CORPORATE

GOVERNANCE AND
BANK
PERFORMANCE

Papanikolaou Ermina
Patsi Maria
MSc in Banking and Finance
2009

31/1/2010
ABSTRACT

Economic scandals and the recent financial crisis made it essential to investigate the
role of Corporate Governance on bank performance. This paper investigates the relationship
between bank performance, Corporate Governance and other financial elements. The study
uses a sample of 79 banks from Europe, Canada, America, Australia and Japan covering the
four year period 2004-2008. Findings of the study confirm earlier studies. We found a negligible
negative relationship between bank performance and Corporate Governance, while we
observed strong relationship between bank performance, Leverage and Sales growth, as it was
expected. We conclude that despite the light that has been diffused on Corporate Governance
issues during the previous years, there is no strong evidence that Corporate Governance affects
bank performance. Finally, we found a positive correlation between inside shareholders and
bank performance indicating that the more shares held by insiders like officers, directors and
large shareholders the better the performance.

[1]
CONTENTS

I. Introduction…………………………………………………………………………page 4

II. Banking System……………………………………………………………………page 6

III. Corporate Governance Framework………….………………………………….page 8

IV. Literature Review………………………………………………………………….page 8

V. Data and Methodology……………………………………………………………page 16

VI. Empirical Results………………………………………………………………....page 21

VII. Conclusion…………………………………………………………………….…..page 29

VIII. References…………………………………………………………………….…..page 31

[2]
TABLES

Ø Table 1: Summarizing results from comparable empirical studies examining the

relationship of performance with other variables…………….……………………page 9

Ø Table 2: Dependent Variables………………………………………………………page 17

Ø Table 3: Independent Variables…………………………………………………….page 17

Ø Table 4: Descriptive statistics for all variables………………………………….…page 18

Ø Table 5: Correlation matrix……………………………………………………….…page 20

Ø Table 6: Model Summary of ROE regression……………………………………..page 21

Ø Table 7: Model Summary of ROA regression......................................................page 23

Ø Table 8: Model Summary of P/E regression........................................................page 24

Ø Table 9: Model Summary of INVESTMENT RETURN regression.....................page 25

[3]
 I. INTRODUCTION

In the middle of 2007 world economy faced the worst financial crisis, according to
leading economists, since the Great Depression of 1930. Failure of key businesses, decline in
economic activity, bank solvency, decline in consumer wealth, losses on the global stock
markets, mergers, acquisitions and bailouts are some of the effects of this credit crunch
globally. Specifically, the banking sector had to confront major issues caused by the over
extension to credit.

Minor problems in the US housing sector unfolded into a widespread credit and liquidity
crisis that lent to the global economic slowdown showing the unbreakably connection of
individual economies both in the developed and developing world. Particularly, the global
banking system, which was the most profitable sector in 2006, now faces severe difficulties that
threaten global economy. Although many explanations exist for the potential cause of the credit
crunch, many of the causal factors are linked to a failure in Corporate Governance (Moxey and
Berendt, 2008). According to this paper, fundamental principles of good Corporate Governance
were breached and it is suggested that more emphasis should be given to the performance of
Corporate Governance than to its regulatory compliance in order for it to be achieved.
Specifically, it is reported that from the Corporate Governance perspective the main problems
were:

• Failure of institutions to evaluate and manage the interconnection between the inherent
business risks and remuneration incentives.
• Remuneration structures and bonuses that encourage short – term decisions without
supporting long – term shareholder’s interests.
• Lack of influence or power in bank’s risk management departments.
• Weaknesses in reporting on risk and financial transactions.
• Lack of accountability within organizations and between them and their owners.
• Information asymmetry between parties.
• Inability of senior management and board members to understand underlying business
models and products.
• Failure to appreciate cultural and motivational factors, lack of desire to change
inappropriate vision and last but not least, human greed.

[4]
 Lloyd (2009) reported, also on a theoretical level, that current financial crisis can be
viewed as a potential breakdown of Corporate Governance suggesting that board members
failed to understand and respond to the financial risks appropriately. But Corporate Governance
is not only connected with the crisis. Looking a few years back there are a lot of corporate
scandals, such as Enron and Worldcom that stated governance weaknesses due to
inappropriate and ineffective control mechanisms (Biswas and Bhuiyan, 2007). Taking into
consideration the scandals and most importantly the breakdown of the banking system due to
the financial crisis, we are willing through this paper to examine the connection between
Corporate Governance and bank performance that would confirm the theoretical background on
this issue.

There are a lot o papers analyzing bank efficiency, stability, accounting performance and
ownership structure, but a few that examine the relationship between bank performance and
Corporate Governance. Taking into account all these previous studies, our paper focuses on the
connection of Corporate Governance and bank performance using a sample of 79 banks
worldwide, for the period 2004 – 2008. Our purpose is to seek any correlation between
performance variables and Corporate Governance rating that would confirm the theoretical
studies mentioned above. Statistical results indicate that there is no significant correlation
between our dependent variables ROE, ROA and Investment Return and Corporate
Governance Rating, while there is a negative correlation between P/E ratio and Corporate
Governance. Overall, our empirical results are in accordance with previous case studies (Love
et al. 2007).

The rest of the paper is organized as follows: Section II examines the history of the
Banking system in UK, Japan, Canada, Europe and the International banking regulations.
Section III examines the Corporate Governance framework worldwide. Section IV presents a
brief review of the literature related to performance and Corporate Governance. Section V
describes the data and the variables used in the estimation and presents summary statistics.
Section VI presents the empirical results. Section VII presents the concluding remarks and
Section VIII presents the references used in our paper.

[5]
 II. BANKING SYSTEM

The origins of international banking date back over 4.000 years, when various civilizations
used letters of credit and bills of exchange issued across sovereign boundaries to finance
trades. The word “bank” with its physical meaning was used more recently in the 15th century
during Renaissance, when Florentine bankers used desks to make their transactions.

International banking is a substantial component of the global economy because it provides

liquidity around the world through capital flow. 1990s was a decade when international banking
system was prevailed by large institutions of the most developed economies, the United States,
Germany, Japan, United Kingdom and Canada. During that period International banks changed
massively due to the intense competition with other financial institutions which led them into
riskier fields. So, by the beginning of 2000s international banking was in process of defining
itself again (Hughes and MacDonald, 2003).

The banking system in developed countries experienced many changes through the years
until it reached its actual form. Takeovers, mergers, acquisitions, privatization, competition and
other reformation, which had taken place in the past, led banks to compose their profile. The
Canadian and German banking systems are considered the strongest and most efficient all over
the world. The Canadian system is highly internationally expanded, mostly in the US (Hughes
and MacDonald, 2003). The German banking system, which is characterized as bank based, is
divided into three banking sectors, offering a wide range of bank activities. Generally, banks in
developed countries were greatly affected by privatization which created a more liberalized
climate, enhancing competition among banks and non financial institutions. Furthermore, many
financial regulations and supervision systems were established in order to control banking
activities and ensure the safety of investors.

Bank institutions in developing and developed countries represent machines of economic

growth. For this reason, Corporate Governance of banks plays important role in the global
financial scene (Arun and Turner, date not available).

INTERNATIONAL BANKING REGULATIONS

The Basel Committee on Banking Supervision was established in 1975 and its main role
is to formulate standards, guidelines and recommendations in banking supervision

[6]
internationally. Its main purpose is to encourage convergence towards common approaches
and standards. Under this aspiration, the Committee published Basel I in 1988 and Basel II in
2006. Basel I established a set of minimum capital requirements for banks by requiring higher
capital ratios in order to strengthen the soundness and stability of the international banking
system. Basel I nowadays is considered outmoded and it has been replaced by Basel II, which
is a more comprehensive measure. The Basel II framework is trying to improve the existing
rules by aligning regulatory capital requirements more closely to the underlying bank risks and
also to encourage banks to identify the risks that they may face today and in the future and
develop their ability to manage them.

The Basel II framework consists of three pillars:

• Pillar 1: sets out the minimum capital requirements firms will be required to meet to
cover credit, market and operational risk.
• Pillar 2: sets out a new supervisory review process. This requires financial institutions to
have their own internal processes to assess their overall capital adequacy in relation to
their risk profile. These are subject to review and evaluation by their supervisors.
• Pillar 3: cements Pillar 1 and 2 and is designed to improve market discipline by requiring
firms to publish certain details of their risks, capital and risk management. The intention
is that these disclosures should be in line with how senior management and the Board
assess and manage the institution’s risks. (Hassan, Wolfe and Maroney, 2004)

Huge corporate and accounting scandals, such as Enron, Tyco International and
WorldCom have led to the establishment of the United States federal law “Sarbanes–Oxley
Act of 2002” on July 30, 2002. The Act contained 11 sections, ranging from additional
corporate responsibilities to criminal penalties and required the implementation of rulings on
requirements by the Securities and Exchange Commission. According to Garneau and Shahid
(2009), Sarbanes – Oxley Act is redundant and imposes additional unnecessary compliance
costs without having any effect on the banking sector.

[7]
 III. CORPORATE GOVERNANCE FRAMEWORK

Corporate Governance can be described as a system that tries to provide guidelines and
principles to the board of directors in order to execute their responsibilities appropriately and to
satisfy shareholders eliminating moral hazard problems. In this point, it is worth mentioning that
a global and unified standard for corporate governance cannot be applied because it could not
be responsive to local economies. The major difference between the corporate governance
systems of the US and UK and that of Europe is that European countries give more emphasis
on cooperative relationships and consensus, while Anglo-Saxon tradition focuses on market
processes and competition (Nestor and Thompson, 2000). Despite their success in the
twentieth century, Anglo-American economies were spotted by Corporate Governance failures.
Since 1991 many reforms of the UK corporate governance model have taken place concerning
disclosure matters, shareholders voting and decision-making factors in order to regulate the
disorganized set of practices (Clarke Thomas, 2007).

In the rest Europe Corporate Governance is characterized by institutional diversity.

However, the German system prevails in Northern Europe, while the Latin model exists in
Southern Europe. For example in Italy, corporate governance is poor, with banks having a stake
in corporate financing but playing a minor role in governance, whereas in France corporate
governance is dominated by cross-shareholdings. The fact is that in recent years both models
tend to adapt elements from the Anglo-American system. Different European countries handle
the issue of corporate governance in different ways. Some of them emphasize on a wider range
of stakeholder interests and others emphasize on the ownership rights of shareholders. Since
1995 many European governance codes were applied in an effort to improve corporate
governance system (Clarke Thomas, 2007).

IV. LITERATURE REVIEW

A lot of attention has been given the last years on corporate governance which has
become an issue of interest across the world, especially during the last economic crisis and the
financial devastation of many companies and banks. However, very little attention has been

[8]
 given on both corporate governance and performance of banking sector globally. What we
should mention is that there are not any previous studies on this topic for International banks,
but there are a lot of case studies on ownership, corporate governance and performance of
banks or firms in general.

Table 1: Summarizing results from comparable empirical studies examining the relationship of
performance with other variables

STUDY SAMPLE VARIABLES TESTED RESULTS

Positive
relationship with
industry
concentration,
INDUSTRY
growth (in sales
CONCENTRATION,
and assets),
GROWTH (in sales and
market share,
assets), MARKET SHARE,
size (in sales)
SIZE (in assets), SIZE (in
and advertising
sales), CAPITAL
Capon, Farley and Hoenig 320 empirical studies INVESTMENT INTENSITY (at
Not significant
(1990) between 1921-1987 industry), CAPITAL
with size (in
INVESTMENT INTENSITY (at
assets) and firm
firm/ business),
control (owners
ADVERTISING, DEBT, FIRM
vs managers)
CONTROL (owners vs
managers)
Negative
relationship with
capital investment
intensity (at firm/
business), debt
Positive
relationship with
growth (in sales
and assets),
GROWTH (in sales and
foreign ownership
1005 Indian firms assets), FOREIGN
and ownership
Douma et al.(2006) between 1999–2000 OWNERSHIP, OWNERSHIP
by owner
BY OWNER MANAGERS,
managers
AGE
Negative
relationship with
age

[9]
 Negative
relationship with
All Kenyan financial FOREIGN OWNERSHIP,
Barako Dulacha G. and foreign
institutions between BOARD COMPOSITION,
Tower Greg (2006 – 2007) ownership, board
2000-2004 GOVERNMENT
composition and
OWNERSHIP
government
ownership
Positive
relationship with
foreign ownership
Choi and Hasan (2005) 77 Korean banks SIZE (in sales), FOREIGN
between 1998 – 2002 OWNERSHIP Not significant
with size (in
sales)
Positive
Mary Hallward-Driemeier, 1,500 Chinese PRIVATIZATION, relationship with
Scott Wallsten and Lixin enterprises, data OWNERSHIP BY OWNER privatization and
Colin Xu (2006) period 2000 MANAGERS ownership by
owner managers
511 banks from 73 Negative
Lensink and Naaborg
countries between FOREIGN OWNERSHIP relationship with
(2007)
1998–2001 foreign ownership
Positive
relationship- no
51 Indonesian banks CORPORATE
Tandelilin et al. (2007) linear effect with
between 1999-2004 GOVERNANCE
corporate
governance
Significant but
economically
107 Russian and 50
Love and Rachinsky CORPORATE unimportant
Ukrainian banks
(2008) GOVERNANCE relationship with
between 2003-2006
corporate
governance
Not significant
561 firms from Czech,
OWNERSHIP relationship with
Spatareanu (2005) Poland and UK
CONCENTRATION ownership
between 1995-1998
concentration
Positive
relationship with
size (in assets),
debt and board
SIZE (in assets), DEBT,
size
Kyereboah-Coleman and 18 Banks in Ghana BOARD COMPOSITION,
Biekpe (2006) between1997- 2004 BOARD SIZE, TENURE OF
Negative
CEO
relationship with
board
composition,
tenure of CEO

[10]
 Negative
1510 German firms
Heiss and Koke (2004) CHANGE IN CONTROL relationship with
between 1986 – 1995
change in control
Positive
Kapopoulos and Lazaretou 175 Greek firms for relationship with
OWNERSHIP STRUCTURE
(2007) year 2000 ownership
structure
Negative
170 banks between MERGERS AND relationship with
Delfino (2007)
1993 – 2000 ACQUISITIONS mergers and
acquisitions
Positive
443 firms between relationship with
Rose (2007) OWNERSHIP BY BANKS
1998 – 2001 ownership by
banks
800 highest paid Negative
Mahajan and Lummer executives employed MANAGEMENT relationship with
(1993) by US corporations DEPARTURES management
between 1972-1983 departures

Specifically, Choi and Hasan (2005) examined the effect of ownership and corporate
governance on Korean bank’s performance during 1998 – 2002 by using a simple ordinary least
squared model reporting that the existence of one foreign director on the board improves bank
performance significantly, but multiple foreign directors on the board do not improve bank’s
performance. Using a sample of recent Korean banking industry for 1994-2000, Lee (2005)
examined how the effectiveness of managerial ownership is affected by the regulatory regimes
in banking industry and the banks' moral hazard incentives. The researcher concluded that the
managers of the banks in the higher moral-hazard group have greater incentives to collimate
their interests to those of stockholders by taking on more risk as managerial ownership rises,
compared to the banks in the lower moral-hazard group, but only over the relatively period
1994-1997. So, manager agency problem of banks is characterized by changes in the holdings
or ownership structure when the banks have relatively higher moral-hazard incentives and
banking regulations are not strong. The model also revealed that the higher the risk taken, the
worse the performance of the bank is. Spatareanu (2005) investigated the relationship between
ownership concentration and performance, while also accounting for the effect of hostile
takeover threats on this relationship of UK, Czech and Polish public firms during 1999 by using
the Generalized Method of Moments (GMM) finding that concentration is insignificant in
explaining performance in both developed and transition countries.

[11]
 Love and Rachinsky (2008) in their paper investigate the connection between
ownership, corporate governance and operating performance in the banking sector for the
period 2003 – 2006. Their sample consists of 107 Russian banks and 50 Ukrainian banks.
Regression results showed some significant but economically unimportant relationship between
corporate governance and operating performance. Tandelilin et al. (2007) examined the
correlation among corporate governance, risk management and bank performance using a
sample of 51 Indonesian banks for the period 1999 – 2004. For the empirical study they used a
Triangle Gap Model with primary data analysis and secondary data analysis. This study
revealed that bank ownership affects both the relationship of corporate governance and bank
performance and corporate governance and risk management. It is worth mentioning that the
model used in this study found no linear effect of corporate governance on bank performance.

Sinha (2008), examined 40 Indian Commercial banks, both public and private using a
Radial DEA model for the period 2000 – 2001 to 2005 – 2006 comparing two variables: Total
Assets and Off Balance Sheet exposures in order to compare their performance. The results
were that for the period 2000 – 2001 public banks outperformed the private ones, while for the
period 2005 – 2006 private banks outperformed the public ones. The main disadvantage of this
research is that it did not use any qualitative factors. Hossain’s (2007) study revealed the level
and extent of the corporate governance disclosure of the banking sector in India. After collecting
the annual reports of 38 Indian banks for the year 2002-2003, he adopted a regression model to
investigate the relationship between corporate governance disclosure and various corporate
attributes such as size, profitability, ownership, listing, status, age etc. The conclusion was that
Indian banks have very high level of compliance in corporate governance disclosure, showing
that the attempt taken by the Indian authority to include ‘Corporate Governance Reporting’ in
the annual report is notable.

Kyereboah-Coleman and Biekpe’s (2006) study investigates the role of boards and
CEOs in the performance of the Ghanaian banking sector examining 18 banks both listed and
not – listed for the period 1997 – 2004 by adopting panel data to support their model. The
conclusion was that the more independent the board is, the worse the profitability of a bank.
Also, the regression results showed a positive relationship between the board size and ROA,
while on the other hand, they showed that CEO's tenure largely indicated a negative impact on
ROA. Goddard, Molyneux and Wilson (2004) investigated the profitability of European banks
during the 1990s. They used data of 665 banks from 6 European countries Denmark, France,
Germany, Italy, Spain and the UK, for the period 1992–1998. In their study, they created cross-

[12]
sectional, pooled cross-sectional time-series and dynamic panel models in order to identify
selected determinants of profitability. The result was that despite the high competition, which is
effective in eliminating abnormal profit, there is significant evidence of abnormal profit from year
to year. However, there is some variation between countries in effectiveness of competition in
eliminating abnormal profits.

Heiss and Koke (2004), investigated the determinants of changes in corporate

ownership and firm failure for German firms. This study included observations of 1510 German
firms for the period 1986 – 1995 for the variables: firm performance, capital structure, ownership
structure and firm size. The result was that many of the determinants of failure also affect
ownership changes in this bank-based economy, including poor performance, weak corporate
governance, high leverage, and small firm size. The ownership structure also plays a role for
both events. Grobi and Levratto (2008) examined, in a theoretical level, the impact of private
ownership on bank performance in Bulgaria and Hungary taking into account, also, principles of
corporate governance. The result was that in both transition countries private ownership plays a
crucial role, especially if it is combined with principles of good corporate governance, which
depend on accepted social norms derived from cultural value orientations, such as rule of law
and accountability. Concentrated on Bulgaria and Hungary, which expose differences on their
value orientations, Ingrid Größl and Nadine Levratto concluded that Bulgaria hindered the
privatization process of banks as a result of corruption, absence of law and maximization of
owners’ interest while Hungary, based more on Western system, supported the creation of
private banks.
Kapopoulos and Lazaretou (2007) used data of 175 greek listed firms in order to
investigate whether there is strong evidence that ownership structure affects firm’s performance,
measured by profitability. Empirical findings indicate that there is a positive relationship between
profitability and ownership structure in greek firms. Specifically, the results state that the greater
the degree to which shares are concentrated in outside or inside shareholders, the more
efficient the firm’s management and as a result the firm’s performance. Delfino (2007) examined
the impact of control changes (due to privatization, foreign acquisition and mergers and
acquisitions) on efficiency and productivity in Argentina’s banking sector. Specifically, she used
panel data for the period 1993 – 2000 in order to construct the regression model and came to
the conclusion that state – owned banks were less efficient than private ones, bank privatization
provided only short – term efficiency gains, foreign acquisitions led to stronger productivity

[13]
performance of acquired banks, though it did not affect efficiency, and finally, mergers and
acquisitions had a negative impact on bank’s performance.
Roe (2004) in his paper outlined the institutions of corporate governance in the West. In
particular, institutions face two problems: vertical governance (between distant shareholders
and managers) and horizontal governance (between close, controlling shareholder and distant
shareholder). Biswas and Bhuiyan (2007) examined, in a theoretical level, the impact of
corporate governance on firm performance. The analysis of the OLS regression indicated
confusion in identifying the direction of causality between corporate governance and firm
performance. In their paper, Hassan, Wolfe and Maroney (2004) presented the agency
problems of the banking sector based on a corporate governance literature review. They found
that in developing countries corporate governance is rather weak due to the information
asymmetries, agency problems, political corruption and absence of stable accounting practices,
which negatively affect all companies’ participants and especially stakeholders.

Rose (2007) used a sample of all Danish firms listed at the Copenhagen Stock
Exchange for the period 1998 – 2001 excluding banks and insurance companies in order to
examine whether ownership affects firm’s performance, measured by Tobin’s q. The cross –
sectional regression analysis showed that increased ownership by institutional investors did not
have an impact firm’s performance. However decomposing the results, it was evident that
ownership by banks had a positive significant impact on performance. Barako and Tower (2007)
investigated the association between ownership structure and bank performance in Kenya.
Their empirical analysis included all financial institutions operating in Kenya and ran a
multivariate regression with variables referring to ownership, bank size and ROA. The results
provided a strong support that ownership structure influence bank performance. Specifically,
board ownership is significantly and negatively associated with performance, institutional
shareholders have no significant influence on performance and foreign ownership has a
significant positive impact o bank’s performance.

Hallward-Driemeier et al. (2006) made a research on 1,500 Chinese enterprises in five

cities in order to investigate the components of the investment climate and their effects on firm
performance. The survey revealed that both ownership and investment climate measures
influence firm performance and more specifically productivity and growth. In particular, having in
their regression firm performance as the dependent variable, they found that it is positively
correlated with foreign and domestic private ownership, light regulatory burdens, limited
corruption, technological infrastructure and labour market flexibility. In order to investigate the

[14]
impact of the diversity of foreign institutional and foreign corporate shareholders on the
performance of emerging market firms, Douma et al. (2006) used financial data of 1005 firms
belonging to the financial year 1999–2000 from different industries. They run a regression taking
as the dependent variable corporate performance, measured by ROA and q ratio. The study
revealed the necessity to separate foreign ownership into foreign institutional and foreign
corporate shareholdings. It also showed the way that foreign institutional investors affect firm
performance is ambiguous. As for the outside domestic shareholders, it was proved that
domestic corporations influence firm performance in a positive way; however, the coefficients of
the regression were less significant compared to foreign corporations.

Another case based on firm performance was that of Lensink and Naaborg (2007). They
examined the impact of a rise in foreign ownership on banks’ interest revenues and profitability
using panel data of 511 banks from 73 countries worldwide. They applied for their estimations
the generalized methods of moments (GMM) technique and they found that a rise in foreign
ownership negatively affects bank performance, particularly the net interest margin and bank
profits, providing evidence for the “home field advantage theory”. In contrast, banks with a
limited degree of foreign ownership provide greater profitability and ability to increase more net
interest revenues. Mahajan and Lummer (1993) analyzed in their paper how an announcement
of changes in senior corporate management and specifically in the case of loss of decision
making power of corporate managers affect the direction and magnitude of changes in stock
prices and stockholder wealth. They collected data for the 800 highest paid executives
employed by US corporations for the period 1972-1983 in order to support their model. The
results of their investigation were in a line with their initial three hypotheses. Firstly,
management departures cause instability, which adversely affects corporate performance.
Secondly, decisions made by the corporate group in order to reshape the management team is
not in the interest of the shareholders, while in time of management change shareholders credit
that the existing management team is not acting in their interest, so, there is a need to replace
it. Finally, shareholders have the conviction that executive changes caused by factors external
to the company are preferable to internally generated changes.

Finally, Capon, Farley and Hoenig (1990) summarized in their paper a meta-analysis of
statistical results in the literature on industry, firm and business financial performance. They
collected 320 empirical studies published between 1921 and 1987 and they presented their
results. They used 2 forms of meta-analysis: counting the occurrence of qualitative relationships
and ANCOVA of regression coefficients associated with 8 frequently-studied causal variables.

[15]
Most of the studies under examination indicated positive and significant relationship between
firm performance and industry concentration, growth in sales and assets, capital investment
intensity (of industry) and advertising. On the other hand, they found negative relationship
between firm performance and debt, while not significant relationship was spotted between firm
performance and size (measured by assets and sales) and firm control (owners vs managers).

In summary, it is not feasible to accept one general conclusion for the relationship
between firm performance and corporate governance. However, empirical results show that
generally ownership structure affects significantly corporate performance. More specifically,
ownership concentration does not have any impact on firm’s performance, in addition to
independent ownership, which has a negative impact on profitability and as a result on
performance. Moreover, it is stated that weak corporate governance leads to poor corporate
performance. As for the banking sector, there are mixed and ineffective results about the link
between performance and corporate governance. In general, ownership structure affects bank
performance. More analytically, there are cases where foreign ownership has a negative impact
on bank’s performance, while in other cases the addition of one foreign director affects
positively the performance, but the addition of more than one foreign director does not improve
it. Furthermore, it is proved that institutional directors do not affect bank’s performance.

As we put forward the perspectives on the impact of corporate governance on bank

performance we should attempt to elaborate on the rationale related with the corporate structure
in a developed market like Europe and America. It is likely that an effective corporate structure
would have a more efficient operating strategy, which would lead to increased profitability and
performance.

V. DATA AND METHODOLOGY

In order to examine the role of corporate governance in bank performance, we have taken a
sample of 79 big banks globally that had a corporate governance rating from 2004 to 2008
according to database Corporate Governance Quotient by RiskMetrics Group. Corporate
Governance data were estimated by using public available documents and website disclosure
on 63 different items referring to the structure of board of directors, audit quality, anti-takeover
mechanisms and compensation / ownership. The accounting data used in our models were

[16]
extracted from Thomson One Banker database. Specifically, our sample contains 38 European
banks, 28 Asian banks (mostly Japanese), 9 American banks and 3 Australian banks.
Unfortunately, we had to exclude a lot of banks because of lack of all the appropriate
information for our model.

Taking into account past literature, we have used as proxies for bank performance the
following variables:

Table 2: Dependent Variables

Variable Measurement
ROE Bank’s Net Income plus Depreciation divided by its book value
of shareholder’s Equity
ROA Bank’s Net Income plus Depreciation divided by its book value
of Total Assets
P/E Bank’s market value per share divided by its earnings per share
Investment Return Market Price Year End + Dividends Per Share + Special
Dividend -Quarter 1 + Special Dividend-Quarter 2 + Special
Dividend-Quarter 3 + Special Dividend-Quarter 4) / Last Year's
Market Price-Year End - 1) *100

We have formulated an Ordinary Least Squared model in order to examine the

correlation between corporate governance and bank performance, with these four dependent
variables that measure performance.

Performance = f (Size, Volatility, Sales growth, Assets growth, Leverage, Age, CGQ,
Y2004, Y2005, Y2007, Y2008, Asia, America, Australia, Ins Trading) + ε

As independent variables we have used the following:

Table 3: Independent Variables

Variable Measurement
Size Logarithm of bank’s Total Assets

[17]
Volatility Standard deviation of stock return
Sales Growth Net sales / Revenues – 1year annual growth
Assets Growth Total assets – 1 year annual growth
Leverage Debt to Total Assets
Age Years since the date of establishment
CGQ Industry Corporate Governance Rating
Y2004 A dummy variable that takes a value of 1 if the observation is in
year 2004 and 0 otherwise
Y2005 A dummy variable that takes a value of 1 if the observation is in
year 2005 and 0 otherwise
Y2007 A dummy variable that takes a value of 1 if the observation is in
year 2007 and 0 otherwise
Y2008 A dummy variable that takes a value of 1 if the observation is in
year 2008 and 0 otherwise
Asia A dummy variable that takes a value of 1 if the observation
belongs to an Asian bank and 0 otherwise
America A dummy variable that takes a value of 1 if the observation
belongs to an American bank and 0 otherwise
Australia A dummy variable that takes a value of 1 if the observation
belongs to an Australian bank and 0 otherwise
Ins Trading Number of closely held shares / Common shares outstanding) *
100
ε Error term

Previous empirical work includes similar control variables. However, we have included year
dummy variables and continent dummy variables in order to examine their correlation with bank
performance.

Table 4: Descriptive statistics for all variables

Variable Mean Median Maximum Minimum Std. Observations

Deviation

ROE 0.143898 0.145868 1.049035 -2.392000 0.091361 268

ROA 0.010981 0.010004 0.168988 -0.042100 0.011490 268
P/E 18.49382 13.76763 810.1000 0.000000 49.20785 268
InvestReturn 0.070510 0.110050 1.561800 -4.820.000 0.462465 268
InsTrading 25.63713 21.29000 96.42000 0.010000 22.51987 268
Volatility 22.95761 20.84000 57.94000 11.77000 8.439332 268
Leverage 22.87567 18.03500 222.9200 0.160000 23.06824 268
AssGrowth 12.48108 9.245000 117.7900 -8.960.000 14.90988 268

[18]
SalesGrowth 12.74104 11.21000 84.57000 -2.737.000 16.17904 268
CGQ 56.17873 59.20000 100.0000 1.400000 29.26405 268
Age 114.5373 91.00000 537.0000 5.000000 90.96842 268
Size 5.091082 5.036253 6.412220 3.605227 0.580159 268

This table contains the descriptive statistics for the variables that investigate the effect of
different financial elements on bank performance. It presents means, medians, maximums,
minimums and standard deviation of all the variables. The sample consists of European,
Australian, American and Japanese banks for the period 2004 to 2008. These banks on
average have not been performing quite well with an annual average ROA of 1% and an
average ROE of 1,4%. The maximum of ROA is 16,8% with a minimum of -4,21%, while the
maximum of ROE is 104,9% and the minimum -239,2% . The same exists for Investment Return
which fluctuates in low levels of 7%. On average, Ins Trading accounted for about 25,6% but it it
varies from 96,4% to 0,01 %. Most of the banks have their assets financed with debt rather than
equity. With regard to assets and sales growth, the mean numbers of about 12,6% are quite
positive. Corporate governance rating seems to be quite high with an average of 56,2%. The
size of the banks varies, indicated by their asset base, as well as their age, which varies
between 5 to 537 years.

Table 5 presents the correlation coefficients of the variables used in our models. This
symmetric matrix measures correlation on a scale with 1 indicating a perfect positive correlation,
zero no correlation and -1 perfect negative correlation. According to our data, performance
variables (ROE, ROA, P/E and Investment return) do not seem to have a very significant
relationship with our control variables and especially CGQ, which is the main variable under
investigation. Analytically, we observe that Sales Growth seems to have a relationship with all
four dependent variables, while Assets Growth, Leverage and Size have a positive relationship
with both ROE and ROA. Specifically for the Corporate Governance rating variable we observe
that it has a positive correlation with ROE and ROA and a negative correlation with Investment
Return, but without ensuring the statistical dependence of these variables, which we are going
to examine with the following regression analysis.

[19]
Table 5: Correlation matrix

Age AssGrowth CGQ InsTrading Leverage P/E InvestReturn ROA ROE SalesGrowth Size Volatility

Age

1.000000
AssGrowth

0.189562 1.000000
CGQ

0.375560 0.292145 1.000000

InsTrading

-0.126426 -0.144278 -0.274059 1.000000

Leverage

0.135291 0.203460 0.090109 0.162902 1.000000

P/E

-0.041987 -0.057755 -0.026548** -0.026211 -0.070646** 1.000000

InvestReturn

-0.054111 -0.016822** -0.110227 0.078009*** 0.026987 -0.070972 1.000000

ROA

0.110069 0.108811 0.210856 0.006593** 0.176740* -0.096414 -0.041583 1.000000

ROE

0.092174 0.123921 0.228835 0.028770 0.250026* -0.123665 0.212480 0.254064* 1.000000

SalesGrowth

-0.009995 0.291358 0.064298 0.087036 0.161469 -0.134548 0.178340 0.137053 0.414375 1.000000
Size

0.323868 0.323002 0.283114 -0.164013 0.212125 -0.106502* -0.171959 0.158457 0.191632 0.062544 1.000000
Volatility

-0.100117 0.066790 -0.109922 0.071868 0.145791 0.037699*** -0.036828 -0.027551 0.227264** 0.047866 0.152200 1.000000
* 1%, ** 5%, *** 10% confidence level

[20]
 VI. EMPIRICAL RESULTS

The empirical investigation includes four regressions with four different proxies for bank
performance (ROE, ROA, P/E and Investment return) in order to trace any impact of corporate
governance on bank performance. Below, we present the tables and the statistical analysis of
the estimated equations.

Table 6: Model Summary of ROE regression

Variable Predicted β Std. error t-statistic Sig. Tolerance VIF

sign
(Constant) (+) 0,121 0,121 0,999 0,319 - -
Size (+) 0,055 0,057 0,952 0,342 0,635 1,574
Volatility (+) 0,122 0,055 2,236 0,026 0,673 1,486
SalesGrowth (+) 0,169 0,054 3,159 0,002 0,706 1,416
AssGrowth (+) 0,089 0,056 1,586 0,114 0,609 1,642
Leverage (+) 0,175 0,062 2,823 0,005 0,475 2,104
Age (+) 0,015 0,055 0,279 0,781 0,621 1,610
CGQ (+) 0,069 0,058 1,195 0,233 0,609 1,643
Y2004 (-) -0,095 0,148 -0,643 0,521 0,577 1,732
Y2005 (-) -0,195 0,138 -1,409 0,160 0,624 1,603
Y2007 (-) -0,136 0,134 -1,011 0,313 0,618 1,618
Y2008 (-) -0,607 0,142 -4,269 0,000 0,559 1,788
Asia (+) 0,015 0,140 0,104 0,917 0,396 2,523
America (+) 0,612 0,181 3,379 0,001 0,583 1,716
Australia (+) 0,561 0,206 2,723 0,007 0,849 1,178
InsTrading (+) 0,017 0,053 0,319 0,750 0,699 1,430

Adjusted R2 0,303

Std. Error of the estimate 0,714

F 8,868

F – Significance 0,000

The R-squared statistic measures the success of the regression in predicting the values
of the dependent variable within the sample. In standard settings may be interpreted as the
fraction of the variance of the dependent variable explained by the independent variable. The
statistic will equal one if the regression fits perfectly, and zero if it fits no better than the simple
mean of the dependent variable. Specifically, the adjusted R-squared is a modification of R-

[21]
squared that adjusts for the number of explanatory variables in the model, in other words it
takes into account the number of control variables. According to Table 6, adjusted R –squared
of ROE regression is equal to 0,303, which does not indicate a very good model because only
30,3% of the dependent variable is explained from the control variables. The explanatory
variables included in this regression are not very good predictors of Return on Equity.

The standard error of the regression is a summary measure based on the estimated
variance of the residuals. The standard error of ROE equation is 0,714 indicating small
statistical noise in the estimates. Table 6 presents the F-statistic probability of ROE regression,
which is equal to 0, so we can reject the null hypothesis that all slope coefficients excluding the
constant are zero with 5% significance level. As a result, the model is considered significantly
better than would be expected by chance and there is linear relationship of ROE to the
independent variables.

Table 6 presents, also, the statistical analysis of the control variables included in our
equation. According to t-statistics, which are the ratios of the estimated coefficient to its
standard error, there are six statistically significant coefficients in a 5% confidence interval:
Volatility, Sales growth, Leverage, Y2008, America and Australia. These coefficients have a t –
statistic bigger than 1,96 in absolute value and also a p – value very close to zero, so we can
reject the null hypothesis that these slope coefficients are zero with 5% confidence level. The
CGQ variable has a t – statistic equal to 1,195 and a p – value equal to 0,233 leading to the
conclusion that this coefficient is not statistically significant neither at a 10% significance level
and we cannot reject the hypothesis that it is equal to zero, in other words, corporate
governance rating does not have any impact on Return on Equity. With regard to the 1%
confidence interval, there are 5 statistically significant variables: Leverage, Sales Growth,
Y2008, Australia and America indicating a stronger relationship. Meanwhile, there is a positive
correlation (according to the coefficients) between each of the following variables: Volatility,
Sales growth, Leverage, America and Australia and Return on Equity and a negative correlation
between Y2008 and Return on Equity, which can be interpreted as a -0,607 decrease in Return
on Equity when the dummy variable for 2008 increases by one unit.

With regard to multicollinearity, both tolerance and VIF give the same result that there is
no correlation between the independent variables, provided that none of the variables has
tolerance close to zero or VIF > 10.

[22]
Table 7: Model Summary of ROA regression

Variable Predicted β Std. error t-statistic Sig. Tolerance VIF

sign
(Constant) (+) 0,148 0,115 1,279 0,202 - -
Size (+) 0,030 0,055 0,548 0,584 0,635 1,574
Volatility (+) 0,042 0,052 0,803 0,423 0,673 1,486
SalesGrowth (+) 0,182 0,051 3,555 0,000 0,706 1,416
AssGrowth (+) 0,057 0,053 1,069 0,286 0,609 1,642
Leverage (+) 0,279 0,059 4,701 0,000 0,475 2,104
Age (-) -0,053 0,052 -1,006 0,315 0,621 1,610
CGQ (+) 0,074 0,055 1,340 0,181 0,609 1,643
Y2004 (-) -0,096 0,142 -0,680 0,497 0,577 1,732
Y2005 (-) -0,204 0,132 -1,546 0,123 0,624 1,603
Y2007 (+) 0,199 0,128 1,553 0,122 0,618 1,618
Y2008 (-) -0,029 0,136 -0,215 0,830 0,559 1,788
Asia (-) -0,229 0,134 -1,710 0,089 0,396 2,523
America (+) 0,449 0,173 2,594 0,010 0,583 1,716
Australia (+) 0,784 0,197 3,989 0,000 0,849 1,178
InsTrading (+) 0,119 0,050 2,374 0,018 0,699 1,430

Adjusted R2 0,385

Std. Error of the estimate 0,682

F 12,352

F – Significance 0,000

According to the Adjusted R–squared of Table 7, the model explains 38,5% of the
variability of the dependent variable ROA. This means that the regression line does not
approximates very well the real data points and thus, the model can predict less of the
movements of ROA. Standard error of 0,68 shows that the coefficient estimate is reliable, or that
it has a small variability, which means that there are not many extreme prices in the model and
thus, the trend is strong. It, also, presents the probability of F-statistic which is 0, meaning that
the explanatory variables do have impact on the dependent variable.

With regard to the ROA regression output, there is a statistically positive relationship
between the Sales growth and Leverage with ROA at a 1% significance level, a positive
relationship between Ins Trading and ROA at a 5% significance level and a weaker negative

[23]
correlation between the dummy referring to Asia and ROA at a 10% confidence level. So, ROA
and thus performance is positively affected by the increase in sales over a specific period of
time, by the debt used to finance the assets of the company and by the voting stocks held by a
small number of shareholders of the company. Also, the dummy variable AUSTRALIA and the
dummy variable AMERICA are statistically significant with probability almost zero. The two
dummies have positive coefficient, 0,449 and 0,784 respectively. This means that when the
dummy AUSTRALIA increases by 1 the dependent variable ROA is 0,449 units more than the
average in the other countries. Similarly, when the dummy AMERICA takes the value of 1, ROA
increases by 0,784 units. With regard to Asia, when the dummy takes a price of 1 decreases the
value of ROA by 0,229.

As for the Corporate Governance variable (which includes factors as board, ownership,
etc), it is not statistically significant, having no real impact on ROA. Checking for
multicollinearity, the model is uncharged from correlation between the variables, using the
common rule of thumb that only tolerance close to zero and VIF higher than 10 indicate a
multicollinearity problem. Individual Standard errors are close to zero, reporting small variability
of the model.

Table 8: Model summary of P/E regression

Variable Predicted β Std. error t-statistic Sig. Tolerance VIF

sign
(Constant) (+) 0,151 0,128 1,179 0,240 - -
Size (-) -0,311 0,061 -5,081 0,000 0,647 1,546
Volatility (-) -0,103 0,058 -1,785 0,075 0,680 1,471
SalesGrowth (-) -0,089 0,057 -1,561 0,120 0,710 1,408
AssGrowth (+) 0,101 0,062 1,631 0,104 0,586 1,706
Leverage (-) -0,155 0,067 -2,311 0,022 0,474 2,109
Age (+) 0,048 0,059 0,818 0,414 0,616 1,623
CGQ (-) -0,136 0,062 -2,206 0,028 0,608 1,645
Y2004 (-) -0,133 0,157 -0,847 0,398 0,579 1,727
Y2005 (-) -0,162 0,146 -1,108 0,269 0,626 1,597
Y2007 (-) -0,258 0,143 -1,807 0,072 0,624 1,602
Y2008 (-) -0,554 0,153 -3,620 0,000 0,574 1,741
Asia (+) 0,418 0,150 2,787 0,006 0,393 2,542

[24]
America (-) -0,214 0,192 -1,113 0,267 0,581 1,723
Australia (+) 0,329 0,218 1,509 0,133 0,848 1,179
InsTrading (-) -0,019 0,057 -0,327 0,744 0,703 1,423

Adjusted R2 0,330

Std. Error of the estimate 0,755

F 9,775

F – Significance 0,000

According to Table 8, the P/E regression has an adjusted R-squared equal to 0,33,
which indicates that only 33% of the Price to Earnings ratio can be explained by the control
variables of our estimated equation. As a result, our regression model cannot predict the
movements of the P/E ratio. The standard error of the model is 0,756, which states the lack of
statistical noise in our estimates. The F-statistic probability presented in Table 8, is zero leading
to the rejection of the null hypothesis that all slope coefficients, except for the constant, are
equal to zero and to the conclusion that the explanatory variables do have impact on the P/E
ratio.

Table 8 presents, also, the statistical analysis of our control variables. Studying the t-
statistics, it is obvious that there are only three out of fifteen control variables that have a price
higher than 2,56 in absolute value and are statistically significant with 1% significance level.
These variables are: Size, Y2008 and Asia and have a p – value smaller than 0,05 leading to
the rejection of the null hypothesis that their slope coefficients are equal to zero. Meanwhile,
with 95% confidence interval, there is statistical significance between Leverage, CGQ and P/E.
Specifically, the Corporate Governance rating variable has a coefficient equal to -0,136
indicating a negative correlation between Corporate governance and P/E ratio, where an
increase of Corporate governance rating by one unit leads to the decrease of P/E ratio by 0,136
units. With 10% significance level we found two statistically significant variables Volatility and
the dummy variable referring to 2007 observations. With regard to the statistically significant
dummy variables, Y2008 has a coefficient equal to -0,554 stating that the observations of 2008
have a P/E ratio decreased by 0,554 in comparison with the previous four years, Y2007 has a
coefficient equal to -0,258 indicating that the observations of 2007 have a P/E ratio decreased

[25]
by 0,258, while the variable Asia has a coefficient equal to 0,418 indicating that Asian banks
present higher P/E ratios than the other countries by 0,418.

Individual standard errors are small, as it was detected also in the previous equations,
indicating a small variability of the estimates. Checking for multicollinearity, none of the
variables have tolerance close to zero or VIF>10 indicating that there is no multicollinearity
problem.

Table 9: Model summary of INVESTMENT RETURN regression

Variable Predicted β Std. error t-statistic Sig. Tolerance VIF

sign
(Constant) (+) 0,498 0,101 4,950 0,000 - -
Size (-) -0,027 0,048 -0,570 0,569 0,634 1,577
Volatility (-) -0,035 0,045 -0,774 0,439 0,674 1,484
SalesGrowth (+) 0,037 0,045 0,817 0,415 0,713 1,403
AssGrowth (+) 0,109 0,047 2,346 0,020 0,609 1,641
Leverage (-) -0,048 0,052 -0,924 0,356 0,475 2,104
Age (-) -0,014 0,046 -0,316 0,752 0,619 1,615
CGQ (-) -0,053 0,048 -1,112 0,267 0,609 1,641
Y2004 (-) -0,035 0,123 -0,282 0,778 0,577 1,732
Y2005 (-) -0,352 0,115 -3,060 0,002 0,625 1,601
Y2007 (-) -1,097 0,112 -9,827 0,000 0,619 1,616
Y2008 (-) -1,851 0,118 -15,659 0,000 0,566 1,767
Asia (+) 0,199 0,117 1,697 0,091 0,391 2,554
America (+) 0,256 0,150 1,702 0,090 0,583 1,716
Australia (+) 0,195 0,171 1,142 0,254 0,849 1,178
InsTrading (+) 0,078 0,044 1,777 0,077 0,699 1,431

Adjusted R2 0,608

Std. Error of the estimate 0,593

F 28,983

F – Significance 0,000

[26]
 The table above presents the regression results having as dependent variable the
Investment return. The value of Adjusted R-squared is 0,608, so, the companies’ performance
based on Investment return is affected by the response variables more than the half. The
variability of the model is small as the standard error is close to zero. F-statistic gives a high
value of 28,98, with p-value almost zero, which leads us to reject the hypothesis that all slope
coefficients are zero.
The table indicates that Assets growth is statistically significant, having positive relation
with Investment Return, while p-value does not overcome the 5% significance level. The dummy
variables for years 2005, 2007 and 2008 are also significant with 1% significance level
indicating a very strong correlation. With 90% confidence interval, we observe a weaker
correlation between the variables Asia, America, Ins Trading and Investment Return. Because
of the negative coefficients we can say that for year 2005 Investment Return was 0,352 less
than the average of the other years. In Year 2007 Investment Return was 1,097 less than the
average of the other years and in Year 2008 it was decreased by 1,851.
Corporate governance does not have strong relationship with Performance, as was
detected before, which means that corporate governance’s elements do not affect Investment
return. Multicollinearity is absent in this regression too, with tolerance not close to zero and VIF
below 10. Standard errors of each variable are small, supporting the statistical reliability of the
coefficients.

The overall findings, so far, suggest that although Corporate Governance relationship
with bank performance seems to be non – existent, it is clear that there is a strong relationship
of bank performance with a number of other financial variables. Specifically, we observe
significant correlation of our performance variables with Leverage and Sales Growth as it was
expected. This relationship is based on the fact that ROA, ROE, P/E, Leverage and Sales
Growth are all financial ratios unbreakably connected with each other. Size, which was
measured by the logarithm of Total Assets, is statistically significant only with P/E ratio without
reporting any correlation with the other performance variables. This result is in accordance with
the study of Goddard, Molyneux and Wilson (2004), which concluded that the relationship
between European bank’s size and performance was unconvincing.

It is worth mentioning that Age had no impact on bank performance neither at a 10%
confidence level, showing the neutrality of performance against bank’s age. On the other hand,
Volatility of the stock price seems to affect bank’s performance due to the fact that it has a small
but significant correlation with Return on Equity and P/E ratio. Without any previous studies

[27]
confirming our results, we reported that the higher the Volatility of the stock the higher the
bank’s ROE and the lower the P/E ratio. Assets Growth is related only with Investment Return in
a positive manner in our sample, which was not expected taking into account the study of
Lipson and Schill (2008) that reported a negative correlation between firm Asset Growth and
Stock Return.

A positive relationship between ROE, ROA and the developed continents is also noticed:
America and Australia indicating that American and Australian banks outperformed the
European. With regard to Australian banks’ outperformance, it is owed to the better structure of
the Australian financial system and its fundamental differences of the American and the
European financial systems (Ord Minnett Research, 2008). Without having any previous
studies examining such a relationship in order to compare our results with, we can assume that
this outcome indicates that the economic development leads to a better structured banking
system, which affects positively bank’s performance.

With regard to the year dummies used in our regression models, we found a strong
negative relationship between year 2008 and ROE, P/E and Investment Return. This was a
presumable result taking into account the fact that in the middle of 2007 the global economy
entered the financial crisis, which still exists. The recent credit crunch created severe difficulties
especially on the banking sector, leading to the collapse of many solvent banks (Moxey and
Berendt, 2008). It was, also, expected that the year dummies would have a strong correlation
with the Investment Return, which became very negative as we entered the financial crisis that
led to its galvanic decrease worldwide. We, also, observed that in 2005, 2007 and 2008 banks
underperformed in comparison to 2006, with regard to their Investment Return. While in 2008
they underperformed 2006 with regard to ROE and P/E ratio, which can be explained, bearing in
mind that in 2006 the international banking system reached its pick.

Concentrating on the relationship of inside trading and bank performance, our regression
models showed a significantly positive correlation with Return on Assets, which is consistent
with the study of Kyereboah – Coleman and Biekpe (2006) that showed a negative correlation
between independent board and profitability. In contrast, our results contradict the study of
Barako and Tower (2007), which found no significant relationship between insider shareholders
and bank performance. The positive association between this variable and bank performance
indicate a proactive and objective interaction between insider shareholders and the bank in
which they are invested.

[28]
 Taking into account Corporate Governance Rating, which is the main variable under
investigation, we observed that it has no significant relationship with ROE, ROA and Investment
Return, in contrast with P/E ratio, where there is a strong negative relationship. Previous
literature results indicate confusion in identifying the direction of causality between corporate
governance and bank performance. Analytically, Love and Rachinsky (2008) found significant
but no economically important relationship between corporate governance and bank
performance in Russia and Ukraine using as dependent variables ROA and ROE, while,
Tandelilin et al. (2007) concluded that there is no linear relationship between these variables.
Our results are, also, in consistence with the theoretical study of Hassan, Wolfe and Maroney
(2004), which states that Corporate Governance has a weak effect in developed and developing
countries, like the ones used in our models because of the asymmetry of information, political
corruption, inexperienced market participants and weak judicial systems. They report that the
ineffectiveness is even larger in financial institutions like banks because of the large number of
their stakeholders and the greater systemic risks that they face. As a result, corporate
governance structures cannot contribute on bank’s stability and performance.

VII. CONCLUSION

The economic crisis that began in 2007 resulted in many firm closures including banks
all over the world. The cause of the financial breakdown may be owed to the fact that boards of
large institutions failed to understand and react properly and immediately to emerging risks
(Lloyd, 2009). Stability and profitability of banking sector started to shiver, putting threats to the
global economy. This paper presents evidence between bank performance and different
financial measures, including corporate governance rating, from 2008 to 2009. The investigation
was based on financial elements of 72 banks from Europe, Canada, Australia and Japan.

Evidence witnesses an inconsiderable relationship between bank performance and

corporate governance rating. We observed that while corporate governance did not have any
impact on ROA, ROE and Investment return, a strong negative relation between corporate
governance and P/E ratio exists. These results show largely that bank performance does not
conform to Corporate Governance and remains negatively or negligibly affected. Despite the

[29]
fact that the issue of corporate governance is very popular during last years, it has a secondary
effect on performance. Our results were in line with previous studies, which found insignificant
and obscure relationship of bank performance and corporate governance. As it was expected
we found strong correlation between our dependent variable, Leverage and Sales Growth as
these two variables are inextricable with bank performance. We also found that economically
developed continents like America and Australia show higher levels of bank performance.

The regression results showed also a strong negative association between bank
performance and year 2008. As it was described above, recent financial crisis affected all types
of financial institutions, including banks, leading to a decline on performance and therefore on
return. Finally, we observed a positive relationship between inside trading and performance
indicating that the more shares are held by insiders like officers, directors and large
shareholders the better the performance is.

The limitations of this paper extend to those associated with the issue of data, which led
to a limited sample of banks due to the fact that there was no corporate governance data
available for the period 2004 – 2008 for more than the 79 banks included in our research. The
use of a bigger sample of banks globally would have been more representative of the global
banking system and it would have been more consistent leading to more solid and informative
empirical results. However, we managed to include banks with a wide range of corporate
governance rating making our results sufficient. It is, also, worth mentioning that all the data
used in our research are reliable and trustworthy, taking into account that they were pooled from
two very reliable databases: Thomson One Banker and Corporate Governance Quotient.

This exercise is the first one, to our knowledge, that relates bank performance with
Corporate Governance during the financial crisis of 2007, using financial elements from banks in
4 different continents. However, in this paper we examine only one aspect of performance
using measures such as ROA, ROE, Investment return and P/E ratio. Further research is
required to investigate Corporate Governance as a means of banks to provide capital flow in
order to prevent dangerous market conditions and provide financial stability.

[30]
 VIII. REFERENCES

Articles and Papers

Barako Dulacha G. and Tower Greg (2007), “Corporate Governance and bank performance:
Does ownership matter? Evidence from the Kenyan banking sector”, Corporate
Ownership and Control Volume 4, Issue 2, Winter 2006 – 2007

Capon Noel, Farley John U. and Hoenig Scott (1990), “Determinants of financial performance: A
meta – analysis *”, Management Science Vol. 36, No. 10, October 1990

Delfino Maria Eugenia (2007), “Control Changes and Firm Performance in Banking”, Int. J. of
the Economics of Business, Vol. 14, No. 2, pp. 261 - 281, July 2007

Garneau Victoria Shahid Abdus (2009), “The Sarbanes – Oxley Act of 2002 (SOX): A redundant
regulation for the banking industry”, Journal of Banking Regulation Vol. 10. 4

Goddard John, Molyneux Phil and Wilson John O. S. (2004), “The profitability of European
banks: A cross-sectional and dynamic panel analysis”, Blackwell Publishing Ltd and The
Victoria University of Manchester, 1463–6786 363–381, 2004.

Größl Ingrid and Levratto Nadine (2008), “The role of social norms for the relationship between
private ownership and the performance of banking sectors in transition countries: the
cases of Bulgaria and Hungary as examples”, Corporate Ownership and Control/Volume
5, Issue 2, 68-86, Winter 2008.

Hallward-Driemeier Mary, Wallsten Scott and Xu Lixin Colin (2006), “Ownership, Investment
climate and Firm performance”, Economics of Transition Volume 14(4), 629–647, 2006.

Hassan M. Kabir, Wolfe David R., Maroney Neal C. (2004), “Corporate control and Governance
in banking”, Corporate Ownership & Control / Volume 1, Issue 4, 94-107 Summer 2004.

Heiss Florian and Koke Jens (2004), “Dynamics in Ownership and Firm Survival: Evidence from
Corporate Germany”, European Financial Management, Vol. 10, No. 1, 167–195, 2004.

Hensher and Shirley (2009), “Was Basel II really faulty?”, Financial Management

Hossain Mohammed (2007), “The corporate governance reporting exercise: The portrait of a
developing country”, International Journal of Business Research, Volume VII, Number 2,
2007.

Kapopoulos Panayotis and Lazaretou Sophia (2007), “Corporate Ownership Structure and Firm
Performance: evidence from Greek firms”, Corporate Governance, Volume 15, 144-158,
Number 2 March 2007.

Kyereboah-Coleman Anthony and Biekpe Nicolas (2006), “Do boards and CEOs matter for bank
performance? A comparative analysis of banks in Ghana”, Corporate Ownership and
Control/ Volume 4, Issue 1, Fall 2006.

Lee Seok Weon (2005), “Moral hazard, agency problem and ownership structure”, Corporate
Ownership and Control/Volume 3, Issue 2, 116-124, Winter 2005-2006.

[31]
Lensink Robert and Naaborg Ilko (2007), “Does foreign ownership foster bank performance?”,
Applied Financial Economics, 17, 881–885, 2007.

Lipson Marc L. Mortal Sandra and Schill Michael J. (2008), “What explains the Assset Growth
Effect in Stock Returns?”

Lloyd Bevan (2009) “Governance Models Lessons from the Global Crisis?”, Accountancy
Ireland Vol. 41

Love Inessa and Rachinsky Andrei (2007), “Corporate Governance, Ownership and Bank
Performance in emerging markets: Evidence from Russia and Ukraine”

Lskavyan Vahe and Spatareanu Mariana, “Ownership concentration, market monitoring and
performance: evidence from the UK, the Czech Republic and Poland”, Journal of Applied
Economics. Vol IX, No. 1, 91-104, May 2006.

Mahajan Arvind and Lummer Scott (1993), “Shareholder wealth effects of management
changes”, Journal of Business Finance & Accounting, 20(3), 0306-686X, April 1993.

Moxey Paul and Berendt Adrian (2008), “Corporate Governance and the Credit Crunch”, The
Association of Chartered Certified Accountants

Panagopoulos Yannis and Vlamis Prodromos (2009), “Bank Lending, Real Estate Bubbles and
Basel II”, Journal of Real Estate Literature Vol. 17

Rose Caspar (2007), “Can institutional investors fix the corporate governance problem? Some
Danish evidence” , J. Manage Governance (2007) pp. 405 – 428

Sinha Ram Pratap (2008), “Ownership and Efficiency: A Non-Radial Bilateral Performance
Comparison of Indian Commercial Banks”, The Icfai University Journal of Managerial
Economics, Vol. VI, No. 4, 67-81, 2008.

Sytse Douma, Rejie George and Rezaul Kabir (2006), “Foreign and Domestic ownership,
Business groups, and Firm performance: evidence from a large emerging market”,
Strategic Management Journal, 27: 637–657, 2006.

Tandelilin Eduardus, Kaaro Hermeindito, Mahadwartha Putu Anom and Supriyatna (2007),
“Corporate Governance, Risk Management and Bank Performance: Does type of
ownership matter?”, Final report of an EADN individual research grant project No. 34

Books

Casu Barbara, Girardone Claudia, Molyneux Philip (2006), “Introduction to banking”, Prentice
Hall Financial Times

Clarke Thomas (2007), “International Corporate Governance- A comparative approach”,

Routledge

Frederikslust Ruhd A.I., Sang James, Sudarsanam P. Sudi (2008),“Corporate Governance and
Corporate Finance”, Routledge

[32]
Federowicz Michal and Aguilera Ruth V. (2003), “Corporate governance in a changing
economic and political environment”, Palgrave Macmillan

Howells Peter and Bain Keith (2005), “The Economics of Money, Banking and Finance - A
European text”, Prentice Hall Financial Times.

Pierre L. Siklos (2002), “The changing face of central banking”, Cambridge University Press.

Websites

Arun T.G. and Turner J. D., “Corporate Governance of Banks in Developing Economies:
Concepts and Issues”. Available:
http://unpan1.un.org/intradoc/groups/public/documents/NISPAcee/UNPAN015471.pdf
[20 December 2009]

Hughes Jane E. and MacDonald B. Scott (2003), “International Banking”, Pearson Addison
Wesley, Google books, Available:
http://books.google.gr/books?id=SEwEpksebBwC&printsec=frontcover&dq=international
+banking&source=bl&ots=k7wgxr5Ym6&sig=fhr-NfkWv92ywm-
71mqaYM_OJkg&hl=el&ei=5TVcS_PFGpG7jAf4wuSVAg&sa=X&oi=book_result&ct=res
ult&resnum=10&ved=0CDMQ6AEwCTgU#v=onepage&q=&f=false [1 December 2009]

Ord Minnett Research, “Australian banks vs. International banks”, September 2008, Available:
http://www.fullermoney.com/content/2008-10-2/AustralianBanksvsInternationalBanks.pdf

Databases

Ø Corporate Governance Quotient

Ø EBSCO
Ø Thomson One Banker

[33]