University of Ghana http://ugspace.ug.edu.gh THE UNIVERSITY OF GHANA CORPORATE GOVERNANCE AND THE RISK-TAKING BEHAVIOR OF BANKS BY GODLOVE LARTEY ASIRIFI A THESIS SUBMITTED TO THE DEPARTMENT OF FINANCE, THE UNIVERSITY OF GHANA BUSINESS SCHOOL- UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR AN AWARD OF AN MPHIL DEGREE IN FINANCE JUNE 2017 University of Ghana http://ugspace.ug.edu.gh DECLARATION I do hereby declare that this work is the result of my own research and has not been presented by anyone for any academic award in this or any other University. All references used in this work have been fully acknowledged. I bear sole responsibility for any shortcomings. …………………………………. ………………………………. GODLOVE LARTEY ASIRIFI Date (10351229) i University of Ghana http://ugspace.ug.edu.gh CERTIFICATION I hereby certify that this work was supervised in accordance with the procedures laid down by the University. ………………………………….... ……….…………………….. AGYAPOMAA GYEKE-DAKO, PhD. Date (SUPERVISOR) ………………………………….. ……………………………… LORD MENSAH, PhD Date (SUPERVISOR) ii University of Ghana http://ugspace.ug.edu.gh DEDICATION I dedicate this work to the Asirifi family- my mum, dad and siblings – Mr. George Kingsley and Mrs. Edith (Adwoa Amoah), Joel (Kwaku Addo), Ruth and Prince- you give me reasons to excel beyond what I dare to dream of, I love you all. iii University of Ghana http://ugspace.ug.edu.gh ACKNOWLEDGEMENTS My utmost gratitude is to the Almighty God, for seeing me through this rough and challenging journey. I am also grateful to the many people who believed in my study and gave so much to make this thesis possible. My deepest gratitude goes to my supervisors Dr. Agyapomaa Gyeke-Dako and Dr. Lord Mensah. Your motivation, advice, disciplinary measures and attentive supervision conceived this work, you are wonderful mentors. Also, to Dr. Patrick Opoku Asuminng for his unflinching help, Dr. Elikplimi K. Agbloyor who was extremely helpful with data as well as the entire lecturers and colleagues of the Finance Department of the University of Ghana Business School. You were all very supportive in building a strong foundation in my academic career. Your efforts are indeed acknowledged and appreciated- a very big thank you. iv University of Ghana http://ugspace.ug.edu.gh ABSTRACT In the world of banking, risk-taking is a normal activity that facilitates the process in generating capital appreciation for investors. This risk-taking agenda has increased governance concerns and thus the relationship between shareholders and managers remains uncertain. Therefore, this study sought to provide a deeper understanding while establishing the relationship between corporate governance and the risk-taking behavior of banks in the Sub-Saharan Africa. The study considers reducing existing gap by probing into the extent to which corporate governance (measure by board strength, power held by the Chief Executive Officer and presence of quality auditors) influences the risk-taking decision of banks in sub-Sahara Africa. With a dataset of 442 banks (sample of 3035 observation) in 25 Sub-Saharan African countries within a period of eight years ranging from 2006-2013, the study employs a GLS-RE quantitative econometric technique. The study found board size (measured by log of board sizes and board size squared) to insignificantly explain risk- taking behavior of banks. Again, the study suggest board independent (negatively) and non- executive members (positively) of banks to be significantly related to the dependent variable, whereas CEO power of banks in sub-Sahara Africa had no significant relationship to risk-taking. Finally, the study also found auditors (as measured as the quality of audit from the big four) to have a negative statistical significance on the risk-taking behavior, thus, opinions from the big four auditing firms on the true and fairness of financial information is significant in reducing the risk- taking decisions of banks. From the findings it was revealed that bank board is very significant to bank risk-taking in Sub-Saharan African, therefore it is recommended that regulators need to monitor more intensely banks within the region where both shareholders and managers interests are aligned in an attempt to control their potential for unnecessary or excessive risk-taking. Future studies should consider the broader African spectrum as well as a comparison of public and non- public banks in the sub-region. v University of Ghana http://ugspace.ug.edu.gh LIST OF TABLES Table 2.1: Taxonomy of related empirical literature……………………………………………...37 Table 3.1: Summary of expected signs for the independent variables…………………………….48 Table 4.1: Descriptive Statistics of Econometric Estimation…………………………………….64 Table 4.2: Pearson Pair Wise Correlation matrix…………………………………………………67 Table 4.3: Model Diagnostics…………………………………………………………………….68 Table 4.4: GLS-Random Effect with Z-score as the dependent variable…………………………70 vi University of Ghana http://ugspace.ug.edu.gh LIST OF FIGURES Figure 1.1: The mean z-score with exponential trend line for sub-Saharan banks………………….4 Figure 2.1: Corporate Governance and Risk-Taking behavior of banks………………………….40 vii University of Ghana http://ugspace.ug.edu.gh LIST OF ABREVIATIONS AQ: Audit Quality BIND: Board Independence BS: Board Size BSIZE: Bank Size CAP: Bank Capital CEO: Chief Executive Officer CEOD: Chief Executive Officer Duality CV: Charter Value FE: Fixed Effect GDP: Gross Domestic Product GLS-RE: Generalized Least Squared- Random Effect LEV: Leverage LIQ: Liquidity NONEXEC: Non-Executive members OLS: Ordinary Least Squared RE: Random Effect ROA: Return on Asset ROE: Return on Equity VIF: Variance Inflation Factor viii University of Ghana http://ugspace.ug.edu.gh TABLE OF CONTENTS DECLARATION............................................................................................................................ i CERTIFICATION ........................................................................................................................ ii DEDICATION.............................................................................................................................. iii ACKNOWLEDGEMENTS ........................................................................................................ iv ABSTRACT ................................................................................................................................... v LIST OF TABLES ....................................................................................................................... vi LIST OF FIGURES .................................................................................................................... vii LIST OF ABREVIATIONS ...................................................................................................... viii CHAPTER ONE ........................................................................................................................... 1 INTRODUCTION......................................................................................................................... 1 1.0 Chapter Introduction ............................................................................................................. 1 1.1 Background to the study ........................................................................................................ 1 1.2 The Statement of Research Problem ..................................................................................... 3 1.3 Research Objectives .............................................................................................................. 5 1.4 Research Questions ............................................................................................................... 6 Hypothesis ................................................................................................................................... 6 1.5 Significance/Purpose of the study ......................................................................................... 7 1. 6 Chapter Disposition .............................................................................................................. 8 CHAPTER TWO .......................................................................................................................... 9 LITERATURE REVIEW ............................................................................................................. 9 2.0 Introduction ........................................................................................................................... 9 2.1 Concepts of Corporate Governance ...................................................................................... 9 2.1.1 Corporate governance ........................................................................................................ 9 2.1.2 Risk-taking behaviour ...................................................................................................... 10 2.2 Theoretical Review ............................................................................................................. 11 2.2.1 Corporate Governance Theories....................................................................................... 11 ix University of Ghana http://ugspace.ug.edu.gh 2.2.1.1 Agency Theory .......................................................................................................... 11 2.2.1.2 Transaction Cost Theory ........................................................................................... 13 2.2.1.3 Resource Dependence Theory ................................................................................... 14 2.2.1.4 Stewardship Theory ................................................................................................... 15 2.2.2 Corporate Governance Elements ...................................................................................... 17 CEO Power ................................................................................................................................ 18 Chief Executive Officer (CEO) Duality ................................................................................ 18 Strong Board ............................................................................................................................. 18 Board size .............................................................................................................................. 18 Board composition and control.............................................................................................. 19 Board Skills levels ................................................................................................................. 21 Auditors ..................................................................................................................................... 22 Internal and external auditors ................................................................................................ 22 Audit Committee ................................................................................................................... 22 2.2.3 Risk-Taking Behavior ...................................................................................................... 24 2.2.4 Types of Risk ................................................................................................................... 25 Operational risk ..................................................................................................................... 25 Liquidity risk ......................................................................................................................... 26 Investment risk ...................................................................................................................... 26 Exchange rate risk ................................................................................................................. 27 Credit risk .............................................................................................................................. 28 2.3 Empirical Review ................................................................................................................ 28 2.3.1 Corporate Governance Mechanisms ................................................................................ 28 2.3.2 Shareholders, Board of Directors and Risk-Taking ......................................................... 30 2.3.3 Chief Executive Officer (CEO) Power and Risk-Taking ................................................. 32 x University of Ghana http://ugspace.ug.edu.gh 2.3.4 Auditors, Corporate Governance and Risk-Taking .......................................................... 34 2.3.5 Banking Characteristics and Governance Issues .............................................................. 35 2.4 Conceptual Framework ....................................................................................................... 38 2.5 Conclusion ........................................................................................................................... 40 CHAPTER THREE .................................................................................................................... 42 RESEARCH METHODOLOGY .............................................................................................. 42 3.0 Introduction ......................................................................................................................... 42 3.1 Research Design .................................................................................................................. 42 3.2 Data Source and Sample ..................................................................................................... 42 3.3 Description and Measurement of Variables ........................................................................ 44 3.3.1 Dependent Variable .......................................................................................................... 44 3.3.2 Independent Variables ...................................................................................................... 46 3.3.2.1 Corporate Governance variables ................................................................................... 46 3.3.2.2 Control Variables .......................................................................................................... 49 3.4 Model Framework and Design ............................................................................................ 53 3.5 Econometric Techniques ..................................................................................................... 55 3.5.1 Random Effect (RE) and Fixed Effect (FE) Estimators ................................................... 55 3.5.2 The Hausman test of Comparison- RE and FE Estimators .............................................. 56 3.6 Ordinary Least Squared (OLS) ........................................................................................... 57 3.7 Generalized Least Squared (GLS)....................................................................................... 57 3.8 GLS-Random Effect ............................................................................................................ 58 3.9 Diagnostic Tests .................................................................................................................. 60 3.9.1 Independence Test ........................................................................................................ 60 3.9.2 Multicollinearity Test ................................................................................................... 60 3.9.3 Heteroscedasticity Test ................................................................................................. 61 xi University of Ghana http://ugspace.ug.edu.gh CHAPTER FOUR ....................................................................................................................... 62 PRESENTATION, ANALYSIS AND DISCUSSION OF RESULTS .................................... 62 4.0 introduction ......................................................................................................................... 62 4.1 Data and description of results ............................................................................................ 62 4.2 Pearson Pairwise Correlation Matrix .................................................................................. 65 4.3 Empirical Results ................................................................................................................ 69 4.4 Presentation of Findings ...................................................................................................... 73 4.5 Chapter Summary ................................................................................................................ 75 CHAPTER FIVE ........................................................................................................................ 77 SUMMARY, CONCLUSION AND RECOMMENDATION ................................................. 77 5.0 Introduction ......................................................................................................................... 77 5.1 Summary of Findings .......................................................................................................... 77 5.2 Conclusion ........................................................................................................................... 79 5.3 Recommendations ............................................................................................................... 81 5.4 Limitations and Gaps for Further Studies ........................................................................... 82 REFERENCE .............................................................................................................................. 83 APPENDIX .................................................................................................................................. 95 xii University of Ghana http://ugspace.ug.edu.gh CHAPTER ONE INTRODUCTION 1.0 Chapter Introduction This study presents a comprehensive introduction to the concept of corporate governance and the risk-taking behavior of banks. The problem of whether or not there exists a relationship between these two elements – that is corporate governance and risk-taking behavior – is the main concern driving the research. 1.1 Background to the study The business of banking has its own twist and turns. It provides diverse opportunities to management who outline strategies and measures to grow and compete advantageously. Africa’s banking sector has seen remarkable growth over the years (Kasekende, 2010). A sector that was previously underinvested and least explored due to its risk orientation and lower profit prospect is today considered one of the expectant prospect within the globe (World Bank, 2011). Like other businesses, the banking industry is established mainly to maximize the wealth and increase the investment of owners. As a result, risk-taking is a part of their daily routine of business. These risk-taking decisions hikes agency differences between owners and managers as to the level to taking risk. These agency problems arise as a result of the disparity of opinions and decisions in the relationship existing between owners of firms’ capital (shareholders) and users of owners’ capital to create value (management). In institutions, corporate governance measures serve as controlling mechanisms in monitoring the activities of managers in line with the ultimate goal of their superiors (Huang & Wang, 2015). Evidently, corporate governance has been found to play a very significant role in the financial systems, as the industry is seen to engage in very intricate and 1 University of Ghana http://ugspace.ug.edu.gh very risky operations. Empirical studies reveals that board size, board composition, management skill level (Abor & Biekpe, 2007) as well as a strong board, CEO power (Pathans, 2009) among others are core aspects of corporate governance which tend to determine the financial decisions and behavior of entities. Ironically, there seems to be no unanimously recognized definition for corporate governance, as different authors define the concept differently. Keasey, Thompson and Wright (1997, p. 3) defines corporate governance as “the process and structure used to direct and manage the business affairs of a company towards enhancing business prosperity and corporate accountability with the ultimate objective of realizing long-term shareholder value, whilst taking into account the interest of other stakeholders”. Similarly, “risk-taking in banking has been long recognized in theoretical and empirical research and, most importantly, in the actual conduct of bank regulators” (Chen, Hwang & Liu, 2012, p. 4). Although the underlining concept has been shown to have a substantial impact on many bank variables, the concept of corporate governance and the risk-taking behavior is only now becoming popular in Africa (World Bank, 2005). Also, the concept of risk is diverse. One of the classic explanations of risk proposed by Arditti (1967), is the plausibility that an actual return will be lesser than the normal return. Hence, the propensity to take risks in the banking industry is dependent on the level of risk appetite relative to the bank management and shareholders. Extant literature including that of Pathans (2009) suggest that shareholders are seen as risk lovers whereas management of banks, risk averse. This means that owners of banks would prefer more risk because they can diversify their portfolios in the capital market as compared to that of managers who can only successfully do so at the firm level (May, 1995; Pathans, 2009). 2 University of Ghana http://ugspace.ug.edu.gh Hence, due to the disparity of their risk appetites, the probability of agency conflict arising cannot be discounted. In sub-Saharan Africa, studies have sought to assess the clear link between corporate governance and a broad spectrum of finance related issues (Abor & Adjasi, 2007; Abor & Biekpe, 2007; Fiador, 2013; Malherbe & Segal, 2001; Rossouw, 2005). Although these studies have looked at corporate governance issues among financial institutions, very few of these studies have been specifically carried out on corporate governance among banking institutions as suggested by Aboayge and Otieku (2010). As banks play important roles in the socio-economic development especially in developing countries (Oten-Abayie & Frimpong, 2011), it is becoming increasingly important to examine the impact of corporate governance on the risk-taking behavior of banks in sub-Saharan Africa. 1.2 The Statement of Research Problem The financial sector of African economies is considered to be one of the significant areas within the region due to its enormous contribution to gross domestic product, growth and development. For instance, developing economies such as Ghana, Nigeria, Botswana and South Africa have their financial service contribution on average twenty-four percent to GDP growth (Naceur & Ghazouani, 2007). Banks, an aspect of the financial sector are one of the driving forces for economic growth due to their distinctive role in financial intermediation such as the disbursement system, liquidity, information control, asset maturity and denomination transformation (De Andres & Vallelado, 2008). Banks, by their operational nature, are risk inherent (Ahn & Choi, 2009). They behave like any other firm with the main objective being wealth and profit maximization. As such 3 University of Ghana http://ugspace.ug.edu.gh good governance is key in maintaining its status as irresponsible risk-taking could be detrimental to economies. A number of empirical findings show varying levels of risk-taking behaviours between shareholders and management (Bokpin, 2016; Pathan, 2009) and provides that bank shareholders, with the expectation of higher returns tend to engage management to take greater risk. To this effect, shareholders require management to outline various strategies to make fruitful investment decisions to their benefits. For fear of the unknown, management on the other hand, are hesitant in making such demands of shareholders a reality. This is because management fears excessive risk might go bad and thus, would prefer to limit their risk-taking behaviour. This leads to agency problem. Because of this conflicting objective between shareholders and management, shareholders appoints Board of directors and Auditors to direct and monitor the behaviour of management in line with their shareholders. This governance system comes at a cost to shareholders. Literature on how much good governance mechanism are able to significantly influence the risk-taking behaviour of banks remains unattended to in the context of sub-Saharan Africa. Chen et al. (2012) posit that risk-taking behaviour has had very minimal recognition in the academic research environment, substantially in the actual demeanor of banks governance. Very few studies on bank risk-taking have taken into account the relevance of board and management diversities (Bokpin, 2016; Chan et al., 2016; Hang & Wang, 2015; Laeven & Levine, 2009; Pathan 2009). Even so, there has been inconclusive results from these literature as to the extent to which good bank governance affect the risk-taking behaviour of banks in sub-Saharan Africa. Further, although various studies have sought to establish the resilience of economies financial system to the characteristics of bank governance (see Alfaro & Chen, 2012; Dietrich & 4 University of Ghana http://ugspace.ug.edu.gh Wanzenried, 2011, Ongore & Kusa, 2013; Sehaeck, Cihak & Wolfe, 2009 and Trinidad, 2006) very few attempts have been pursued to touch on the open secret that good governance largely influence the risk-taking behaviour of banks as well as other financial bodies in an economy’s financial system in the sub-Saharan African region (see Bokpin, 2016; Tsorhe, Aboagye & Kyereboah-Coleman, 2011). Therefore, given the need, this study uncovers the open secret of how good governance influences the risk-taking behaviour of banks in sub-Saharan Africa. The study incorporates major governance variables to provide an in-depth analysis to this effect. 1.3 Research Objectives Traditionally, the study sought to assess the level at which good corporate governance affects the risk-taking behavior of banks in sub-Saharan Africa. More specifically the study seeks: 1. To assess the effect of board strength (strong board) on the risk-taking decision of banks in sub-Saharan Africa. 2. To examine the impact of power held by the Chief Executive Officer (CEO power) on the risk-taking decision of banks in sub-Saharan Africa. 3. To assess the effect of the presence of quality auditors on the risk-taking decision of banks in sub-Saharan Africa. 4. To determine the relevance of board structure on the risk-taking behavior of banks in sub- Saharan Africa. 5. To determine the relevance of management structure on the risk-taking behavior of banks in sub-Saharan Africa. 5 University of Ghana http://ugspace.ug.edu.gh 1.4 Research Questions To accomplishing the research objectives, the researcher is motivated to formulate these research questions and it specifically include the following: 1. Of what impact does the board of directors have on the risk-taking behavior of banks in sub-Sahara? 2. How does the power held by the Chief Executive Officer of a bank influence the risk-taking behavior of sub-Saharan banks? 3. How does the presence of quality auditors influence the risk-taking behavior of sub- Saharan banks? 4. What is the relevance of the bank’s board structure to bank risk-taking behavior? 5. To what extent is the bank’s management structure relevant to bank risk-taking behavior? Hypothesis H1: Risk-taking behavior of banks in sub-Saharan Africa is positively related to strong board (size and independence) H2: Risk-taking behavior of banks in sub-Saharan Africa is inversely related to the power of CEO (CEO duality). H3: Risk-taking behavior of banks in sub-Saharan Africa have no relation with presence of Auditors (Audit quality). H4: Bank board structure is relevant in the risk-taking behavior of banks in sub-Saharan Africa. H5: Bank management structure is relevant in the risk-taking behavior of banks in sub-Saharan Africa. 6 University of Ghana http://ugspace.ug.edu.gh 1.5 Significance/Purpose of the study The main purpose of the study could be drawn from three different perspectives: the area of research (Academia), practice (Industry) and policy implementation (Government and The Central Banks). Concerning the research significance of this study, it will go beyond current or available research on corporate governance in sub-Saharan Africa because it will examine corporate governance with risk-taking behavior dimensionalities. The research contributes immensely to the body of knowledge since it is the first of its kind to measure the strength of a bank board, CEO power and influence from auditors on the risk-taking behavior of banks in the sub region of Africa and the world at large. This will serve as a guide to future researchers working either on corporate governance, risk taking behavior of the banking industry or both. Additionally, it also contributes to the scarce prevailing bank risk-taking literature considering the period (2006–2013), a period during which many banks especially in the developed economies suffered severe financial distress due to the financial crisis. Concerning the significance of the research to practice, this study will highlight the significance of corporate governance mechanisms and best risk-taking exposures associated with corporate governance structure within the sub-Saharan region. This will provide an in-depth measurement of the relationship that exist between corporate governance and risk-taking decision in the banking industry as well as providing the necessary advice to shareholders, investors and management. With reference to policy, the research provides a link (if any), between corporate governance mechanisms including board structure, management structure and risk-taking decisions in banking institutions. This will provide a guide to policy makers and regulators such that they are able to 7 University of Ghana http://ugspace.ug.edu.gh collaborate with banking institutions more in their bid to promote the belief that good corporate governance proliferates monitoring effectiveness and enhance a country’s economic development through a guaranteed sound financial system. 1. 6 Chapter Disposition The study is carefully organized into the following chapters: Chapter one provides an introduction to the research work, hence, providing a general background and problem statement to the study, objectives of study and research questions, significance of research and data. Chapter two provides a review of related literature. It critically defines concepts of and examines corporate governance and risk-taking behavior of banks, providing necessary theories and frameworks to address issues of concern. Further, Chapter three examines the methodology, thus analyzing the data and sources of relevant information for the research work. Chapter four, however, focuses on the presentation of the results and discussions, by presenting a detailed discussing into the results obtained. Finally, the last of the chapters, Chapter five, will focus on the summary, conclusion and recommendations. It summarizes the research findings and outlines recommendation and gap for future research to other writers. it should be noted that risk-taking behavior is interchangeably used with risk-taking decision. 8 University of Ghana http://ugspace.ug.edu.gh CHAPTER TWO LITERATURE REVIEW 2.0 Introduction In this episode, the study examines in critical detail, the concepts of corporate governance, and the risk-taking decisions of banks. The chapter looks in particular at some of the best practices associated with corporate governance around the world, how risk behaviour is measured and how risk-taking behaviour is affected by corporate governance mechanisms. As a result, we define the concept of corporate governance and risk-taking behaviour, the measurement of risk behaviour among banking institutions and a theoretical and empirical review of literature related to corporate governance and risk-taking behaviour. 2.1 Concepts of Corporate Governance 2.1.1 Corporate governance Conventionally, the concept of corporate governance has gone through a radical evolution with several authors propounding various theories to explain corporate governance and the issues surrounding it. The concept of corporate governance takes its roots from the writings of Berle and Means (1932) whose work throws considerable light on the agency problems that arise from separating company ownership from its management. This has also resulted in various definitions of corporate governance. However, there seems to exist no generally accepted definition for corporate governance (Fiador, 2013). Therefore, in providing a considerable agreement, Goergen and Renneboog (2006) establishes the concept of corporate governance from the agency theory approach to comprise a system of combining mechanisms that ensure management (the agent) runs 9 University of Ghana http://ugspace.ug.edu.gh the firm for the benefit of one (principal, shareholders) or several (creditors, government, unions etc.) stakeholders. Abor and Adjasi (2007; 5) broadens the scope and describes the concept of corporate governance as “the process and structure used to direct and manage the business affairs of the company towards enhancing business prosperity and corporate accountability with the ultimate objective of realizing long-term shareholder value, whilst taking into account the interest of other stakeholders”. Zingales (2008), in addition, defined corporate governance to mean the mechanisms, procedures and relations through which organizations are governed and controlled. With this, corporate governance in effect caters for the conflicts of interests that exist between the sponsors of corporate funds and their managers and the deterrence of these conflicts of interests (Goergen, 2012). From the numerous definitions of corporate governance, corporate governance can be concluded as both a structure and a relationship determining the direction and performance of a corporate. Hence, corporate governance simply is a system in which an entity (both profit and nonprofit) is governed and controlled. 2.1.2 Risk-taking behaviour Risk, unlike governance, is centred on uncertainties. Generally, risk appears can be explained as potentially losing or gaining something of value. The concept of risk is defined in several ways according to diverse disciplines. For instance, risk in finance is explained as the possibility that the real return from an asset will contrast that of the anticipated return (Arditti, 1967; Li, Jahera & Yost, 2013). Also, in Insurance, risk is examining as a condition where the possibility of a variable is identified but the means of occurrence, as well as the expected value of the occurrence, is not 10 University of Ghana http://ugspace.ug.edu.gh (Baker & Simon, 2010; Keeley, 1990). However, in literature, scholars such as Antunes & Gonzalez (2015) defines risk as simply the intentional interaction with uncertainty. Therefore, financial risk is the possibility of losses due to financial variables and this is as a result of the influences from the financial market variables (these includes foreign exchange risk, interest rate risk or credit risk), or with the business environment (including capital and liquidity risk) (Trinh, Duyem & Thao (2015). The propensity of risk-taking decisions may not be a distinguishing characteristic. Prior empirical evidences, Agoraki, Dalis and Pasiouras (2011): Bushman and Williams (2012): Delis and Kouretas (2011) and Houston, Lin, Lin and Ma (2010) suggested, “Bank risk-taking behavior is related to their charter value, stable shareholder holdings, ownership structure, bank financial characteristics, and macroeconomic situations”. This means that risk- taking behaviour functions from all possible areas of an institution to provide the paramount returns for the institution. 2.2 Theoretical Review The most popular theoretical models that examine the concept of corporate governance include the agency theory, the transaction cost theory, resource dependency theory, the stewardship theory and the stakeholder theory. However, for the sake of this study, we ground on the agency theory and support it with the transaction cost and the resource dependency. These theories are discussed below. 2.2.1 Corporate Governance Theories 2.2.1.1 Agency Theory One of the earliest theories, which was propounded to explain corporate governance, is the agency theory. The agency theory examines the relationship that exists between two entities (i.e. the 11 University of Ghana http://ugspace.ug.edu.gh principal and the agent). In this context, an agent is a person under contractual agreement to act on behalf of another (principal) and performs all duties in the interest of the person on whose behalf he acts upon. The principal, on the other hand, is the one on whose behalf the agent acts. The agency theory attempts to clarify the intricacies connected with the agency relationship. This problem emerges when the parties to the contract have different interests and there exist asymmetric information (the agent harbours information), such that the principal cannot reliably ensure that the agent is always acting in his best interest (Eisenhardt, 1989). As a result, shareholders bear the huge cost in monitoring the activities of managers while reaping a little portion of the benefits accrued. Therefore, some shareholders find it less rewarding to effectively monitor firms (Dong, Meng, Firth & Hou, 2014). Jensen and Meckling (1976) elaborated that although the agent is primarily employed by the principal to carry out the principal’s activities, there are certain benefits that the agent expects which sometimes leads to a situation of great conflict of interest. For instance, as many bank shareholders require their managers to take excessive risk in other to earn them abnormal profit, management would also consider the risk of losing their benefits, incentives and when these excessive risks backfires. In this vein, management, the sole controller of owners resources, tend to reduce the level of risk-taking not adhering to the demands of their owners. Fama (1980) points out that the agency relationship is extremely beneficial from an economic standpoint. This is so because bank managers consider all other stakeholders in making risky decisions in other to favour the greater goal. This behaviour of management depends on the conflict with shareholders (Denis, 2001). As such, as the problem in the agency relationship deepens, rick-loving shareholders are forced to spend more and more through monitoring managers to align their interest with that of 12 University of Ghana http://ugspace.ug.edu.gh owners. The study relies heavily on this theory as the theory provides the basics of our arguments in literature. This is so because banking business operates in this like manner (principal-agent relations). 2.2.1.2 Transaction Cost Theory The transaction cost theory was grounded on the study from Cyert and March (1963) in support of the agency problem. According to them, transaction cost encompasses the way an organization organizes and governs its control over transactions. The Institute of Chartered Accountant Ghana (2015) defines transaction cost to mean all cost expended in ensuring an economic transaction. For instance, the cost banks incur in providing value to their shareholders through making several risky investments assessments, cost and benefits analysis of portfolios. This result in most bank management both domestic and foreign willfully harbouring most of their risk inherent transactions in order to reduce the outrageous cost (Williamson, 1981). In examining this, the theory is broadly categorized into three: search and information cost (the cost required in determining the availability of profitable investment options to shareholders), bargaining cost (the cost required to reach a gainful agreement in favor of the shareholders) as well as policy and enforcement cost (the cost to secure the fulfilment of a bargain) (ICAG, 2015). Likewise, Campbell (2009) also categorized the theory into four; including transaction costs, contracting costs, coordination costs, and search costs. In illustrating these, bank management in the process of their acquisitions, investments and securing the best available fund inflows to increase capital portfolio faces varieties of transactional costs. To begin with, search costs, are usually the costs of finding potential investors and determining the state of their asset value. 13 University of Ghana http://ugspace.ug.edu.gh Secondly, bargaining costs involve the costs of negotiating a fair price (interest received on their investment) with the investors. Policing and enforcement costs lastly are costs associated with ensuring that the funds from investors flow into the business as promised (Campbell, 2009). All these transactions are geared towards obtaining the best value to maximize shareholder wealth as well as increase profitability. The problem arises when management who are to align strategic measures in ensuring that transactional costs are incurred to its minimum level for the benefit of bank owners, turn to benefit from the line of transaction. Further, Wilkinson (2005) include motivation cost (hidden information and hidden action) to affirm the agency cost. With this, Wilkinson (2005) suggest that bank owners suffer the cost of reducing the level of information asymmetry to a contract (hidden information) and the cost of monitoring management behaviour to ensure contracts are upheld in the interest of owners (hidden actions). This illustrates that, because bank managers usually possess substantial information, they tend to behave in a manner that does not favour their owners and as a result, shareholders incur a huge cost to mitigate this menace. Steenkamp and Geyskens (2012) posit that excessiveness of the cost associated with a transaction is resolved through expansion (Steenkamp & Geyskens, 2012). Therefore, banks need to expand in size and incorporates more transactions internally (carrying most activities in-house) other than outsourcing. 2.2.1.3 Resource Dependence Theory The resource dependence theory, as developed by Pfeffer (1973) was further improved by Pfeffer and Salancik (1978). Here they emphasized the point that non-executive directors are able to provide the necessary support to improve the ability for banks to protect itself against the 14 University of Ghana http://ugspace.ug.edu.gh environmental conditions, diminish vulnerability, or co-opt resources that increase the firm’s capacity to raise finances or increase its position and recognition. In order to succeed, corporations ensure the availability of resources for their survival and growth through the attempt of reducing the uncertainty of outsider control. In this approach, therefore, the board is viewed as one of the various instruments that may encourage access to assets basic to the risk-taking success of a company. From the resource dependence approach, boards of directors provide four primary types of assets. These are:  Advisory service, counselling, and technological know-how;  Legitimacy and source of goodwill;  Channels for disseminating information between outsider institutions and the firm; and  Preferential access to commitments or provision from significant actors outside the firm (Pfeffer & Salancik, 1978). In recent times, the resource dependence theory has seen a lot of scrutiny in numerous review and studies (see Davis and Cobb, 2010; Drees & Heugens, 2013; Hillman, Withers, & Collins, 2009; Sharif & Yeoh, 2014). These studies indicated and discussed the significance of the theory in discussing the actions of banks, through developing interlocks, associations, joint ventures, mergers and acquisitions, in the determination to overcome dependencies and facilitate an institutional independence and legitimacy. 2.2.1.4 Stewardship Theory The stewardship theory significantly disagrees with the agency theory. The theory posits that managers would be good stewards because the only way they can be considered as such and be rewarded accordingly, is by promoting the aims and objectives of the principal for whom they 15 University of Ghana http://ugspace.ug.edu.gh work. Stewardship theory argues that the problem of “managerial opportunism” as espoused by proponents of the agency theory does not exist in reality, because, managers desire to be considered as a good steward. Hence, management of banks will always put the interest of shareholder above theirs in all situations possible. The relationship between bank owners and managers in this scenario is inextricably woven and thus managers would not see any reason to make decisions, which would be inimical to the success of the business (Abor et al., 2007: Muth & Donaldson, 1998). Literature from Herzberg (1966) and McClelland (1961) supports the assertion raised. They believe that, compared to the agency theory, stewardship theory identifies a variety of non- monetary motives for managerial behaviour including the need for achieving and recognizing the intrinsic and inherent fulfilment of performance, reverence for authority and good work ethics. According to Etzioni (1975) when management are challenged with a numerous course of actions considered to be personally fruitless, bank managers may conform grounded in the sense of duty and identification to the bank. This assumption suggests that managers have a broader range of motives and behaviour beyond just self-interest. Therefore, the ‘so-called’ conflict of interest by bank management may not be inherent in the separation of bank ownership from controlling firm’s activities. With this assertion, the theory exposes that risk-taking decision by the management of banks are always in consonance (positive) to the requirement of bank owners and hence management see no reward to do otherwise. 16 University of Ghana http://ugspace.ug.edu.gh The major weakness of this theory is that the theory overemphasizes the ego of senior management to risk-taking and lacks empirical evidence to support its assertion. Again, the role of a ‘steward’ in the context is over-simplified and unrealistic. Practically, this is to `suggest that, management of banks are not probable to always act in the sole interest of bank owners. 2.2.2 Corporate Governance Elements There are a number of key elements that are used by experts in the area of finance to explain practices that constitute good corporate governance. One of the biggest challenges, however, when dealing with corporate governance issues is the slippery nature of the concept (Fiador, 2013). There are a number of elements which various authors have examined when trying to examine corporate governance among banks, SMEs or businesses in general. For instance, in examining corporate governance and value relevance, Kyereboah-Coleman and Osei (2006) in examining corporate governance among Microfinance Institutions in Ghana examined it in two main approaches i.e. The Stakeholder and Shareholder approach. Abor and Biekpe (2007) who also examined corporate governance among SMEs in Ghana focused on Board Size, Board Composition and Control, Board and Staff Skills levels, CEO duality, Inside Ownership, Family Ownership and Foreign Ownership. In a similar fashion Fiador (2013) adopts three main factors; CEO duality, Board Composition and Board size. Therefore, in this context, our study categorizes these mechanisms into CEO Power, Strong Board and Auditors. 17 University of Ghana http://ugspace.ug.edu.gh CEO Power Chief Executive Officer (CEO) Duality CEO duality simply refers to a state where the CEO of a firm is at the same time the board chair. This concept has gained an increasing level of research focus in this current era because of the great interest researchers and practitioners place on higher codes of governance (Fiador, 2013). In general, CEO duality is seen as a restriction on the independence of the board. As a result, in situations where the CEO doubles in strategic duty as the board chair, the effectiveness and efficiency of the board in terms of quality performance in increasing owners’ value will be compromised (Bliss, 2010). To elaborate further, excessive authority in the domain of a single person usually result in non- value maximizing and ineffective behaviour. As a result of this, a report from the South African king, provides the codes for best practices in corporate governance, unmistakably outlines the two powers and requires that they should be vested in two people (Fiador, 2013). Abor and Biekpe (2007) argue that the system where the CEO and the Board Chair are two separate and distinct individuals has been preferable to the scenario where the two are occupied by the same person because of the apparent inefficiency associated with the latter. They, however, point to studies that found CEO duality to be unrelated to the financial performance of companies. Strong Board Board size Board size focuses on the number of members allowed on a business’s board. The impact from the size of the board on its capability to deliver result has received high research interest over the years. Boards can be large or small. Literature suggests that larger boards are better for corporate 18 University of Ghana http://ugspace.ug.edu.gh efficiency due to a range of knowledge which helps members improve decisions and as well deters powerful CEOs from dominating (Abor and Biekpe, 2007). In contrast, Pathans (2009) argues that smaller board sizes with less restrictive boards are much preferable since they positively affect bank risk-taking and hence increase performance. However, recent studies have supported smaller boards (see Huang & Wang, 2015; Pathans, 2009). Emphasizing that large boards are less efficient, are highly liable for easy manipulations, and control from CEOs. It is further highlighted that, when a corporate board becomes flooded with more members, it is difficult to coordinate and has the potential of creating a lot of problems (see Huang & Wang, 2015). On the contrary, smaller boards have also been found to reduce the probability free riders and increase the responsibility of individual directors. Thus, the benefit of enhancing team growth and expansion through a large board of firm has strongly been academically argued to be a significant basis for promoting good corporate governance in small firms (Cadbury, 2000). The underlined consensus here remains that board size is essential in influencing the efficiency of the board. In addition, the association between board size and performance is however presumed to be non-linear (see Agoraki et al., 2009: Karamanou & Vafeas, 2005 and Pye, 2000). Board composition and control Usually, board composition serves as a good measure for board independence which examines the ratio of the number of non-executive directors on a board. Generally, it is acknowledged that a board is more independent and efficient if there is a higher ratio of external directors (directors without stakes in the form of investment and family relations in the business) (Yermack, 1996). 19 University of Ghana http://ugspace.ug.edu.gh That is to say, the more outside directors are on a corporate board, the more efficient in the performance of its core wealth maximization duties to shareholders. This is because it increases the capability to monitor, direct and control ineffective managerial behaviour (Fiador, 2013). From other opposing views (like Dow, 2003), internal directors are far more familiarized with the firm’s operations and for that reason, they should be able to act as watchdogs to top management and to indeed replace top managers in cases where they exhibit incompetence. Contrasting this, Abor and Biekpe (2007:9) suggest that “non-executive directors may act as professional referees to ensure that competition among insiders stimulates actions consistent with shareholder value maximization”. Studies argue that the perceived level of independence of the board of directors increases as the percentage of the non-executive directors increases. Various evidence from the literature on independent non-executive directors on bank board agree that monitoring and advisory functions executed on behalf bank owners are very significant and beneficial (see Aguilera, 2005; De Andres & Vallelado 2008; Kumar & Singh, 2012). However, Fosberg (1989) found no significant link in the proportion of non-executive directors and firm performance. Other studies also (see Bhagat & Black, 2002; Hermalin & Weisbach, 1991), similarly found no significant relationship between board composition and performance. Again, Yermack (1996) also provided that, the percentage of non-executive directors does not significantly affect firm performance. In auxiliary, evidence from Ghana, according to Torshe et al. (2011) suggest that board strength has no correlation with a firm's risky decisions of credit, capital and liquidity nature. 20 University of Ghana http://ugspace.ug.edu.gh Board Skills levels In literature, it is argued that the average level of education and development of board member could have a strong influence on the performance of the firm. Lybaert (1998) contends that better performance among members of the board (entrepreneurs) is as a result of the positive correlation of larger amounts of knowledge in education and their eagerness to use another source of information, create systems, make utilization of strategic or grow in-depth accountability and observing frameworks. Be that as it may, there exist contradicting proof about the level of development among investors and bank management. Evidently, Lawrie (1998) demonstrates that “gaps in management expertise are less of a recognized barrier to institutional development than the availability of specialist employee skills, chiefly in IT and languages”. Consequently, albeit higher significant level of managerial qualifications might be valuable, there is still uncertainty as to their importance. Subsequently, Powell (1991) keeps up that there is a negative impact on firm execution in risk-taking decisions because of the word related and proficient affiliations of exceptionally qualified administrators which may empower expanded agency behaviour. In sum, the board of directors with the necessary strength (size, composition and skills) exist for setting the strategic direction of the bank, supervising the risk management (Torshe et al., 2011) decisions and policies as well as reducing the agency and transaction cost problems of banks. The board, as appointed by and in the interest seeking of the shareholders of companies, representatively monitors, directs and controls the line of operational activities of management of banks in line with the wealth maximization of shareholders. Through this, managers are made accountable to all risky decision taken in their operational activities. 21 University of Ghana http://ugspace.ug.edu.gh Auditors Internal and external auditors Extant internal and external literature is far-reaching, extensive and delivers useful insight into the contemporary and potential roles of the auditor’s functions in corporate governance. In governance issues, auditors serve as the main points of neutrality for monitoring the effectiveness of the management. Consistent with the view of Fan and Wong (2005), high-quality auditors enhance the confidence of external investors and mitigates the enormous cost associated with their agency problems. To this effect as auditing is done by the “big four” firms, it strengthens investors’ confidence in an entity (banks) and helps reduce the cost of employing other monitoring mechanisms to reduce fraud and increase more investment. Conventionally, there exists a clear-cut difference existing between internal and external auditors. The difference in internal and external in the governance mechanism emanates from the focus of each group. Colbert, (2002) argues that while the mandate for the external audit arises from external users’ desire for credible reports (financial statements), the internal audit duties emanates from the board’s desire for information useful in pursuance of responsibility to help the accomplishment of the objectives of an organization. In addition, thus, the scope of the external audit is limited to the issues in association to the financial statement while in contrast, the work of the internal audit encapsulates all functioning, compliance and financial issues. Audit Committee The audit committee performs a very significant role in the corporate governance regarding the focus, control and the going concerns of an organization. As board representative and the foremost anchor of the corporate governance mechanism, the overall work of the audit committee involves 22 University of Ghana http://ugspace.ug.edu.gh “both internal and external audits, internal control, accounting and financial reporting, regulatory compliance, and risk management” of an entity (Al-Baidhani, 2014). Comparable to external auditors, the audit committee operates as representatives of the board to execute its corporate governance responsibilities including supervision, monitoring and risk management activities (Al-Baidhani, 2014). This is applicable to all entities either banks or non- banking institutions, public or private, as well as nonprofit making organizations (Fan and Wong, 2014). The audit committee assesses the going concern to provide the board of directors the necessary opinion (advice) and recommendations. These include ensuring:  That the underlined firm or institution complies with all the relevant regulations that the individual firms adhere to with social ethical and standards;  That the internal auditors remain highly independent auditors and capability;  that the financial information is organized correctly and accurately on due time; and that the all compensation paid to the firm’s executives were in consonance with firms fairness and structure (Al-Baidhani, 2014). Auditors (internal, external and audit committee) of banks serve as the pinnacles of ensuring good corporate governance. In the effort in assessing the going concern (future prospects) of a business, they address, examines and monitor risky decisions by managers in line with that of shareholders, in this manner, providing a safeguard approach to address the problems resulting from agency and transaction cost. 23 University of Ghana http://ugspace.ug.edu.gh 2.2.3 Risk-Taking Behavior The concept of risk-taking behaviour in the banks has been significant to both academic and evidence-based research and most importantly, the definite behaviour of bank regulators. Nonetheless, few aforementioned empirical studies (Chen, Hwang & Liu, 2012: Park, 1997) of bank risk-taking have considered bank variables such as charter value, ownership structure and regulatory environment to define risk, both at the bank-level and country level. Varied empirical evidence documented in the literature suggests various backgrounds to the risk appetite of managers and owners. Risk-taking involves the act of taking risky transactions to achieve the desired goals of an entity (Joseph, 2013). Risk-taking behavior in corporate banking industries are significantly different from other non-banking institutions (including IMF, insurance etc.) in relation to firm complexity, capital adequacy, transparency and regulations (Mehran, Morrison & Shapiro, 2011), and as such banks have the tendency to engage in risk-taking behaviors not privy to bank directors and shareholders. From the perspective of Pathans (2009), there exist three essential measures of bank risk-taking; total risk, idiosyncratic risk, and systematic risk (market risk). He examined that the systematic risk emanates from market structures which produce various forms of uncertainties and shocks faced by all market participants. Such shocks are usually from the nature of economic environment inhabited with government policies, universal financial powers, or demonstrations of nature. Conversely, the idiosyncratic error is the risk with which a particular industry is susceptible (Elosegui, 2003). All banking institution in the world are susceptible to these risks and should structure their activities or operations to ensure maximum profitability and minimize cost. The sum total of the systematic and unsystematic risk (idiosyncratic risk) is termed to be the total risk. 24 University of Ghana http://ugspace.ug.edu.gh 2.2.4 Types of Risk Financial transactions and activities are highly exposed to numerous kind of risk. That is to say, macroeconomic conditions such as inflation, exchange rates, government regulation and political stability affect investment. With this, identifying and outlining measures to reduce their negative impact on returns is very curial and necessary to institutional progress. This provides us with various types of risks that could affect Banking actives. These varied risks include operational risk, liquidity risk investment risk, exchange rate risk and credit risk. Operational risk The Basel Committee revised report defines operational risk “as the risk which results from the inadequacy or unsuccessful internal processes, individuals and systems or from external events” (The Basel Committee on Bank Supervision, 2005;2). Accordingly, this description of risk involves legal risk (such as but is not restricted to, introduction to fines, punishments, or reformatory harms coming about because of supervisory activities, and private settlements). Jarrow (2007) categorizes operational risk into the risk of loss as a result of a firm’s functional structure (failure in a firm’s transactional or business activity, or due to legal considerations) and the risk of a loss as a result of monetary motivations (including both fraud and mismanagement). In view of this, operational risk characterizes agency cost, due to the separation of shareholders and management (Brealey & Myers, 2004). More so, Shalizi (2002) points out that, operational risk results from stupidity (financial risk) and malice (corruption and deception). Hence, management of operational risk should involve all mechanisms which fundamentally serve the purposes of reducing average losses and avoiding the 25 University of Ghana http://ugspace.ug.edu.gh occurrence of catastrophic losses (Chapelle, Crama, Hübner & Peters, 2008). The revised Basel Capital Accord necessitates banks to meet the required capital for functioning risk as a feature of a general risk-based capital structure (Moscadelli, 2004). Liquidity risk Liquidity, as a concept, generally indicates the ability of an entity to trade large quantities/volumes quickly, and without any movement in the price level. In asset pricing theory, liquidity is seen as a good tool for price state variable and for investment decisions. Liquidity risk is defined by Jarrow (2007) “as the risk of a loss due to the inability to liquidate an asset or financial position at a reasonable price in a reasonable time period”. In banking, liquidity risk is attributed to deposits and their potential to spark runs or panics (Gatev, Schuermann & Strahan, 2007). Literature provides that banks that relied heavily on depository from customers and owners’ value financing kept on advancing with respect to different banks while their partners that held more illiquid resources on their accounting reports, conversely, expanded resource liquidity and lessened advancement (Cornetta, McNuttb, Strahanc & Tehranian 2011). Hence, off-balance sheet liquidity risk appeared on the accounting report and compels a new credit advancing to expanded whiles take-down request displaces advancement limit (Cornetta et al., 2011). Again, Gatev et al. (2007) posit that transactions deposits enable banks to hedge liquidity risk from unused loan obligations. Investment risk Another type of risk worth discussing is the investment risk. Investment risk is literally defined as the possibility that losses relative to its expected return on any particular investment might 26 University of Ghana http://ugspace.ug.edu.gh occur. Investment management suggests that managers, shareholders and customers share a common and the underlining definition to the concept of investment risk, whereas portfolio managers and individual investors share a conjoint conception of investment risk (Olsen, 1997). Investment risk cannot be fully insured with optimal contracts because shocks are mainly private information (Meha & Quadrini, 2005). Investment risk, as suggested by Olsen (1997) functions in four attributes. These include; the potential for a huge loss, the potential for an under-target return, the sensation of control, and the perception of the level of knowledge. One important tool of an international investment decision is the perception of an investor on risk. This could probably lead to investors having limited information and no universally agreed understanding of how investment risk should be conceptualized. Exchange rate risk Exchange rate risk provides inconsistency in the worth of a venture, or and benefits in the venture, that results from random dissimilarity in the exchange rate. Gray and Irwin (2003) in their study highlighted two types of exchange rate risk: project related and financing related. According to them, whereas a project exchange rate risk emanates from the value a project’s inputs or outputs, financing choices, on the other hand, affect the amount of exchange rate risk borne by different participants (stockholders, creditors, clienteles, taxpayers). Particularly, advances necessitating settlement in foreign currency highly expose shareholders to exchange rate risk (Gray & Irwin, 2003). Chow, Lee and Solt, (2016) emphasizes that the exchange- rate exposure on bonds is 27 University of Ghana http://ugspace.ug.edu.gh principally due to fluctuations in interest rates related to the unanticipated variations in the real exchange rate. Credit risk Credit risk is simply the risk that arises as a result of default to repayment of credit facility (Jarrow 2007). Expanding this definition, credit risk could be expressed as the risk of default on a debt that may arise from debtors failing to meet required payments. Current studies, especially, the study of Huang and Huang (2012) shows that “credit risk accounts for a small fraction of yield spreads for investment-grade bonds of all maturities, with the fraction lower for bonds of shorter maturities, and that it accounts for a much higher fraction of yield spreads for high-yield bonds”. Further, developments in commercial lending (resulting from the rise in a number of insolvencies, expanded disintermediation by the most credit-commendable borrowers, progressively focused edges on business advances loaning) have spawned the need for improvements in the measurement of credit risk (see Crouhy, Dan Galai & Mark, 2000; Gordy, 2000). The overall justification for analyzing credit risk demands for the full integration of market risk and credit risk. 2.3 Empirical Review 2.3.1 Corporate Governance Mechanisms Many researchers have identified and suggested a number of mechanisms adopted by companies in managing their governance issues (see Abor & Biekpe, 2007; Fiador, 2013; Guo, Smallman & Radford, 2013; Kyereboah-Coleman & Osei, 2006; Maher & Anderson, 2002). As a result, the interaction between law and finance has been acknowledged as one of them, particularly the 28 University of Ghana http://ugspace.ug.edu.gh international differences in investor legal protection (Talamo, 2011). Legally, the key approach to corporate governance mechanism has to do with the protection of outside investors (be it shareholders or creditors) (Gonenc & Aybar, 2008) through strong governmental and industry regulations. Further, another corporate governance mechanism termed ‘sharing control’ has been recognized by Gomes & Novaes (2005). According to this mechanism, negotiating difficulties among manifold controlling shareholders may avert unproductive investment results that harm mostly the few sectional shareholders (Gonenc & Aybar, 2008). However, on the other hand, these same negotiating problems may block efficient investment decisions. Therefore, in providing solutions to these trade-offs, there is the likelihood that shared control will enhance the increases three characteristics of a firm including overinvestment problems, the expense incurred in validating cash flows, and funding requirements (Gonenc & Aybar, 2008). Generally, the varying activities of businesses, their ownership and leadership style of managers also influence the mechanisms of governance adopted. On one side, Maher & Anderson (2002) assert that firms that do not adopt the cost-minimizing corporate mechanisms are less efficient and are most likely to be taken over in the long-run. Agreeably, the appropriate mechanisms to be adopted by firms are dependent on several factors including the size/structure, nature/type of the company and the national laws of many countries. In association, for instance, the corporate governance controls that are efficient in European countries have no important effect on firm performance in some part of Asia (China) and some 29 University of Ghana http://ugspace.ug.edu.gh other developing countries (Guo et al., 2013). An apparent reason for this could be as a result of the significant correlations between public companies and the government. 2.3.2 Shareholders, Board of Directors and Risk-Taking Indeed, it has long been identified in the literature that the concept of corporate governance conveys new and diverse strategic viewpoint through external autonomous directors and enhances firms’ corporate entrepreneurship, risk-taking propensities and competitiveness. Abor and Adjasi (2007) argue that the riskiness and problem of credit restriction and managerial ineffectiveness in the Ghanaian small and medium enterprise sector could be curtailed with a carefully designed corporate governance structure by the board of directors. According to Trinh et al (2015) banks as characterized by high industry rivalry, high rule, agency-related issues and high information asymmetric leads to the corporate governance concerns in the banking system as their operational habit involves risk-taking decisions. Empirically, with a model consisting of 7015 firm-year observation from 2001 to 2004, Jiraporn, Chatjuthamard, Tong and Kim (2015) found that organizations with more proactive and competent governance exhibit corporate structure are significantly less risky. Therefore, effective corporate governance mechanisms decrease the tendency of risk-taking significantly. Also, Dong, Meng, Firth and Hou (2014) in their examination on the ownership structure and risk management-related corporate governance as an impact on the risk-taking behaviour found that banks directed by the public (government) enjoy taking more risk than that managed and directed by private investors. In addition, their revelations highlighted that substantial amount of control positively 30 University of Ghana http://ugspace.ug.edu.gh facilitates and contribute to the monitoring mechanism of management which intend promotes prudent operating procedures. The study on concentrated ownership (influential owners) by Laeven and Levine (2009) have been found to be correlated with high bankruptcy risk in the era of capital regulation and greater return instability (Haw, Ho, Hu &Wu, 2010). This finding is supported by Pathans (2009), who found that strong bank boards that reflect shareholder’s interest, predominantly small and less restrictive, significantly positively impact bank risk-taking. With this, using a sample of 212 US bank holding establishments from the period of 1997 to 2004, he argues that since shareholders are at liberty to differentiate their portfolio in the stock market, they turn to be more risk loving. As the board reflects the interest of shareholders in banks they also turn to positively comprehend to the risk- taking appetite of shareholders. Hence, the strong board being the apex body of the internal governance structure of an organization is expected to adequately monitor CEO decisions and risk- taking performances. In the same vein, Huang and Wang (2015) also found a supporting evidence using 1990 firm data in 81 some three selected China Security Commission from the period of 2003-2011, concluded that entities with less board experience higher variability in long-term performance. This means that entities who turn to reward executive performance, tend to make riskier investment decisions, and participate more regularly in earnings management. In contrast, contextual evidence within Spanish commercial banks from Garcia-Marco and Robles- Fernández (2008) found ownership concentration (quantity of stocks owned by individual investors and the total shareholders) to be significantly related to low risk-taking decisions. This 31 University of Ghana http://ugspace.ug.edu.gh leads to healthier loan value, reduce the risk of an asset and a decrease bankruptcy risk (Iannotta et al, 2007) as against reduces delinquent loans ratio to improve a better capital adequacy ratio (Shehzad, de Haan & Scholtens, 2010). These disparities in the findings may be partly accredited to the difference in settings which implant different organizational features from a number of countries and regulatory establishments. 2.3.3 Chief Executive Officer (CEO) Power and Risk-Taking Pathan (2009) posits that CEO power measured by a firm’s long-term involvement (internally hired) and chair of the board (duality) tends to take fewer risk decisions. As a result, managers would do anything possible to constrain the information dissemination to other board of directors which intends decreases bank board’s independent oversight of management. To affirm this assertion, Pathan (2009) used a sample of 212 US bank holding businesses from the period of 1997-2004 and found CEO power (a CEO’s capability to influence board decisions) to be associated to bank risk-taking negatively. In addition, Kim and Lu (2011) showed the relation between CEO proprietorship and entity valuation on the forte of external governance and risk orientations. Their results suggested that CEO proprietorship and external governance are alternatives for controlling agency problems when ownership is minimum. Nevertheless, maximum levels of stock ownership decrease the value of a firm by establishing the CEO and discouraging them from taking risk. In contrast, using a cross-sectional difference in the structure of compensation contracts, Houston and James (1995) studied whether executive motivation encourages risk-taking in banks. Their results exposed that averagely, CEOs of banks who receive a less monetary reward, are not likely 32 University of Ghana http://ugspace.ug.edu.gh to contribute in a capital market formulation, possess low share options, and receive a minor proportion of their aggregate reward in the form of options and shares than that of CEOs in other industries. This, as a result, promotes the risk-taking of CEOs in banking. Similarly, Li and Tang (2010) related CEO hubris to firm risk-taking. Based on the “upper echelons theory” and the “behavioural decision theory” and a sample data from 2,790 CEOs of various manufacturing firms in China found a positive relationship between CEO hubris and firm risk-taking. Also, Jiraporn et al. (2015) argued that firms with more effective governance demonstrate strategies that are significantly less risky. That is managers, left to adopt their own strategies and control, tend to take excessive risk. As a matter of fact, effective corporate governance reduces the degree of risk- taking decisions by these managers significantly. A study by Mullineux (2006) on the implications of corporate governance in relation to banks fiduciary duty to their owners (including other stakeholders as well the government’s fiscal duty to taxpayers) argued that banks managers adhere to their fiduciary duty to depositors (more risk averse) as well as shareholders (more risk susceptible). Thus, a solution to the “principal-agent problem” aimed at maximizing shareholder value is strengthened. Here Mullineux (2006) argues that managers balance the interest of all stakeholders. Also, a study that analyzed the relationship between a CEO’s individual risk-taking behaviour, institutional risk-taking, and total firm risk by Gain and McKeon (2016) found evidence that firms led by CEOs with higher personal risk have a higher return on capital volatility, beyond the amount described by compensation components that financially reward risk-taking. As a result, prefers to take more risk. This is contrary to CEOs with less personal risk. 33 University of Ghana http://ugspace.ug.edu.gh 2.3.4 Auditors, Corporate Governance and Risk-Taking The influence of Audit as a corporate governance mechanism on the risk-taking of the firm has seen very little in literature. Nonetheless, the studies that have incorporated it have found no significant relationship (Beasley, 1996; Tsorhe et al., 2011). Ideally, auditors provide quality assurance regarding corporate governance, control mechanisms and risk-taking processes (Torshe et al., 2011). Auditors, in addition, provide an independent review of annual financial information to ensure that appropriate international accounting standards and procedures are used in their preparation as a measure of assessing the risk of firms (banks). Additional, board has the responsibility to appoint independent auditors. In this light, auditors protect the interest of shareholders as well as their risk-taking appetite. This means that the expression of opinions by these auditors is held in high esteem by both the management body and bank boards (DeZoort & Salterio, 2001). Hence, this body of persons is seen as influential in financial decisions of a firm. However, studies (see Torshe et al., 2011; Trinh et al., 2015) have found no significant effect on bank risk-taking decisions. For instance, Trinh et al. (2015) in his study assessed the impact of corporate governance on financial risk in Vietnamese commercial banks. The study revealed that auditors (audit committee) had less or minimal influence on the risk-taking level of banks to a noticeable degree. This is because the main duty of an audit committee is only to oversee the integrity of bank financial statements, internal auditing and risk management. Spira and Page (2003) sought to provide a radical redefinition of the changes in the nature of internal audit as a mechanism of corporate governance in the UK, clearly bring into line internal control with risk management. Their study explored this variation, using a sociological viewpoint on risk and its conceptualization to structure 34 University of Ghana http://ugspace.ug.edu.gh the debate about internal controls and management of risk and found out that corporate governance information reporting frameworks offer various advantages for the appropriation of risk and its management by groups willing to promote their own benefits. 2.3.5 Banking Characteristics and Governance Issues Players in financial intermediation differ from country to country depending on the level of financial development within that country. Amidu and Wolfe (2012) assert that banks primary function in intermediation is to direct funds from surplus units to deficit units. In Kwakye’s (2012) study it is asserted that the higher the level of financial intermediation in any economy, the greater the possibility of the financial sector contributing significantly to economic growth. He explained that this was due to the high level of intermediation resulting from financial assets being turned into real assets. Also, Halling and Hayden (2006) assert that due to the fundamental role of financial intermediation by banks, there is the need for greater efficiency in managing them (cited in Siaw, 2013). Early researchers have over the years studied the effect of financial intermediation on economic growth and found that the two are positively related in thus financial development led to growth in any economy (Goldsmith, 1969; Greenwood & Jovanovic, 1990; Montiel, 1995). Swank (1996) cited in Alu (2013), describes the services banks supply as essential services that are needed for the appropriate functioning of an economy. The sensitive nature of banking activities draws concerns to governance issues through regulations, monitoring and controlling in other to channel activities to the interest of shareholders. Thereby, solving the problem in the agency relations. Mullineux (2006) suggest that corporate governance varies in banking institution since banks act as stewards to shareholders and customers. In effect, good corporate governance 35 University of Ghana http://ugspace.ug.edu.gh of banks necessitates good risk-related management, prudential regulation and attention to conflict of interest and competitive issues that arise in banking operations. The figure below outlines the various empirical evidence from the most recent related literature concerning corporate governance and the risk-taking behaviour in the world. The evidence provided in the taxonomy provides varying appetite to this issue of the risk-taking behaviour by management and shareholder in the various institutional discipline. 36 University of Ghana http://ugspace.ug.edu.gh Table 2.1 Taxonomy of Related Empirical Literature Author/Title Study Objective Methodology Key Findings Bokpin (2016). Bank governance, The effect of corporate governance on Econometric panel data from the period 2000 The study found that reserve requirement regulations and risk-taking in Ghana. bank risk-taking conduct with a directing to 2013 under the fixed effect model. regulations influence risk-taking positively. result of various types of proprietorship on the relationship in regulatory condition. Chan et al. (2016). The Chinese banks’ To examine the impact of director’s socio- The GLS method with Arellano and Bover’s The study found smaller board sizes with a directors and their risk-taking economic background on the risk-taking (1995) generalized method of moment for 16 high proportion of independent directors behaviour. A governance and finance behaviour of the listed commercial banks listed commercial bank for the period 2003- low to risk-taking perspective. in Ghana. 2011 Dong, et al. (2014). Ownership structure This study investigates the effect of An OLS method of estimation was used with The research found that banks controlled by and risk-taking: a comparative evidence ownership structure on Chinese banks' a sample of 667 yearly observations covering the government tend to go for risk than from private and state-controlled banks risk-taking practices. 108 Chinese commercial banks over the those controlled by the state-owned in China. period from 2003 to 2011. ventures or private investors. Huang & Wang (2015). Corporate The study investigates the relationship An unbalanced panel data consisting of 1990 The study found a firm with smaller board governance and risk-taking of Chinese: between board size and firm risky policy firms over 2003–2011 across 81 different experience larger tendencies in pursue the role of board size. choices. three-digit CSRC industry codes. riskier investment and frequent earnings management. Jiraporn et al (2015). Does corporate The study examines the effect of corporate Data from firms reported by the Institutional The study found effective corporate governance influence corporate risk- governance on the extent of corporate risk- Shareholder Services (ISS) which consist of governance strategies to be less significant taking? Evidence from the Institutional taking. 7015 firm-year observations from 2001 to to risk-taking. Therefore, Efficient Stakeholders Services (ISS). 2004. governance, decrease the grade of risk- taking significantly. Pathan (2009). Strong boards, CEO The study explores the relevance of bank Using a GLS-RE method with a data of 212 The study found out that strong bank board power and bank risk-taking. board structure on bank risk-taking. large US bank holding companies from 1997– to positively impact risk-taking and CEO 2004 (1534 observations) power to negatively affect bank risk-taking. Mohsni & Otchere (2014). The risk- The study sought to determine the risk- With a univariate test and a data set sample The study found that privatized banks taking behaviour of privatized banks. taking behaviour of privatized banks prior consisting of 242 bank privatizations in 42 knowledge in risk significantly decreases to and after privatization countries that occurred between 1988 and after privatization; nevertheless, they 2007. continue to show higher riskier decisions than their counterparts. Trinh et al (2015). The impact of The study sought to examine the effect of An empirical quantitative study investigates a The study found that board strengths, corporate governance on financial risk corporate governance mechanisms on the sample of 26 joint-stock commercial banks in foreign capital, information disclosures and in Vietnam Commercial bank. capital risk, credit risk, and liquidity risk in the period of 2009-2013. stakeholder roles have a significant effect Vietnamese commercial banks. on financial risk management in the banking system 37 University of Ghana http://ugspace.ug.edu.gh 2.4 Conceptual Framework The rooted concept on the impact of corporate governance on the risk-taking decision of banks is described in two ways; the internal (via, management and ownership structure) and external approaches (via, market discipline and system regulations). Whereas the internal corporate governance mechanism encompasses accountability, monitoring, directing and controlling the risk-taking and the utilization of resources by firm management, the external corporate governance provides a highlight of the rules and regulations within the banking industry. This study ignores the external components’ influence on the risk-taking decision of banks as shown in the conceptual framework and focuses only on the internal structures with respect to their varying influence on the risk-taking the behaviour of banks as established in the literature. The internal mechanism is sub-divided into the management structure and the ownership structure. The management structure involves a team of bank managers headed by a CEO, who represents and provides organizations with the significant level of strategic decisions to move the firm ahead. In addition, the ownership structure incorporates the board of directors and the auditors, who stand in to monitor the risk-taking decisions of management on behalf of shareholders. Here, we show that governance mechanisms vary significantly with their influence on risk. This is because as one party may prefer higher risk, the other party, on the contrary, may prefer low and vice versa. To achieve a common interest between the varying difference between shareholders and management, good corporate governance measures must be instituted to align both interests. It is explained that shareholders in their quest to take risk rely on the board of directors and opinions from auditors since they are closely connected to the affairs of bank’s activities. Hence 38 University of Ghana http://ugspace.ug.edu.gh favourable feedback from bank boards and auditors implies more risk-taking. From these assertions, the section provides a diagrammatic review of how internal corporate governance mechanisms impact on the risk-taking behaviours of banks as suggested in the literature. In detail, the model in Figure 2.1 suggests that all the governance mechanism, with a measure of degree influence the risk-taking behaviour in the banking environment. Lipton and Lorsch (1992) suggest that smaller board size with adequate independence are more effective since a larger board is likely to encounter difficulties in the supervision of management. This is strengthened with the inclusion of the auditors who provides neutral assessment into the financials and risky decisions of bank management. On the other hand, bank’s CEO, if given the necessary power in terms of duality will be likely to take decisions of self-interest which will be detrimental to shareholders. With this, CEOs are likely to restrict the information flow to the board and hence causing disintegration 39 University of Ghana http://ugspace.ug.edu.gh Figure 2.1: Corporate Governance and Risk-taking behaviour of Banks CORPORATE GOVERNANCE Internal Mechanisms External Mechanisms Management Structure Ownership Structure Bank CEO Board of Director Auditors RISK-TAKING BEHAVIOR IN BANKS Source: adapted from Trinh et al., (2015) 2.5 Conclusion This chapter has critically examined in particular some of the best practices associated with corporate governance around the world as well as the risk-taking behaviour of banks and how risk- taking behaviour is affected by corporate governance. It is documented that the significant players in the banking industry (investors/owners and management) have varying decisions in respect to risk-taking. As a result, as one party prefers more to risk the other moves in an opposite direction, deepening the problem as suggested in the agency theory. 40 University of Ghana http://ugspace.ug.edu.gh In the nutshell, from the various studies that have been examined and discussed in this section, it is undoubtedly clear that the performance of banks is affected by bank risk-taking decisions. Hence, corporate governance mechanisms have a significant effect on bank risk behaviour. 41 University of Ghana http://ugspace.ug.edu.gh CHAPTER THREE RESEARCH METHODOLOGY 3.0 Introduction This chapter discusses the methodological approach and procedures used in this study. Essentially, the methodology consists of the model framework and the techniques used to attain sample data, analyze the data, and present the empirical outcomes by establishing the expectations, reasons and theoretical concept of the approaches or techniques. The foremost items that are discussed in this chapter include the research design to the study, data sample, the source of data, model framework, description of variables as well as the econometric measurement and data analysis procedures. 3.1 Research Design The research is designed to provide the preferably adopted strategy useful to assimilate the various mechanisms of the study in a comprehensible and logical way. This is guaranteed effectively by addressing the research problems, establishing the blueprint for the collection of data, providing the measurement mechanism, and analyzing the data. The study employs the quantitative econometric approach to investigate the impact of sub-Saharan banks’ board strength, CEO power and auditors on banks' risk-taking. Data was sourced and collected from the Bank Scope Database. 3.2 Data Source and Sample In the quest to investigate the impact of good corporate governance (as represented by the strength of the board, the power held by the CEO and the influence of external auditors) on bank risk-taking 42 University of Ghana http://ugspace.ug.edu.gh behaviour, the researcher extracts bank-specific accounting data from the Bank Scope database. The data covers four hundred and forty-two (442) banks from twenty-five (25) sub-Saharan African countries over the period eight (8) years ranging from 2006 to 2013. The data set contains all the adequate information about the selected banks, their financial conditions, audit systems, corporate governance and board structure, as well as their financial leveraging, assets and the major roles for which bank CEOs play in all significant and sensitive transactions. Similar to Adams and Mehran (2008), the good governance data structure is measured as proxies of board strength (non-executive members’ independence and size), CEO duality (chair of bank board) and Audit (the quality of audit from the big four). The researcher’s data collection procedure was also adjusted to account for the proxies to reveal some governance information (for instance, the proportion of CEO controlling rights) and others for the following (like that of the number of directors on a specific bank board). In addition, the financial information on the Banks was obtained from Bank Scope database. Based on the similarities of African bank characteristics, a sample of 442 banks from 25 countries out of the 54 countries in sub-Saharan Africa was sampled for the study. These countries were selected based on the available data to the researcher. In essence, the data was consistently cleaned to provide for them as well as arrive at the needed variables (dependent and independent) necessarily for the study. Further, the study encompasses an intersection of data on Sub-Saharan Banks from the Bank Scope database, and with minimum two consecutive years’ data over 2006– 2013, consisting of approximately 3520 observations. 43 University of Ghana http://ugspace.ug.edu.gh 3.3 Description and Measurement of Variables Here, we discuss and describe all the variables of interest to our study as well as providing their respective measurements. Our variables are of two categories; dependent and independent, with the independent variables comprising of the variables of interest and control variables. 3.3.1 Dependent Variable RISK This asserts probability of an investment’s actual returns deviating from the expected return. Risk- taking is proxied by the Z-score. The study measures z-score as the return from an asset plus ratio divided by the standard deviation of asset returns. As stated in the literature (see Bokpin, 2016; Kick & Prieto, 2015; Laeven & Levine 2009; Saghi-Zedek & Tarazi, 2015), the Z-score is described to be the status from bankruptcy, where bankruptcy is described as the condition in which bank losses exceed equity. From the prevailing literature, bank risk is measured in numerous ways. According to the study by Pathan (2009), “risk is classified into three primary measures; namely total risk, idiosyncratic risk, and systematic risk.” In a similar fashion, Anderson and Fraser (2000) in their study employed the use of the total risk measure but computed bank total risk as the standard deviation of its daily shares return. Here, the daily share return was computed as the natural logarithmic function of the ratio of equity returns, i.e., ln(Pit/Pit-1), where Pit represents the share values adjusted for capital modification (dividend and stock split). This includes the general changes in bank stock returns mirroring the market's observations about the risk intrinsic in the bank's benefits, liabilities, and off-balance sheet positions (Pathan, 2009). 44 University of Ghana http://ugspace.ug.edu.gh “The main drawback to this measure of bank total risk is that it is applicable to only listed firms (firms listed on the stock exchange market)” (Bokpin 2015: 8). On the contrary, this study includes all the banks in the banking industry in the sub-Saharan countries of Africa and out of which only a few numbers of these banks are listed. Hence, we, therefore, find it appropriate to employ the Z- score as the commonly accepted and appropriate measure of bank total risk for the study. Empirically, increasing number of literature (see Agoraki, Delis and Pasiouras, 2011; Bopkin, 2016; Laeven and Levine, 2009 and Mohsni & Otchere, 2014) helps us to measure RISK mathematically as: (𝑅𝑒𝑡𝑢𝑟𝑛 𝑜𝑛 𝐴𝑠𝑠𝑒𝑡) + 𝐴𝑣𝑒𝑟𝑎𝑔𝑒(𝐸𝑢𝑖𝑡𝑦/𝐴𝑠𝑠𝑒𝑡) 𝑍𝑠𝑐𝑜𝑟𝑒 = 𝑆𝑡𝑑(𝐴𝑠𝑠𝑒𝑡𝑠 𝑜𝑛 𝐴𝑠𝑠𝑒𝑡) In simplified terms the mathematical formula for the Z-score is deduced and presented as; 𝑅𝑂𝐴 + 𝐶𝐴𝑅 𝑍𝑠𝑐𝑜𝑟𝑒 = 𝜎(𝑅𝑂𝐴) From the literature, since the Z-score is a measure of the degree of stability. Therefore, a higher Z-score indicates that an African bank is more stable suggesting that the Z-score is highly skewed. For the sake of robustness check, the research additionally employs the variations of earning as an alternative measure for risk-taking in the banking industry. 45 University of Ghana http://ugspace.ug.edu.gh 3.3.2 Independent Variables 3.3.2.1 Corporate Governance variables Strong Board Bank board monitors the administrative behaviour of bank managers to limit management’s opportunistic behaviour and actions against the interest of shareholders. This monitoring is achieved efficiently depending upon the effectiveness of boards constructs (such as board size, composition and independence). A strong board is a proxy for Board Size (BS), Board Independence (BIND) and the strength of non-executive members (NONEXEC) on a board. Similar to the views of Pathans (2009) and with scare theories as to the most significant board mechanisms, an ad hoc selection of bank board size, non-executive members and independence as highlighted most in the empirical literature is used as a proxy of a “strong bank board”. BS represents the numerical members on a bank board. It measures the numerical strength of the board of directors of a firm. Lipton and Lorsch (1992) as cited in Torshe et al (2011) provides that lesser board size is more effective since larger boards encounter more complications in the supervision of managers and are slow in making decisions. However, on the contrary Pearce, and Zahra (1992) propose a larger board size strengthens the level of capacity to supervise, monitor and promotes information sources. Following from these arguments, we square board size (BS2) to increase the number of members on bank board and test the impact on the explained variable. In addition, BIND in the research is estimated as the proportion of the number of explanatory non- executive members to the aggregate number of director’s expense to the capital employed. Board composition as suggested in the literature specifies a portion of non-executive board members that 46 University of Ghana http://ugspace.ug.edu.gh offer an unbiased opinion and judgement in the discussion and decision-making process of the board (Adams & Mehran, 2008; Brickley, Coles & Jarrell, 1997). Mathematically BIND is expressed as; 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑖𝑛𝑑𝑒𝑝𝑒𝑛𝑑𝑒𝑛𝑡 𝑑𝑖𝑟𝑒𝑐𝑡𝑜𝑟𝑠 𝐵𝐼𝑁𝐷 = 𝑇𝑜𝑡𝑎𝑙 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑑𝑖𝑟𝑒𝑐𝑡𝑜𝑟𝑠 The NONEXEC is measure as the total number of external independent board members who have no operational and financial stake in the business and as a result, have no reason to be in the best books of bank management. As board members represent the interest of shareholders of banks, we assume to have a positive effect for all board variables signifying an increasing monitoring mechanism on bank management to increase the risk-taking decision in line with the expectations of shareholders. CEO Duality (CEOD) Literature from Torshe et al (2011) recommends that, ideally, corporate bodies should not have the Chief Executive Officer (CEO) as Chairman of the Board of directors. In the event of this situation, it leads not solely to the concentration of power and control in the hands of one individual but could also result in a serious impediment of the process of checks and balances. CEOD, a proxy for dual CEO power is measured as a dummy variable. This provides the study as to whether the CEO of an entity is also the head (chair) of the board or otherwise. This is represented by 1 if CEO is the chair of the bank board or 0 otherwise. 47 University of Ghana http://ugspace.ug.edu.gh Essentially, the study expects to see a negative relationship with z-score (a proxy for risk-taking). This will help establish that bank CEOs are risk-taking averse, hence, prefer to take less risk in other to protect their interest (position). Audit Quality (AQ) According to existing literature, the most important duties of an auditor are to provide due diligence and assurance regarding the going concern corporate governance, control systems and risk management processes (Knechel & Salterio 2016). In an attempt to ensure good corporate governance, auditors evaluate all documents prepared by a firm (such as the financial statements, policy documents, etc.) to ensure that the appropriate accounting standards, policies and procedures, are adhered to whiles seeking the interest of owners. The researcher considerate for the existence of internal auditors. Hence, audit quality represented by AQ shows the possibility of detecting significant errors or misstatement in the financial information. We measure AQ as a dummy represented by 1 if audited by any of the big 4 audit firms (PWC, KPM, Deloitte & Touché and EY) or 0 otherwise. The researcher expects AQ to be negatively related with risk-taking (Z-score) to indicate that opinions from the big four (audit quality) are a significant catalyst to restrict firms from taking an excessive risk which might affect the prospect of a banks’ going concern. 48 University of Ghana http://ugspace.ug.edu.gh Table3.1: Summary of expected signs for the independent variable Authors and Findings Expected Sign Independent Variables Strong Board Board size Salhi and Boujelbene (2012) found small board’s size to help reduce risk- taking activities with a dataset of 10 Tunisian banks over an 8-year period _ from 2002-2009 Bokpin (2016) with data on banks in Ghana found that large boards increase the monitoring mechanisms of bank managers in line with the + interest of shareholders Board independence Booth, Cornett and Tehranian (2002) provided that external members characterize independence and strength of the board. They found that the + larger proportion of outsider directors increases the risk-taking actions of banks due to agency conflicts. Non-executive members Pathan (2009) found that board members who have no financial stake in a business tend to take more risk (hence prefer more risk initiatives). + CEO Power CEO duality Pathan (2009) concluded with data from US banks that CEO are risk- averse individual (hence prefer fewer risk initiatives) in the quest to _ protect their jobs. Audit Audit quality Beasley (1996) found an evidence that audit committee (who ensures the quality of audit) has no relationship with financial statement fraud +/_ Source adapted from Trinh et al (2015) 3.3.2.2 Control Variables Bank Size (BSIZE) The bank size represents the total asset (the value of current and non-current assets available within a year) of a bank for each fiscal year. Bank Size, denote by BSIZE represents the total value of the bank. This represents how large or small a firm is. Mathematically, it is calculated as the natural logarithm of the Total Assets of the bank. Total Asset = ln (TA) It is evident from the literature that the higher the size of a bank (in terms of total asset value) the more probable the bank is to risk-taking. This is because bigger banks have better contact to 49 University of Ghana http://ugspace.ug.edu.gh financial facilities and loan clienteles which in turn results in the better portfolio diversification. As a consequence, it is anticipated that risk-taking would be positively related to bank size. Charter Value (CV) The Charter Value of bank is denoted by CV. This represents the going concern of a bank. It expresses the overall value of a bank's ability to continue in operation into the foreseeable future. It is measured as the sum of equity and liability divided by the value of the total asset. 𝑇𝑜𝑡𝑎𝑙 𝐸𝑞𝑢𝑖𝑡𝑦 + 𝑇𝑜𝑡𝑎𝑙 𝐿𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝐶𝑉 = 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡 CV is predictable to be negatively correlated with the dependent variable (risk-taking). This is because as a bank predict and strive to be in existence for the foreseeable future, they tend to prefer low to excessive risk-taking in order to protect the banks’ future. Bank Capital (CAP) Also known as Equity ratio is the bank capital. CAP is used to represent the Bank Capital. In the literature, bank capital represents the overall value of the bank's total capital divided by its total asset. Banks with larger equity ratios are seen to be safer and better placed than their counterparts and hence would prefer to lower their risk-taking behaviour to protect their future uncertainty. The researcher expects a negative relationship with the dependent variable to that effect. However, we measured CAP as the percentage of Bank total equity as a percentage of total asset. This is expressed as; 50 University of Ghana http://ugspace.ug.edu.gh 𝑇𝑜𝑡𝑎𝑙 𝑒𝑞𝑢𝑖𝑡𝑦 𝐶𝐴𝑃 = 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡 Return on Asset (ROA) Return on assets, proxied ROA, indicates the level of profitability relative to a bank’s total asset. As documented in the literature, the ROA provides an underlining idea as to the level of efficiency bank management are in utilizing the assets of banks to create earnings. Mathematically, it is calculated by dividing the bank's yearly profit by its total assets, ROA is usually reported in a percentage form and it is represented as; 𝐵𝑎𝑛𝑘 𝑛𝑒𝑡 𝑝𝑟𝑜𝑓𝑖𝑡 𝑅𝑂𝐴 = 𝑥 100% 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡 Since ROA measure profitability, it is expected that when a ban has a higher profitability index (ROA), the higher their risk orientation. This is because the bank would possess more funds and would be in a better position to contribute to a better diversification of banks’ portfolio. As a result, it is expected that risk-taking would be positively related. Return on Equity Return on Equity as proxied ROE. In plain language, ROE is explained as the amount of operating income after tax returned as a proportion of shareholder’s fund. It also measures a bank’s earnings by advertising how much profit a bank generates with the funds invested by owners. In practice, “all other things constant” a higher and rising ROE represents an efficient use of owner’s resources 51 University of Ghana http://ugspace.ug.edu.gh and vice versa. In the light of this, it is expected that ROE should positively relate with the dependent variable suggesting that banks are willing to make riskier transactions in order to increase the value of shareholder’s investment. 𝐵𝑎𝑛𝑘 𝑛𝑒𝑡 𝑝𝑟𝑜𝑓𝑖𝑡 𝑅𝑂𝐸𝑇 = 𝑥 100% 𝑜𝑡𝑎𝑙 𝐸𝑞𝑢𝑖𝑡𝑦 Liquidity LIQ, representing liquidity, indicates the availability of asset to a company. This provides the tendency at which an asset can be easily convertible (bought or sold) in the asset market with no change in price or value. It is calculated as a bank’s current assets per its current liabilities. The research expects a negative relationship between liquidity and risk-taking, this is because banks are highly sceptical of ‘runs’ and hence would prefer to take a low risk in other to meet the demands of customers in due time, adhere to regulations and as well do away with distress. 𝐶𝑢𝑟𝑟𝑒𝑛𝑡 𝐴𝑠𝑠𝑒𝑡 𝐿𝐼𝑄 = 𝐶𝑢𝑟𝑟𝑒𝑛𝑡 𝐿𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑖𝑒𝑠 Leverage Leverage proxied by LEV is explained by a bank’s total debt scaled by total asset at a point in time within a given fiscal year (Huang & Wang, 2015). This ratio identifies the amount of debt obtained to finance assets. Therefore, the higher the leverage ratio the more debt is acquired to finance banks asset. Hence, the researcher expects a negative relationship between leverage and risk-taking (as 52 University of Ghana http://ugspace.ug.edu.gh measured by the z-score) since banks will not be willing to risk the ownership of the firm to external parties (creditors, debenture holders). 𝑇𝑜𝑡𝑎𝑙 𝐷𝑒𝑏𝑡 𝐿𝐸𝑉 = 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡 3.4 Model Framework and Design This framework seeks to confirm the relationship between corporate governance and the risk- taking behaviour of banks in the sub-Saharan region of Africa. With this, panel regression model involving a pool of observations on a cross-section of variables over the period from 2006 to 2013 is employed for this study. This is because analyzing panel data involves “the pooling of observations on a cross-section of units over several time periods and provides results that are simply not detectable in pure cross-sections or pure time-series studies” (Abor, 2007: 86). Generally, the study used the regression model to examine the effect of corporate governance on the risk appetite of banks. The framework proposes that good corporate governance practices of a bank will have a significant impact on the bank risk behaviour. As we seek to address the overall objective of the research, that is, to investigate the extent to which corporate governance influences the risk-taking behaviour of banks in sub-Sahara Africa. A general regression model is formulated following from Anderson and Fraser (2000); Bokpin (2016) and Pathans (2009) to test empirically the three main hypothesis, H1, H2, H3, as discussed in literature; 𝐼𝑛(𝑅𝐼𝑆𝐾)𝑖𝑡 = 𝛽1(𝑆𝐵)𝑖𝑡 + 𝛽3(𝐴𝑄)𝑖𝑡 + 𝛽4(𝐶𝐸𝑂)𝑖𝑡 + 𝛽𝑛(𝐶𝑂𝑁𝑇𝑅𝑂𝐿𝑍)𝑖𝑡 + 𝜀𝑖𝑡……….eqn. 1 53 University of Ghana http://ugspace.ug.edu.gh From the above SB represents the strong boards, AQ the quality of audit, CEO power as they affect the risk-taking behaviour of banks (RISK). The model examines the impact of the various corporate governance variables on the risk-taking behaviour of banks. This is specified to determine the effect of corporate governance on the Z- score among countries in Sub- Saharan Africa. Broadly, we expand the model in equation 1 to include all the corporate governance variables as well as control variables. This is represented as; 𝑍𝑠𝑐𝑜𝑟𝑒𝑖𝑡 = 𝛽1𝐼𝑛(𝐵𝑆)𝑖𝑡 + 𝛽2(𝐵𝑆𝑆)𝑖𝑡 + 𝛽3(𝐵𝐼𝑁𝐷)𝑖𝑡 + 𝛽4(𝑁𝑂𝑁𝐸𝑋𝐸𝐶)𝑖𝑡 + 𝛽5(𝐴𝑄)𝑖𝑡 + 𝛽6(𝐶𝐸𝑂𝐷)𝑖𝑡 + 𝛽7(𝐶𝑉)𝑖𝑡 + 𝛽8(𝐶𝐴𝑃)𝑖𝑡 + 𝛽9(𝐵𝑆𝐼𝑍𝐸)𝑖𝑡 + 𝛽10(𝑅𝑂𝐴)𝑖𝑡 + 𝛽11(𝑅𝑂𝐸)𝑖𝑡 + 𝛽12(𝐿𝐼𝑄)𝑖𝑡 + 𝛽13(𝐿𝐸𝑉)𝑖𝑡 + 𝜀𝑖𝑡………..…eqn. 2 Where: Zscore the explained variable signifies the Risk-taking behaviour of Banks in Sub Saharan Africa, BS denotes Board Size, BIND is the Board Independence, CEOD is the CEO Duality, and AQ is the Audit Quality. The subscript i denotes the specific banks (i=1, 2,…, 434), t the time frame (t=2006, 2007, …, 2013), ln is the natural logarithm 𝜷1,2…10 are the parameters to be estimated with 𝜺 being the idiosyncratic error. Following from other studies (see Anderson &Fraser, 2000; Demsetz et al., 1997; Pathans, 2009; Saunders et al., 1990) the researcher includes CV which represents Charter value, CAP is the Bank Capital, BSIZE as Bank Size and ROA representing the Return on Capital Employed. In addition, 54 University of Ghana http://ugspace.ug.edu.gh bank variables such as ROE representing Return on Shareholders’ Equity, LIQ, Liquidity and LEV which is a proxy for Leverage are also added to beef up the control variables. 3.5 Econometric Techniques 3.5.1 Random Effect (RE) and Fixed Effect (FE) Estimators Random-Effect show expects that information to be examined is drawn from a progressive system of various populaces whose distinctions identify with that pecking order. The justification for random-effect is that variations across institutions are not related to the independent variables included in the model: “…the crucial distinction between fixed and random effects is whether the unobserved individual effect embodies elements that are correlated with the regressors in the model, not whether these effects are stochastic or not” (Green 2008, p. 183) The RE is always utilized when the model difference across the entities have some influence on the dependent variable. The underlined advantage of the RE is the inclusion of time-invariant variables which are not so with the fixed-effect (Wooldridge, 2002). Also, RE assumes that the random error term is not correlated with the independent variables. RE specify all the specific characteristics which may or may not influence the explanatory variables. The only disadvantage here is that a few factors may not be accessible along these lines prompting precluded variable predisposition in the model. RE permits to sum up the inference beyond the sample used in the model (Baayen, Davidson & Bates, 2008). 55 University of Ghana http://ugspace.ug.edu.gh The fixed-effect, on the other hand, is utilized when examining the influence of variables which vary over time. With this, FE examines the relationship between an independent and the dependent variables within an entity. Also, in employing the FE, it is assumed that variables within may impact on the other and as a result the need to control for this. This justification supports the rationale underlined in the econometric assumption of the correlation between independent variables (Wooldridge, 2002). FE best eliminates the effect of all the time-invariant characteristics to help assess the net effect of the independent variables on the dependent variables. Additionally, another important underlining assumption of the FE model is that the time-invariant characteristics are unique to the individual and should be correlated with individual characteristics (Baltagi, Bresson & Pirotte, 2003). On the contrary, when each entity is different therefore the random error team and the constant should not be correlated with other variables. If the error terms are correlated, then FE is not suitable and as a result, the inferences may not be correct to model that relationship (Wooldridge, 2002). 3.5.2 The Hausman test of Comparison- RE and FE Estimators Durbin Wu-Hausman in the year 1978 suggested a test in light of the distinctions that exist between the RE and FE estimates. Basically, it is basic to have a philosophy for testing presumptions in thinking about the decision between RE and FE. The Hausman test of correlation gives the appropriate estimator to be done. That is, in other to choose the FE and RE for a model, a Durbin- Wu-Hausman test (otherwise called Hausman specification test) where the null hypothesis is that the favoured model is random as against the alternative, the FE (see Green, 2008, section 9) is preferred. In total, this test assesses the consistency of an estimator when contrasted with an option. Basically, the test assesses if the measurable model relates to the data set (Baltagi et al., 2003). 56 University of Ghana http://ugspace.ug.edu.gh 3.6 Ordinary Least Squared (OLS) The Ordinary Least Square (OLS) is an econometric technique for evaluating the obscure parameters in the linear regression model with the point of diminishing the whole of squares of the contrasts between the observed factors in a given data set (Harrell, 2015). The OLS estimator is predictable when the explanatory factors are exogenous, and consequently the ideal class of linear unbiased estimators when the errors are homoscedastic and serially uncorrelated (Wooldridge, 2002). Under these different conditions, the techniques for OLS gives least fluctuation mean unbiased estimation when the blunder have limited change. Under these, an extra asymptotic presumption is the mistakes are typically conveyed, subsequently, OLS is the most extreme probability estimator. The OLS is most used since it has the benefits of being the Best Linear Unbiased Estimates (BLUE). In any case, the OLS characteristics are regularly one-sided and may in outrageous cases pick wrong factors more than frequently expected under the irregular decision. It is in addition that the fluctuation covariance network ends up solitary as factors approach zero and along these lines, the GLS might be best (Rzhetsky and Nei, 1992). 3.7 Generalized Least Squared (GLS) With the shortcoming in the OLS as expressed above, the examination received a numerous relapse strategy with reaction variable risk-taking behaviour (z-score) and the explanatory factors comprising of board quality, CEO duality and the presence of autonomous evaluators (auditors) of bank financial reporting related detailing in agreement to set models. The econometric reason for utilizing the numerous relapse display is because of the accompanying underlined presumptions. 57 University of Ghana http://ugspace.ug.edu.gh It is normal that the arbitrary idiosyncratic noise of the model is factually autonomous and indistinguishably circulated by the ordinary conveyance with a consistent variance (homoscedasticity). The error terms are likewise expected to be straight. The study sought to use the Generalized Least Square (GLS) method as underlined the estimation in the regression procedure since it delivers a superior coefficient (BLUE) when contrasted with the OLS that are vital basic for the model following Baltagi and Wu (1999) and Pathan (2009) methodology. As indicated by them, this econometric philosophy is extremely powerful to first-order autoregressive (AR (1)) unsettling influence inside unbalanced-panel relationship and heteroscedasticity across panels. In utilizing the General Least Squares (GLS) estimation system, the study analyzed successfully the impact of corporate governance observing component on the risk-taking conduct of banks. GLS is the econometric technique for evaluating the parameters that are behaving in a linear regression framework (Anselin, 2013). This is finished with the point of limiting the aggregate of the square errors; the contrasts between the reactions that are seen in a particular or given dataset and those that are anticipated by a linear. The linear function is the explained variable communicated as a direct set of the combination of logical factors. The GLS strategy was picked as the fitting technique for the data estimation because of its effectiveness in limiting the errors in expectation. 3.8 GLS-Random Effect The research out a Hausman particular test for covariance in other to decide between the two, which estimation strategy (FE and EF) is the most appropriate for the information. Subsequently 58 University of Ghana http://ugspace.ug.edu.gh the invalid theory was defined (H0= cov {xi, ɛ) =0). This connoted the covariance of the coefficients and the error term is equivalent to zero and thus both the RE and FE are steady estimators. After the test when the hypothesis was achieved, it implied that the RE was best since the standard of RE is not as much as that of FE. Despite what might be expected, with a p-value of 0.0035 (see section 4) the determination proposed that since the p-value< alpha of 0.05 we are to dismiss the null hypothesis that both FE and RE are predictable estimators. By so doing, the study embraced the FE for this examination. in the presence of bank unobserved FE, panel fixed- effect estimation is the most prescribed (Wooldridge, 2002, pp. 265– 291). In any case, such fixed-effect technique for variable estimation is not well fit for this study for few reasons. In the first place, the FE econometric estimation strategy requires a significant "within bank variation" of the factors esteems to create reliable and productive appraisals (Pathan, 2009). At the point when the critical factors that are the different independent factors do not vary considerably after some time, similar to the board characteristics, (for example, the bank board size) in this research, the FE assessments would be erroneous (Wooldridge, 2002, p. 286). Also, Baltagi (2005) recommend that the FE estimates may intensify the impact of multicollinearity if comprehend with least squares dummy factors. Taking everything into account, be that as it may, for a vast number of bank perception utilized (i.e. roughly 3,521) in this study and a time period of 8 years, the FE econometric estimation will again be conflicting (Baltagi, 2005). Thus, in instances of substantial perception the FE econometric estimation may prompt a huge loss of degrees of freedom. Thus, an elective strategy appropriate for the study as proposed by Pathan (2009) is the Generalized Least Squared Random-Effect. 59 University of Ghana http://ugspace.ug.edu.gh 3.9 Diagnostic Tests To set up a solid estimated model sufficient to upgrade our depiction of the information the study progressed to perform some diagnostics. The underlining established linear regression model is obviously based on the various hypothetical suppositions involving a direct relationship, the test for normality, no or little multicollinearity, no auto-correlation and homoscedasticity. In other to effectively accomplish the principle goal of the study the accompanying diagnostic tests were conducted on the data. 3.9.1 Independence Test Globally, one of the ordinarily known suppositions in linear regression examination requires that there exist no or little autocorrelation among factors in the dataset. Autocorrelation (likewise known to be independence test) happens when the idiosyncratic error terms are not autonomous from each other. The Durbin – Watson statistical technique (1.5VIF>5 or 10. Once more, the Hausman test gave a p-value>chi2 suggesting that we reject the null hypothesis of no relationship. The study considered the GLS- Random effect in light of the contention brought up in section three. The outcomes for the model diagnostics are given below; Table 4.3: Model Diagnostics Hausman (b-B)'[(V_b-V_B)^(- 1)](b-B) chi2(4) 15.670 Prob>chi2 0.0035 Mean VIF 4.5900 Source: Author’s computations 2017. 68 University of Ghana http://ugspace.ug.edu.gh 4.3 Empirical Results From the model specifications, the researcher regressed the independent variable against the dependent as a process in finding the impact of corporate governance mechanism on the risk-taking behaviour of banks in Sub-Saharan Africa. Here, we show the resultant effects of the risk-taking behavior of banks (measured with Z-score) with the board characteristics variables (measured with board size, board independence and the non-executive members), the influence of a bank’s chief executive officer (measured with CEO duality) and the significance and quality of an external body of evaluators (measured as audit quality). The resultant effect from the GLS-Random effects with Z-score as the dependent variable are represented below in table 4.4 69 University of Ghana http://ugspace.ug.edu.gh Table 4.4: GLS-Random effect with Z-score as the dependent variable zscore Coef. Std. Err. z P>z [95% Conf. Interval] ceod -2.797 4.416 -0.63 0.526 -1.453 5.859 logbs -5.409 0.139 -0.76 0.449 -1.401 8.583 logbss 2.049 0.216 0.39 0.694 -0.173 2.272 bind -4.345 1.245 -4.97 0.000*** -3.982 -4.707 nonexec 5.898 0.975 5.28 0.000*** 3.387 9.409 aq -0.010 3.262 -2.15 0.032** -1.404 -0.617 bsize -0.542 1.602 -0.34 0.735 -3.681 2.598 cv -2.151 1.473 -0.53 0.598 -1.036 3.735 bcap 0.782 0.329 1.23 0.219 -1.623 2.187 roa -0.615 0.238 -2.59 0.010*** -1.082 -0.149 roe 0.065 0.029 2.23 0.026** 0.008 0.122 liq 0.002 0.001 3.09 0.002*** 0.001 0.004 lev -0.218 0.104 -2.11 0.035** -0.421 -0.015 _cons 20.579 21.403 0.98 0.325 -20.170 21.327 R-sq: within 0.310 Obs 3035 between 0.367 overall 0.304 Source: Author’s computations 2017. The table shows the GLS-RE effect with the Z-score. It shows the coefficient, the standard error, the p-values and the 95% confidence intervals. Z-score the dependent variable represents the risk behaviour of banks in sub Saharan Africa, bs denotes board size, bind is the board independence, ceod is the ceo duality, and aq is the audit quality. cv which represents charter value, cap is the bank capital, bsize as bank size and roa representing the return on capital employed, as well as roe representing return on shareholders’ equity, liq, liquidity and lev which is a proxy for leverage makeup the control variables. The superscript ** and *** indicates statistical significance of 5% and 1% respectively. 70 University of Ghana http://ugspace.ug.edu.gh To continue, we show results from the econometric estimations in much accordance to the research objectives. The econometric results from the regression model with the Risk (Z-score) being the dependent variables are shown in the econometric Table 4.4. The goodness-of-fit of the model specifications show that there is a significant linear relationship between the independent variables and the dependent variable. Also, the probability chi-squared shows a p-value of 0.000 suggesting that we do not reject the null hypothesis of no linear relationship between the dependent and independent variables and conclude by establishing that there exists a significant relationship between the dependent and independent variables. Although, the R-square in the econometric model provides that the predictor variables together account for less than 50% (that is, 30%) of the variation in the dependent variable. The researcher is of a strong belief that the variables drawn as independent variables are significant enough in explaining the variation in the risk-taking decisions of the banks within the Saharan banking sector. Conventionally, however, the literature documents other variables that in addition helps to explain the dependent variable well. These other control variables, as suggested by the literature are nonetheless not really the main focus. From the above GLS-Random Effect econometric estimation, all the variables with the exception of the squared of log of board size square (with a coefficient of 2.1), the number of non-executive directors on the board (coefficient, 5.9), return on shareholders fund (coefficient, 0.07), liquidity (coefficient, 0.002) and the capital value of banks (with its coefficient of 0.78) in the Sub-Saharan Africa are negatively correlated with the Z-score (risk), which suggest an inverse relationship with 71 University of Ghana http://ugspace.ug.edu.gh the dependent variable Z-score (risk). These independent variable with negative related to the Z- score (risk) includes CEO duality (-2.8), log of board size (-5.9), board independence (-4.4), audit quality (-0.01), bank size (-0.5), charter value (-2.2), the return on asset (-0.6) and leverage with a coefficient (-0.22). The output of the regression provides the results of the fitting multiple linear regression model to describe the relationship between the risk-taking behaviour of Sub-Saharan banks (Z-score) and independent variables (made up of Board, CEO and Auditor characteristics). The results from Table 4.3 provide a p-value less than 0.05, hence there exist a statistically significant relationship between the variable at the 95.0% confidence level. On the average, with an observation of 3035 banks in the Sub-Saharan Africa, the empirical results show that the CEO duality, the board size (log of board size and log board size squared), with the control variables bank size, charter value and bank capital were all statistically insignificant at the 95.0% confidence level with the p-values of (0.99) (0.82, 0.97), (0.32), (0.44), (0.99), (0.69) respectively. These suggest these variables have no significant relationship with risk-taking. On the contrary, independent variables such as the independent board, the number of non- executive members on a bank board, the quality of audit, return on asset and equity, liquidity and the bank leverage were all statistically significant at a 5% and 1% with the Z-score at the 95.0% confidence level with p-values (0.00), (0.00), (0.032), (0.01), (0.003), (0.002) and (0.005) respectively. These variables have a strong relationship or influence with the dependent variables since their p-values are less than 0.05. 72 University of Ghana http://ugspace.ug.edu.gh From the specified model, a 1% (percentage) change in the number of non-executive directors on a bank board in sub-Sahara Africa will lead to an over 5 points increase in the risk-taking behaviour of the bank. Similarly, a 1% percentage increase in the number of independent board members will lead to an over 4.4 point’s reduction in the risk-taking behaviour of the bank. In addition, the risk propensity of banks in Sub-Sahara decreases by a little over 2.1 to an increase in the bank board size. In the event of auditing, we also provide that the risk (as measured by the z-score) decrease close to 1% to increase in the audit quality. Further, a change in the influential power of a bank CEO will also lead to a less than the 3-point decrease in the risk-taking activities of the bank. 4.4 Presentation of Findings The study finds board size to insignificantly express the risk-taking decisions of banks in the sub- Sahara Africa. The coefficient was found to be positive, meaning that increased board size does not significantly lead to a more stable banking system. This is because a higher z-score implies a more stable bank. This finding substantiates the assertions that corporate boards exert some significant level of monitoring on management and that increasing number on board will mean that new expertise is brought on board to increase the effectiveness and quality of monitoring. Although not part of the results, we investigated the possible non-linearity in the relationship. The econometric outcome substantiates the hypothesis on the free-rider argument, which recommends that expanding the number of individuals on a board prompts majority of the board to loose and ride on the hard work of a minority of bank board members engaged in the activity. These findings which recommend that extensive board size is related with less risk-taking similar 73 University of Ghana http://ugspace.ug.edu.gh to that given by Pathan (2009) and Cheng (2008), as the two authors found small boards to be fundamentally related with more risk-taking. Again, the study found board independent (which is calculated as a fraction of the number of external non-executive directors to the total number of a bank boards) was negative and significantly related to the z-score variable. These results imply that increasing the number of the non-executive directors will lead to a decrease in the risk-taking behavior of banks. The finding contradicts with that of Pathan (2009) and serves as an evidence to reinforce the theory that suggest shareholders are risk lover and would prefer more risk to that of the managers (risk averse) especially within the banking industry where a substantial fraction of the capital structure is made up of debt financing. Non-executive directors are “those directors who are not directly responsible for the day to- day running of the firm” (Bopkin 2016, p. 62). However, the non-executive directors, unlike the executive directors, are not considered as part of the management board, however, have no incentives to work in the interest of Chief Executive Officer. The literature provides that the inclusion of external directors on bank boards, therefore, fortifies the monitoring role of bank board in aligning CEO behaviour with owners’ preference. An increase in the number of non-executive directors on the bank board will mean that there is an increase in the quality of governance and independence aimed at aligning the actions and dealings of managers to the desires of shareholders, owners and providers of bank equity funds. The results suggest that the more aligned managers’ actions are to that of the shareholders (by way of increase in non- executive directors), the riskier a bank becomes through a decrease in the independent variable. 74 University of Ghana http://ugspace.ug.edu.gh However, Pathan (2009) in his paper on strong boards, CEO power and risk-taking behaviour, found that independent directors are also sovereign of bank owners and do not support shareholder actions of desiring more risk. He adds that independent directors are much more apprehensive about rules/regulation on how to avoid unnecessary lawsuits due to excessive risk-taking. That notwithstanding, Sumner and Webb (2005) evidently found both proportion of outsiders and a board strength index to be directly related to growth in risk-taking. These findings contest the assertions of Pathan (2009), on the contrary, the researcher's findings affirm the statements of Pathan (2009) as we found a negative but significant relationship with the z-score. 4.5 Chapter Summary Comprehensively, the chapter presented an in-depth analysis and interpretation of data, descriptive and empirical results of the study in line with existing theories and empirical literature. The objectives of the study examined the relationship that existed between risk-taking decisions (z- score) and corporate governance characteristics (board, CEO and audit). The study measured risk using Z-score (measured as the asset plus the capital asset ratio divided by the standard deviation of the return on asset). The summary statistics showed a minimum of -12.2 and a maximum of 804.9. The banks that recorded the least level of Z-score are those that experience a higher level of financial insolvency and instability in their country. The correlation matrix showed that the Z-score was positively correlated with all the independent variables with the exception of CEO duality and board independence. Board independence and non-executive members also have significant (positive and negative) correlation with the Z-score. The multiple regression results show that Z-score has a significant negative effect on board independence but a positive significant effect of non- 75 University of Ghana http://ugspace.ug.edu.gh executive members. This implies that a rise in board independence as a percentage of non- executive members is more likely to decrease the risk-taking attitude of banks. 76 University of Ghana http://ugspace.ug.edu.gh CHAPTER FIVE SUMMARY, CONCLUSION AND RECOMMENDATION 5.0 Introduction This final chapter summarizes the key findings of the study. It also highlights crucial lessons drawn from the study based on conclusions from data generated and the estimated results from the study as well as literature reviewed in the research. The chapter concludes with recommendations and a gap for policy deliberation and future research. 5.1 Summary of Findings The research sought to empirically explore the level at which corporate governance impact on the risk-taking behavior of banks in sub-Saharan Africa. Specifically, the study sought to examine the overall impact and relevance of board strength (strong board), power held by the Chief Executive Officer (CEO power) and presence of quality auditors on influencing decisions on the risk-taking activities of banks in sub-Saharan Africa. Various banking institutions take some amount of risk, this is because risk in the banking industry is good. Ironically, there seem to some kind of conflict between the management body and shareholders of banks as to the level of risk-taking activities. Studies from the literature suggest that shareholders prefer to take excessive risk in other to achieve greater rewards as the theory of utility provide that the higher the risk, the higher the expected return, but on the contrary, their agent (management acting in trust of owner) prefer lower level of risk for the fear of losing their jobs. In the quest to achieve the research objectives, that is, assessing the impact and relevance of good corporate governance (as represented by the strength of the board, power held by the CEO 77 University of Ghana http://ugspace.ug.edu.gh and the influence of external auditors) on bank risk-taking behavior, the researcher extracts the bank-specific accounting data covering 442 banks in 25 sub-Saharan African countries over the period from 2006 to 2013, consisting of 3035 observations to arrive at the following results. The study finds board size (measured as a logarithmic function) to insignificantly explain risk- taking behavior of banks in sub-Saharan Africa. This means that no matter how large or small a bank board is, it has no significant influence on the risk-taking decisions of banks in Sub-Saharan Africa. This implies that the overriding effect as theoretically suggested to occur in too large a board was not found in the study. Additionally, this suggests that even though in practice a large board size is likely to strengthen the monitoring framework of putting management on their toes, the findings provide that in terms of risk-taking behavior the size of the board does not matter in real life. This finding is consistent with the studies from Tsorhe et al., (2011) and Trinh et al., (2015), as their studies carried out in Ghana and Vietnam. Further, the research found board composition made up of independent bank board members to be significantly and negatively related to the dependent variable. Relative to the findings from Tsorhe et al., (2011) that board strength has no significant relationship with credit risk as well as liquidity risk. Also, this finding is supported by Pathan (2009) as he also arrived at the same outcome. This therefore explains that more independent external directors on a bank board are also independent from bank shareholders which results in low risk-taking decisions among banks in sub-Saharan Africa. This may predominantly be due to the fact that independent directors are more sensitive to national regulatory compliance. As such, in the presence of continuous and close monitoring by regulator, most African managers of banks, irrespective of the ownership as well as some significant number of directors act more conservatively to avoid any lawsuit in case of any default just to maintain a good name in the books of shareholders. Non-executive members, on the other 78 University of Ghana http://ugspace.ug.edu.gh hand was significantly and positively related to risk-taking behavior (measured by the Z-score). This successfully affirms the theory in literature that shareholders are said to be risk lovers and hence will consider taking more risk. Similar results were shown by Bopkin (2016) and Pathan (2009) as they all found out that shareholders in Ghana and USA (given majority of its board composition made up of more non-executive directors) would prefer more risk. Surprisingly, although the CEO power as well as the financial assurance and advice from the Audit had negative impact on risk (the Z-score), the CEO variable was found to have no significant in relations to the risk-taking behavior whereas the quality of audit was found to be statistically significant with the risk-taking behavior. This finding is contrary to that of Beasley (1996) and Torshe et al. (2011). This shows that the quality of an audit is essential in influencing the risk- taking decision of banks in the Sub-Saharan of Africa. Therefore, the report from auditor’s suggestion on the true and fairness of financial statement is likely to induce shareholders to limit the probability of risk-taking. 5.2 Conclusion The study seeking to examine the risk-taking behavior of banks in sub-Saharan is far more important and motivating to consider as an area. Essentially, the problems associated with poor bank governance are severe and their failures have even more significant cost costs to bank owners than any other. This is because banks are “highly special” economic unit with distinguishing characteristics and roles in financial intermediation, payment systems, liquidity management, evidence and maturity. Banking institutions are evidently significant as they offer stricter supervisory roles in the governance of their depositors and borrowers; such as by reducing borrowers’ earning management decisions (Ahn & Choi, 2009). Therefore, the study sought to 79 University of Ghana http://ugspace.ug.edu.gh provide a deeper understanding while establishing the relationship between corporate governance and the risk-taking behavior of banks in Sub-Saharan Africa. The study considered reducing existing gap in literature through assessing the impact and relevance of board strength (measure by the board size and independence), power held by the Chief Executive Officer (measure by the CEO duality) and presence of quality auditors (measured with audit quality) on influencing decisions on the risk taking activities of banks in sub-Saharan Africa with a data set of 442 banks in 25 Sub-Saharan African countries within a period of eight years ranging from 2006-2013. Board size was not significantly in predicting the risk-taking decision of banks in sub-Saharan African. It supports the statement that corporate bank boards exercise a good degree of monitoring on management and that growing board size explains that other new expertise is introduced on board to upsurge the quality of the monitoring. The results do not substantiate the theory on the free rider debate, which suggest that increasing bank board size will lead to some lazy board members riding on the hard works of a few bank board members doing the job. Again, the study suggest board independence which is measured as a fraction of the total number of non-executive directors on bank boards in sub-Saharan Africa to be significant and negatively related to the dependent variable, whereas CEO power was not significantly related to risk-taking. These results provide that an increasing number of independence on bank board of directors will lead to a reduction in the risk-taking behavior of banks. These findings serve as evidence to reinforce the agency theory and suggest shareholders who are risk loving prefer more risk-taking than managers (risk averse) especially within the banking industry where substantial fraction of 80 University of Ghana http://ugspace.ug.edu.gh the capital structure is made up of debt financing. Finally, the study also provided that auditors (as measured as the quality of audit from the big four) had an influence on the risk-taking behavior of banks in Sub-Saharan Africa and as such it was statistically significant according to the Z-score. 5.3 Recommendations Motivated from the findings of the research, the researcher recommends the following for policy and academic consideration as well as providing room for future research. From the findings in this study, the researcher indicated that bank board structure is an important determining factor of bank risk-taking in Sub-Saharan Africa. Hence, board structure is considered a very instrumental tool in bank risk-taking and as a result industry regulators and governmental agencies should monitor banks within the region more intensely where owners and management interest are aligned in attempt to control their potential for excessively irresponsible risk-taking. This will help stabilize the financial structure of the economy, thereby restricting irresponsible risk-taking actions by the financial entities that might lead to the African economies to be more vulnerable to financial crisis. Additionally, the Central Banks of African countries should tighten regulations and money supply (through increasing the required reserve ratio etc.) in order to mop up excess amount of money in circulation or held by the banking institutions. This would limit unnecessary loan disbursement (in the quest to receive more in return) and any other riskier investments that could lead to non- performing loans and bank run within the banking environment. This would reduce the amount of money held by the banking sectors and in effect reduce irresponsible risk-taking attitude. 81 University of Ghana http://ugspace.ug.edu.gh 5.4 Limitations and Gaps for Further Studies The researcher limited the study to different mechanisms of corporate governance in banks in sub- Saharan Africa as the researcher considers the impact of internal mechanism to corporate governance on the risk appetite of banks. Also, the researcher’s study is constrained to the period of 2006- 2013 data to provide insights on how the corporate governance variables impact risk- taking behavior evolution due to market reforms. The study on the brighter side provides room for further studies. This study is limited to only the sub-Saharan African countries and hence further studies could increase the scope to consider the whole of Africa at large. Furthermore, other studies could move beyond using only the Z-score measure of risk to consider financial risk, credit risk, liquidity and capital risk in assessing the risk-taking behavior of banks in African. Also, a look into the comparison of the risk taking behaviours of listed and non-listed banks could be explored. 82 University of Ghana http://ugspace.ug.edu.gh REFERENCE Abor, J., & Adjasi, C. K. (2007). Corporate governance and the small and medium enterprises sector: theory and implications, Corporate Governance. The international journal of business in society, 7(2), 111-122. Abor, J., & Biekpe, N. (2007). Corporate governance, ownership structure and performance of SMEs in Ghana: implications for financing opportunities. The international journal of business in society, 7(3), 288-300. Adams, R.B., Mehran, H., 2008. Corporate performance, board structure and its determinants in the banking industry. Working Paper, Federal Reserve Bank of New York. Agoraki, M. E. K., Delis, M. D., & Pasiouras, F. (2011). Regulations, competition and bank risk- taking in transition countries. Journal of Financial Stability, 7(1), 38-48. Aguilera, R. V. (2005). Corporate governance and director accountability: An institutional comparative perspective. British Journal of Management, 16(s1). Ahn, S., & Choi, W. (2009). The role of bank monitoring in corporate governance: Evidence from borrowers’ earnings management behavior. Journal of banking & finance, 33(2), 425-434. Al-Baidhani, A. M. (2014). Review of Corporate Governance Bundle. Alfaro, L., & Chen, M. X. (2012). Surviving the global financial crisis: foreign ownership and establishment performance. American Economic Journal: Economic Policy, 4(3), 30-55. Anselin, L. (2013). Spatial econometrics: methods and models (Vol. 4). Springer Science & Business Media. 83 University of Ghana http://ugspace.ug.edu.gh Antunes, R., & Gonzalez, V. (2015). A production model for construction: A theoretical framework. Buildings, 5(1), 209-228 Arditti, F. D. (1967). Risk and the required return on equity. The Journal of Finance, 22(1), 19- 36. Baayen, R. H., Davidson, D. J., & Bates, D. M. (2008). Mixed-effects modeling with crossed random effects for subjects and items. Journal of memory and language, 59(4), 390-412. Baker, T., & Simon, J. (Eds.). (2010). Embracing risk: The changing culture of insurance and responsibility. University of Chicago Press. Baltagi, B. H., & Wu, P. X. (1999). Unequally spaced panel data regressions with AR (1) disturbances. Econometric Theory, 15(06), 814-823. Baltagi, B. H., Bresson, G., & Pirotte, A. (2003). Fixed effects, random effects or Hausman– Taylor?: A pretest estimator. Economics letters, 79(3), 361-369. Beasley, M. S. (1996). An empirical analysis of the relation between the board of director composition and financial statement fraud. Accounting review, 443-465. Berle, A., & Means, G. (1932). The modern corporate and private property. McMillian, New York, NY. Bhagat, S., & Bolton, B. (2008). Corporate governance and firm performance. Journal of corporate finance, 14(3), 257-273. Bliss, M. A. (2011). Does CEO duality constrain board independence? Some evidence from audit pricing. Accounting & Finance, 51(2), 361-380. 84 University of Ghana http://ugspace.ug.edu.gh Bokpin, G. A. (2016). Bank governance, regulation and risk-taking in Ghana. Journal of African Business, 17(1), 52-68. Booth, J. R., Cornett, M. M., & Tehranian, H. (2002). Boards of directors, ownership, and regulation. Journal of Banking & Finance, 26(10), 1973-1996. Brealey, M. (2004). Marcus. Fundamentals of Corporate Finance. Bushman, R. M., & Williams, C. D. (2012). Accounting discretion, loan loss provisioning, and discipline of banks’ risk-taking. Journal of Accounting and Economics, 54(1), 1-18. Cadbury, S. A. (2000). The corporate governance agenda. Corporate Governance: An International Review, 8(1), 7-15. Cain, M. D., & McKeon, S. B. (2016). CEO personal risk-taking and corporate policies. Journal of Financial and Quantitative Analysis, 51(1), 139-164. Chan, S. G., Chan, S. G., Koh, E. H., Koh, E. H., Karim, M. Z. A., & Karim, M. Z. A. (2016). The Chinese banks’ directors and their risk-taking behavior: A corporate governance and finance perspective. Chinese Management Studies, 10(2), 291-311. Chapelle, A., Crama, Y., Hübner, G., & Peters, J. P. (2008). Practical methods for measuring and managing operational risk in the financial sector: A clinical study. Journal of Banking & Finance, 32(6), 1049-1061. Cheng, S. (2008). Board size and the variability of corporate performance. Journal of financial economics, 87(1), 157-176. Chow, E. H., Lee, W. Y., & Solt, M. E. (1997). The exchange-rate risk exposure of asset returns. Journal of Business, 105-123. 85 University of Ghana http://ugspace.ug.edu.gh Colbert, J. L. (2002). Corporate governance: communications from internal and external auditors. Managerial Auditing Journal, 17(3), 147-152. Coles, J. L., Daniel, N. D., & Naveen, L. (2006). Managerial incentives and risk-taking. Journal of financial Economics, 79(2), 431-468. Crouhy, M., Galai, D., & Mark, R. (2000). A comparative analysis of current credit risk models. Journal of Banking & Finance, 24(1), 59-117. Cyert, R. M., & March, J. G. (1963). A behavioral theory of the firm. Englewood Cliffs, NJ, 2. Davis, G. F., & Adam Cobb, J. (2010). Chapter 2 Resource dependence theory: Past and future. In Stanford's organization theory renaissance, 1970–2000 (pp. 21-42). Emerald Group Publishing Limited. De Andres, P., & Vallelado, E. (2008). Corporate governance in banking: The role of the board of directors. Journal of banking & finance, 32(12), 2570-2580. Denis, D. K. (2001). Twenty-five years of corporate governance research… and counting. Review of financial economics, 10(3), 191-212. DeZoort, F. T., & Salterio, S. E. (2001). The effects of corporate governance experience and financial-reporting and audit knowledge on audit committee members' judgments. Auditing: A Journal of Practice & Theory, 20(2), 31-47. Dietrich, A., & Wanzenried, G. (2011). Determinants of bank profitability before and during the crisis: Evidence from Switzerland. Journal of International Financial Markets, Institutions and Money, 21(3), 307-327. 86 University of Ghana http://ugspace.ug.edu.gh Dong, Y., Meng, C., Firth, M., & Hou, W. (2014). Ownership structure and risk-taking: Comparative evidence from private and state-controlled banks in China. International Review of Financial Analysis, 36, 120-130. Dong, Y., Meng, C., Firth, M., & Hou, W. (2014). Ownership structure and risk-taking: Comparative evidence from private and state-controlled banks in China. International Review of Financial Analysis, 36, 120-130. Dow, G. K. (2003). Governing the firm: workers' control in theory and practice. Cambridge University Press. Drees, J. M., & Heugens, P. P. (2013). Synthesizing and extending resource dependence theory: A meta-analysis. Journal of Management, 39(6), 1666-1698. Eisenhardt, K. M. (1989). Agency theory: An assessment and review. Academy of management review, 14(1), 57-74. Etzioni, A. (1975). Comparative Analysis of Complex Organizations, Rev. Simon and Schuster. Fama, E. F. (1980). Agency Problems and the Theory of the Firm. Journal of political economy, 88(2), 288-307. Fan, J. P., & Wong, T. J. (2005). Do external auditors perform a corporate governance role in emerging markets? Evidence from East Asia. Journal of accounting research, 43(1), 35- 72. Fiador, V. O. (2013). Corporate governance and value relevance of financial information: evidence from the Ghana Stock Exchange". Corporate Governance: The international journal of business in society, 13(2), 208-217. 87 University of Ghana http://ugspace.ug.edu.gh Fosberg, R. H. (1989). Outside directors and managerial monitoring. Akron Business and Economic Review, 20(2), 24. García-Marco, T., & Robles-Fernández, M. D. (2008). Risk-taking behaviour and ownership in the banking industry: The Spanish evidence. Journal of Economics and Business, 60(4), 332-354. Gatev, E., Schuermann, T., & Strahan, P. (2007). How do banks manage liquidity risk? Evidence from the equity and deposit markets in the Fall of 1998. In The risks of financial institutions (pp. 105-132). University of Chicago Goergen, M. (2012). International corporate governance. Pearson Higher Ed. Goergen, M., Renneboog, L., & Khurshed, A. (2006). Explaining the diversity in shareholder lockup agreements. Journal of Financial Intermediation, 15(2), 254-280. Goldsmith, R. W. (1969). Financial structure and development (No. HG174 G57). Gomes, A. R., & Novaes, W. (2005). Sharing of control as a corporate governance mechanism. Gordy, M. B. (2000). A comparative anatomy of credit risk models. Journal of Banking & Finance, 24(1), 119-149. Gray, P., & Irwin, T. (2003). Exchange rate risk: Reviewing the record for private infrastructure contracts. Green, W. H. (2008). Discrete choice modelling. Handbook of Econometrics, 2. Greenwood, J., & Jovanovic, B. (1990). Financial development, growth, and the distribution of income. Journal of political Economy, 98(5, Part 1), 1076-1107. 88 University of Ghana http://ugspace.ug.edu.gh Guo, L., Smallman, C., & Radford, J. (2013). A critique of corporate governance in China. International Journal of Law and Management, 55(4), 257-272. Halling, M., & Hayden, E. (2006). Bank failure prediction: a two-step survival time approach. Harrell, F. (2015). Regression modeling strategies: with applications to linear models, logistic and ordinal regression, and survival analysis. Springer. Hausman, J. A. (1978). Specification tests in econometrics. Econometrica: Journal of the Econometric Society, 1251-1271. Hermalin, Benjamin E., and Michael S. Weisbach. "The effects of board composition and direct incentives on firm performance." Financial management (1991): 101-112. Herzberg, F. I. (1966). Work and the nature of man. Hillman, A. J., Withers, M. C., & Collins, B. J. (2009). Resource dependence theory: A review. Journal of management, 35(6), 1404-1427. Houston, J. F., & James, C. (1995). CEO compensation and bank risk Is compensation in banking structured to promote risk taking? Journal of Monetary Economics, 36(2), 405-431. Houston, J. F., Lin, C., Lin, P., & Ma, Y. (2010). Creditor rights, information sharing, and bank risk taking. Journal of Financial Economics, 96(3), 485-512. Huang, L., Chen, Q., Zhao, Y., Wang, W., Fang, F., & Bao, Y. (2015). Is elective cesarean section associated with a higher risk of asthma? A meta-analysis. Journal of Asthma, 52(1), 16-25. Huang, Y. S., & Wang, C. J. (2015). Corporate governance and risk-taking of Chinese firms: The role of board size. International Review of Economics & Finance, 37, 96-113. 89 University of Ghana http://ugspace.ug.edu.gh Jarrow, R. A. (2007). A critique of revised Basel II. Journal of Financial Services Research, 32(1), 1-16. Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of financial economics, 3(4), 305-360. Jiraporn, P., Chatjuthamard, P., Tong, S., & Kim, Y. S. (2015). Does corporate governance influence corporate risk-taking? Evidence from the Institutional Shareholders Services (ISS). Finance Research Letters, 13, 105-112. Karamanou, I., & Vafeas, N. (2005). The Association between Corporate Boards, Audit Committees and Management Earning Forecasts: An Empirical Analysis. Journal of Accounting Research, 4(3), 453-486 Kasekende, L. (2010). Developing a sound banking system in sub-Saharan African countries. African finance in the 21st century. Keasey, K., Thompson, S., & Wright, M. (Eds.). (1997). Corporate governance: Economic and financial issues. OUP Oxford. Keeley, M. C. (1990). Deposit insurance, risk, and market power in banking. The American Economic Review, 1183-1200. Kim, E. H., & Lu, Y. (2011). CEO ownership, external governance, and risk-taking. Journal of Financial Economics, 102(2), 272-292. Knechel, W. R., & Salterio, S. E. (2016). Auditing: Assurance and risk. Taylor & Francis. Kumar, N., & Singh, J. P. (2012). Outside directors, corporate governance and firm performance: Empirical evidence from India. Asian Journal of Finance & Accounting, 4(2), 39. 90 University of Ghana http://ugspace.ug.edu.gh Kyereboah-Coleman, A., Adjasi, K.D.C and Abor, J. (2005), “Corporate governance and Firm Performance: Evidence from Ghanaian Listed Firms”, Corporate Ownership and Control Journal, Vol. 4 (2), pp. 123-132. Laeven, L., & Levine, R. (2009). Bank governance, regulation and risk taking. Journal of Financial Economics, 93(2), 259-275. Li, H., Jahera Jr, J. S., & Yost, K. (2013). Corporate risk and corporate governance: another view. Managerial Finance, 39(3), 204-227. Li, J., & Tang, Y. I. (2010). CEO hubris and firm risk taking in China: The moderating role of managerial discretion. Academy of Management Journal, 53(1), 45-68. Lipton, M., & Lorsch, J. W. (1992). A modest proposal for improved corporate governance. The business lawyer, 59-77. Lybaert, N. (1998). The information use in a SME: its importance and some elements of influence. Small business economics, 10(2), 171-191. Montiel, P. J. (1995). Financial Policies and Economic Growth: theory, evidence and country specific experience from Sub-Saharan Africa. African Economic Research Consortium, Nairobi, KE. Moscadelli, M. (2004). The modelling of operational risk: experience with the analysis of the data Press. Collected by the Basel Committee. Mullineux, A. (2006). The corporate governance of banks. Journal of Financial Regulation and Compliance, 14(4), 375-382. 91 University of Ghana http://ugspace.ug.edu.gh Naceur, S. B., & Ghazouani, S. (2007). Stock markets, banks, and economic growth: Empirical evidence from the MENA region. Research in International Business and Finance, 21(2), 297-315. Olsen, F. E. (1997). Feminism in Central and Eastern Europe: Risks and Possibilities of American Engagement. The Yale Law Journal, 106(7), 2215-2257. Ongore, V. O., & Kusa, G. B. (2013). Determinants of financial performance of commercial banks in Kenya. International Journal of Economics and Financial Issues, 3(1), 237. Pahlevan Sharif, S., & Kyid Yeoh, K. (2014). Independent directors’ resource provision capability in publicly-listed companies in Malaysia. Park, S. (1997). Risk-taking behavior of banks under regulation. Journal of Banking & Finance, 21(4), 491-507. Pathan, S. (2009). Strong boards, CEO power and bank risk-taking. Journal of Banking & Finance, 33(7), 1340-1350. Pfeffer, J., & Salancik, G. R. (1978). The external control of organizations: A resource dependence approach. NY: Harper and Row Publishers. Powell, W. (1991). Expanding the scope of institutional analysis. The new institutionalism in organizational analysis, Chicago, 183-203. Rzhetsky, A., & Nei, M. (1992). Statistical properties of the ordinary least-squares, generalized least-squares, and minimum-evolution methods of phylogenetic inference. Journal of molecular evolution, 35(4), 367-375. 92 University of Ghana http://ugspace.ug.edu.gh Schaeck, K., Cihak, M., & Wolfe, S. (2009). Are competitive banking systems more stable?. Journal of Money, Credit and Banking, 41(4), 711-734. Shehzad, C. T., de Haan, J., & Scholtens, B. (2010). The impact of bank ownership concentration on impaired loans and capital adequacy. Journal of Banking & Finance, 34(2), 399-408. Spira, L. F., & Page, M. (2003). Risk management: The reinvention of internal control and the changing role of internal audit. Accounting, Auditing & Accountability Journal, 16(4), 640-661. Steenkamp, J. B. E., & Geyskens, I. (2012). Transaction cost economics and the roles of national culture: a test of hypotheses based on Inglehart and Hofstede. Journal of the Academy of Marketing Science, 40(2), 252-270. Talamo, G. (2011). Corporate governance and capital flows. Corporate Governance: The international journal of business in society, 11(3), 228-243. Trinh, T. H., Duyen, T. T., & Thao, N. T. (2015). The Impact of Corporate Governance on Financial Risk in Vietnamese Commercial Banks. International Journal of Economics and Finance, 7(7), 123. Trinidad, A. M. (2006). Goodbye Washington consensus, hello Washington confusion? A review of the World Bank's economic growth in the 1990s: learning from a decade of reform. Journal of Economic literature, 44(4), 973-987. Tsorhe, J. S., Aboagye, A. Q. Q., & Kyereboah-Coleman, A. (2011). Corporate Governance and Bank Risk Management in Ghana. University of Ghana Business School. 93 University of Ghana http://ugspace.ug.edu.gh Wilkinson, N. (2005). Managerial economics: a problem-solving approach. Cambridge University Press. Williamson, O. E. (1981). The economics of organization: The transaction cost approach. American journal of sociology, 87(3), 548-577. Yermack, D. (1996). Higher market valuation of companies with a small board of directors. Journal of financial economics, 40(2), 185-211. Zingales, L. (2008). Causes and effects of the Lehman Brothers bankruptcy. Committee on Oversight and Government Reform US House of Representatives. 94 University of Ghana http://ugspace.ug.edu.gh APPENDIX Figure 1.1: the mean z-score with exponential trend line for sub-Saharan banks Mean Z-scores of Banks y = 17.065e-0.153x 16 R² = 0.1752 14 12 10 8 Mean Expon. (Mean) 6 4 2 0 2006 2007 2008 2009 2010 2011 2012 2013 Source: Authors computations (2017) The figure above indicates the mean z-score of banks in sub-Sahara Africa for the period 2006- 2013. It provides that there is a continuous reduction in the average risk-taking behavior of banks in the sub-Saharan Africa during and after the experience of global financial crises. This suggest most banking institutions are shunning away from irresponsible risk-taking. 95 University of Ghana http://ugspace.ug.edu.gh Below outlines the list of the sampled countries in the Sub-Sahara, the number of banks and the percentage to the total sample size. Table 2.1: List of sampled countries and number of banks Number Country Number of Banks Percentage to Total Sample (%) 1 Benin 10 2.27 2 Botswana 13 2.95 3 Burkina Faso 9 2.05 4 Cameroon 12 2.73 5 Cote D’Ivoire 19 4.32 6 Ethiopia 24 5.45 7 Ghana 23 5.23 8 Kenya 43 9.77 9 Malawi 14 3.18 10 Mali 10 2.27 11 Mauritius 18 4.09 12 Mauritania 8 1.82 13 Mozambique 13 2.95 14 Namibia 8 1.82 15 Nigeria 31 7.05 16 Rwanda 8 1.82 17 Senegal 12 2.73 18 Sierra Leone 13 2.95 96 University of Ghana http://ugspace.ug.edu.gh 19 South Africa 34 7.73 20 Sudan 15 3.41 21 Swaziland 6 1.36 22 Uganda 17 3.86 23 Tanzania 33 7.50 24 Zambia 24 5.45 25 Zimbabwe 23 5.23 Total 440 100.00 Source: Research data 2017 97