence of

institutional factors negatively affects the interest margin obtained.

When we turned to analyze the effect of bank financial performance on the corporate

governance issues in Non-Gulf banks, we found that only the property rights index is significant

and negative, while WGI, as a proxy for governance at country-level, is significant and positive.

At the same time, the participation of the state in the capital of the banks erodes the result. The

same happens when we analyse exclusively the banks included in Gulf countries, and again the

family ownership is important in explaining the performance. Nevertheless, corporate

governance variables are not significant, and we explain this result based on the high

homogeneity of the countries in terms of these factors.

Concerning the behavior during the crisis and after crisis period, differences between the

results appeared in some variables. First, World Governance Indicator hasa positive effect on the

performance either before or after the crisis. In other words, the World Governance ndicator

matters, regardless the crisis, which suggests that good governance at a country-level is good in

all the contexts and has a significant and positive effect on performance. Second, before the

crisis, corporate governance at a firn level has a significant and positive effect on the

performance, while after the crisis it does not have. Thís can be explained because before the

crisis, banks took more risk and thus, more profitability, but after the crises they suffered the

previous policy. Third, before the crisis, Islamic banks exerted a positive effect on the

performance, but before the crisis this 'Islamic effect' turns insignificant.

5.2 Relating the effect of corporate governance on banks financial and economic

performance in MENA countries

The second objetive of our work is to study the effect of corporate governance, property

and type of bank on banks' performance in MENA countries. More specifically, in this research

we used three measures to gauge the performance of the banks: ROAA, ROAE and NIM/TA. As

independent variables we included again those related to corporate governance issues: corporate

governance indices at both country and bank-level. Additionally, we consider Islamic banks,

property rights protection/investor protection, deposit insurance, the non-duality of the CE0, and
the share of government, family and institutional investor's ownership. We also controlled by a

set of banks ratios and macroeconomic indicators.

The results allow us to observe how certain variables that imply an increased risk also

negatively impact on the level of performance. In this sense, Islamic banks, despite having a

higher level of risk, do not obtain an improvement in the results in any of the measures

considered. This means that lower performance may be acting as a factor of financial instability.

By contrast, the variables related to corporate governance showed that the indicator at the

company level is significant and the sign indicates a positive relationship with the bank's

performance measured by financial profitability, confirming that greater exposure to risk allows

them to obtain greater profitability. The same is true for corporate governance at the country-

level, where there is a positive impact on all the performance variables. The protection of

CHAPTER5

161

investor rights also has a positive impact, suggesting that better protection of shareholders is

empirically associated with a higher performance. Additionally, there is a positive impact of

family ownership on banks'performance, which in turn achieve greater financial stability, unlike

those in which the government participates in a significant way. Finally, the presence of

institutional factors negatively affects the interest margin obtained.

When we turned to analyze the effect of bank financial performance on the corporate

governance issues in Non-Gulf banks, we found that only the property rights index is significant

and negative, while WGI, as a proxy for governance at country-level, is significant and positive.

At the same time, the participation of the state in the capital of the banks erodes the result. The

same happens when we analyse exclusively the banks included in Gulf countries, and again the

family ownership is important in explaining the performance. Nevertheless, corporate

governance variables are not significant, and we explain this result based on the high

homogeneity of the countries in terms of these factors.

Concerning the behavior during the crisis and after crisis period, differences between the

results appeared in some variables. First, World Governance Indicator hasa positive effect on the

performance either before or after the crisis. In other words, the World Governance ndicator

matters, regardless the crisis, which suggests that good governance at a country-level is good in

all the contexts and has a significant and positive effect on performance. Second, before the
crisis, corporate governance at a firn level has a significant and positive effect on the

performance, while after the crisis it does not have. Thís can be explained because before the

crisis, banks took more risk and thus, more profitability, but after the crises they suffered the

previous policy. Third, before the crisis, Islamic banks exerted a positive effect on the

performance, but before the crisis this 'Islamic effect' turns insignificant.

Other future research considering the Arab Spring in countries like Tunisia, Egypt,

Libya, Syria and its effect on banks corporate governance variables. on risk-taking,

performance, and other variables.

Another additional area for future research is to examine and analyze banks from other

emerging markets such as Latin American countries, and South Asian countries.

More could be still done to test the benefits of large shareholding were highlighted

under the 'convergence-of-interest' and the efficient monitoring' hypotheses, which

argue that large shareholders are likely to be more efficient than small and dispersed

shareholders in monitoring bank management.

Finally, regarding the suggestions for policy implications, the key policy recommendations

include the following:

Central banks should consider the characteristics of corporate governance, analyzing the

possibility of imposing higher capital requirements for those banks whose only thinking

on the interests of shareholder could lead to an excessive increase in risk.

Central banks will be better prepared to monitor the banking system if they understand

that corporate governance has a significant and positive impact on banks' risk-taking,

the investor protection is driving the risk-taking incentives positively and that Islamic

banks tend to take more risk than the conventional one. If bank regulators are aware of

these relationships, perhaps the bank regulatory oversight function can be made more

effective preventing more bank failures and banking crises in future.

Enabling the national teams for some countries for updating and further improvement

for their corporate governance codes.

In addition to writing laws that support private investors, governments must also enforce

these laws to effectively protect minority investors.