Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Board Independence and Firm Performance: Evidence from ASX-Listed Companies Yi Wang A Thesis Submitted for the Degree of Doctor of Philosophy Faculty of Business and Enterprise Swinburne University of Technology August 2009 Abstract There is an explosion of research on corporate governance in the past two decades; two major corporate governance models have been identified in the literature - the outsider system in the U.K. and U.S., and the insider system in Germany and Japan. Commentators tended to favour the insider system during the 1980s, when the economies of Germany and Japan outperformed others, and tended to favour the outsider system in the 1990s, when the economies of the U.S. looked better. The Australian market is often described as forming a part of the Anglo-Saxon outsider model; some scholars, however, raised questions about this classification, and pointed out that the Australian market may have more in common with Germany and Japan in terms of corporate ownership and control. From 2003, Australian listed companies have been subject to new corporate governance guidelines, which identify independent directors as a key component of effective governance. It appears that these recommendations are based on the Anglo-Saxon model, whose focus is on the agency conflict between managers and shareholders. Matching the trend towards greater board independence, the research on the empirical link between board characteristics and firm performance, primarily from the US and more broadly in recent years, is growing; most of them use agency theory as their underlying theoretical arguments. The mixed evidence, however, suggests the need for an in-depth investigation. This study gives an examination of various theoretical perspectives on boards of directors, to enhance our understanding on the potential relationship between board independence and corporate performance. It is found that agency theory, stewardship theory and organizational portfolio theory offer different expectations. From an agency perspective, where the board of directors is more independent of management, company performance would be higher. Stewardship theory proposes that board of directors with a lower level of independence would lead to better performance. The organizational portfolio model, as a theory waiting for empirical testing, suggests that board independence would be both proactive and reactive to performance. To identify the effect of board independence on firm performance, and the effect of firm performance on board independence, this study follows an archival research approach using secondary data, which has been typically adopted by the research surrounding this ii topic. It may be the first Australian study which presents direct evidence on the potential consequences of the recently altered regulatory environment with respect to board composition and structure. It is also designed to overcome the methodological limitations identified in prior research to provide improved evidence in this field. The results indicate that, for Australian listed firms, there is no strong relationship between board independence, and past or subsequent performance. The evidence casts doubts on the hope that promoting board independence would add value to Australian corporations. It appears that appointing independent members to the boards may merely represent firms’ attempts to comply with institutional pressures, and therefore would not be linked to performance. It is suggested that, despite agency theory’s policy influence, whether certain corporate governance practices recommended by the theory would lead to better performance need to be empirically tested. The regulatory bodies in Australia and other economies should be mindful of the differences between the markets when they look for the solutions to their corporate governance issues. iii Declaration This thesis contains no material which has been accepted for the award to me of any other degree or diploma, except where due reference is made in the text of the thesis. To the best of my knowledge this thesis contains no material previously published or written by another person except where due reference is made in the text of the thesis. Yi Wang Faculty of Business and Enterprise Swinburne University of Technology 16 August 2009 iv Acknowledgements I would especially like to thank Dr Albie Brooks and Dr Judy Oliver for the time and effort, and undoubted frustration from time to time, in guiding me capably and professionally through this task. I would also like to thank my parents, Yungang Wang and Xuezhen Hu, whose encouragement and confidence gave me the strongest motivation to take up this challenge. A special thank you goes to my son, Yuqing Wang, who has had to live with me through the ups and downs of completing this study. v Table of Contents Page List of Appendices ix List of Figures x List of Tables xi List of Abbreviations xiii Chapter 1. Introduction 1 1.1 Aim of the Study 1 1.2 Context of the Study 2 1.3 Motivation 3 1.4 Contribution to Knowledge 4 1.5 Conceptual Framework 5 1.6 Research Method 7 1.7 Major Findings 8 1.8 Structure of the Thesis 9 Corporate Governance Reforms 11 2.1 Introduction 11 2.2 Corporate Governance Systems 12 2.3 Corporate Governance Standards 18 2.3.1 19 Chapter 2. 2.4 Recommendations in the U.K. 2.3.2 Listing Rules in the U.S. 24 2.3.3 29 Australian Guidelines Summary 33 Empirical Evidence 36 3.1 Introduction 36 3.2 Australian Evidence 37 3.3 Evidence from the U.S. 40 3.3.1 Cross-Sectional Studies 41 3.3.2 Event Studies 52 Evidence from Other Regions 54 Chapter 3. 3.4 vi 3.5 Chapter 4. Summary and Limitations 62 3.5.1 Australian Studies 63 3.5.2 Overseas Studies 65 Theoretical Development 71 4.1 Introduction 71 4.2 Theories of Board of Directors 71 4.2.1 Legalistic View 72 4.2.2 Agency Theory 74 4.2.3 Stewardship Theory 77 4.2.4 Resource and Strategy Theories 79 4.2.5 Organizational Portfolio Theory 81 4.3 Board Independence and Firm Performance 84 4.4 Testable Hypotheses 86 4.5 Summary 88 Research Method 90 5.1 Introduction 90 5.2 Research Approach 91 5.3 Sample and Data Collection 93 5.4 Research Variables 94 5.4.1 Measurement of Board Independence 94 5.4.2 Performance Measures 98 Chapter 5. 5.4.3 Control Variables 101 5.5 Data Analysis 104 5.6 Summary 110 Univariate Analysis 111 6.1 Introduction 111 6.2 Preliminary Statistics 113 6.3 Correlations: The Sample Period 2000-2003 115 6.4 Correlations: The Sample Period 2003-2006 117 6.5 Summary 119 Multivariate Analysis 121 Introduction 121 Chapter 6. Chapter 7. 7.1 vii 7.2 Regressions: Board Independence and Past Performance 122 7.3 Regressions: Board Independence and Subsequent Performance 130 7.4 Regressions: Board Independence and Firm Risk 141 7.5 Summary 144 Discussion and Conclusions 146 8.1 Introduction 146 8.2 Discussion of Findings 146 8.2.1 Board Independence and Past Performance 146 8.2.2 Board Independence and Subsequent Performance 147 8.2.3 Board Independence and Firm Risk 149 8.2.4 Summary of Other Findings 149 8.3 Conclusions and Recommendations 151 8.4 Limitations and Future Research 157 8.5 Summary 159 Chapter 8. References 161 Appendices 190 viii List of Appendices Number Page 1. Summary of Australian Research 190 2. Summary of Overseas Research 193 3. Theories and Hypotheses in Prior Research 204 4. Pearson Correlations: 2000-2003 214 5. Pearson Correlations: 2003-2006 220 6. OLS and Logit Regressions: Board Independence 226 and Past Performance (ROA) 7. OLS and Logit Regressions: Board Independence 227 and Past Performance (ROE) 8. OLS and Logit Regressions: Board Independence 228 and Past Performance (Shareholder Return) 9. OLS and Logit Regressions: Board Independence 229 and Past Performance (Tobin’s Q) 10. OLS Regressions: Full Board Independence 230 and Subsequent Performance 11. OLS Regressions: Audit Committee Independence 231 and Subsequent Performance 12. OLS Regressions: Nomination Committee Independence 232 and Subsequent Performance 13. OLS Regressions: Remuneration Committee Independence 233 and Subsequent Performance 14. OLS Regressions: Chairman Independence 234 and Subsequent Performance 15. OLS Regressions: Board Independence and Firm Risk ix 235 List of Figures Number Page 4.1. Theoretical Development: High Board Independence 86 4.2. Theoretical Development: Low Board Independence 87 x List of Tables Number Page 1.1 Relationships: Board Independence, Firm Performance and Firm Risk 7 5.1. Measures of Firm Performance 100 5.2. Measures of Control Variables 103 6.1. Abbreviations of Research Variables 112 6.2. Descriptive Statistics: Boards of Directors 113 6.3. Descriptive Statistics: Other Research Variables 114 6.4. Jarque-Bera Statistics 115 6.5. Pearson Correlations: 2000-2003 116 6.6. Pearson Correlations: 2003-2006 118 7.1. OLS and Logit Regressions: Board Independence 123 and Past Performance (ROA) 7.2. OLS and Logit Regressions: Board Independence 125 and Past Performance (ROE) 7.3. OLS and Logit Regressions: Board Independence 127 and Past Performance (Shareholder Return) 7.4. OLS and Logit Regressions: Board Independence 129 and Past Performance (TOBQ) 7.5. OLS Regressions: Full Board Independence 132 and Subsequent Performance 7.6. OLS Regressions: Audit Committee Independence 134 and Subsequent Performance 7.7. OLS Regressions: Nomination Committee Independence and Subsequent Performance xi 136 7.8. OLS Regressions: Remuneration Committee Independence 138 and Subsequent Performance 7.9. OLS Regressions: Chairman Independence 140 and Subsequent Performance 7.10. OLS Regressions: Board Independence and Firm Risk 143 8.1. Relationship between Board Independence and Past Performance 147 8.2 Relationship between Board Independence and Subsequent Performance 148 8.3 Results: Board Independence, Firm Performance and Firm Risk xii 152 List of Abbreviations Abbreviation AEOA Australian Employee Ownership Association AICD Australian Institute of Company Directors ASA Australian Shareholders’ Association ASX Australian Stock Exchange BRT Business Roundtable CEO Chief Executive Officer CFROTA CIMA Cash Flow Return on Total Assets Chartered Institute of Management Accountants EBIT Earnings before Interest and Tax EPS Earnings per Share EVA Economic Value Added HKSE Hong Kong Stock Exchange IBGC Brazilian Institute of Corporate Governance ICAEW Institute of Chartered Accountants in England and Wales IPO Initial Public Offering LSE London Stock Exchange MBT Market-to-Book Ratio MVA Market Value Added NACD National Association of Corporate Directors NED Non-Executive Director xiii NZSE New Zealand Stock Exchange NYSE New York Stock Exchange OECD Organization for Economic Co-operation and Development OLS Ordinary Least Squares R&D Research & Development ROA Return on Assets ROE Return on Equity S&P Standard & Poor’s SEC Securities and Exchange Commission SET Stock Exchange of Thailand SG&A Selling, General & Administrative TSE Toronto Stock Exchange xiv Chapter 1. Introduction 1.1 Aim of the Study With the publicity surrounding recent corporate collapses, the issues of board composition and structure generally and independent directors specifically, have become a fertile area of interest and research. There is a global movement to enhance board independence in public companies, initiated by the concern about agency conflict between managers and shareholders. The definitions of “independence” proposed by the regulatory bodies vary; it appears that they are sourced from the statement in the Cadbury Report (1992, Code 2.2) - an independent director “… should be independent of management and free from any business or other relationship which could materially interfere with the exercise of their independent judgement, apart from their fees and shareholding.” The purpose of this study is to test the applicability of several theories which make different predictions about the effect of board independence on firm performance and vice versa; address some of the limitations of prior studies, including small sample size, short-term observation of firm performance, limited control variables and performance measures, and simplistic dichotomy of inside and outside directors as an empirical proxy for board independence; and shed some light on the potential influence of a recently altered regulatory environment with respect to corporate governance mechanisms, particularly those relating to the requirements for a majority of independent directors on the board and board committees, on firm performance. However, it should be noted that the focus here is on the empirical correlation between board independence and firm performance, rather than the performance of boards of directors or individual directors. This chapter offers an overview of the thesis. Section 2 and 3 provide an introduction to the context and motivation for the research. There is a discussion in Section 4 of the contribution that it makes to the literature. Section 5 and 6 outline the conceptual 1 framework and method used to conduct this study, followed by a summary of the major findings in Section 7. The structure of the dissertation is described in the last section. 1.2 Context of the Study Australia had watched closely as big corporate failures such as Enron, WorldCom and Arthur Andersen had prompted the U.S. regulatory authorities to launch significant corporate governance reforms. To promote and restore investor confidence, in 2003 the Australian Stock Exchange endorsed Principles of Good Corporate Governance and Best Practice Recommendations released by its Corporate Governance Council, which reflect “best international practice” by highlighting the importance of board independence. The stock exchange requires that a majority of each listed company’s directors qualify as independent directors. Although it is for the board to decide in particular cases whether the definition of independence is met, there is a list of the categories of persons who should not be considered independent. It is also recommended that the roles of chairman and chief executive should not be exercised by the same individual; the chairman should be an independent director. Firms should establish an audit committee, nomination committee and remuneration committee, and a majority of each committee’s members should be independent. In the long-term the recommendations may result in a shift in the power of boards of directors in favour of independent directors, and away from management. The literature suggests that there are two major governance models around the world (e.g., Hall and Soskice, 2001; Denis and McConnell, 2003; Murphy and Topyan, 2005; Gillan, 2006); the first is the outsider system in the U.K. and U.S., in which the primary corporate objective is to maximize profit, and managers must ensure the firm is run in the interests of shareholders. From the agency perspective the main concern of corporate governance is the conflict between strong managers and weak dispersed shareholders. The second is the insider system in Germany and Japan, in which corporations must fulfil wider objectives and have responsibilities to parties other than shareholders; it is assumed that the basic conflict is between weak managers and minority owners, and strong majority owners (Bouy, 2005). 2 Australia’s system of corporate governance has been described as forming part of the Anglo-Saxon outsider model of ownership and control (Scott, 1997; Weimar and Paper, 1999; Campbell, 2002). Recently some academics raised questions about this classification, and argued that the Australian system might have more in common with the insider system (e.g., Lamba and Stapledon, 2001; Dignam and Galanis, 2004). Dignam and Galanis (2004) demonstrated that Australia was in the process of reforming its corporate governance based on an assumption that it is an outsider model; if that assumption is incorrect, recent reforms may have a destabilizing effect. 1.3 Motivation Scholars, in general, have taken two approaches to investigate the impact of board composition and structure on firm performance (e.g., Bathala and Rao, 1995; Lawrence and Stapledon, 1999; Bhagat and Black, 1999, 2000). The first approach is based on relating board composition and structure to certain corporate events, such as executive turnover and remuneration, financial reporting, making or defending against a takeover bid, management buyout and shareholder litigation. It is believed that “[t]he principle weakness of this approach is that it cannot tell us how board composition affects overall firm performance. Firms with majorityindependent boards could perform better on particular tasks, such as replacing the CEO, yet worse on other tasks, leading to no net advantage in overall performance” (Bhagat and Black, 1999, p.3). The second approach, which is the focus of this study, involves examining directly the link between board characteristics and financial performance - the “bottom line” of firm performance; as suggested by some authors (e.g., Bathala and Rao, 1995; Lawrence and Stapledon, 1999; Bhagat and Black, 1999, 2000), it may avoid the weakness inherent in the first group of studies. However, after a survey of some Australian and overseas studies, it is concluded that prior research does not establish a clear correlation between board composition and structure, and financial performance. The recent corporate governance reforms, and the resulting pressure on public companies for greater board independence, suggest the need for a further investigation in the Australian context. 3 So far there is no direct empirical evidence in Australia supporting the introduction of “best practice recommendations”. As noted by Lawrence and Stapledon (1999), unquestioned acceptance of overseas practice may result in proposals for regulatory reform which are not suited to the local environment; the regulatory requirements imposed on companies may add to the compliance costs for these companies, and, indirectly, their shareholders. Therefore, it may be necessary to ask whether the costs of imposing governance regulations on all listed companies would be outweighed by the benefits; in this research the consequences of “best practice recommendations” supported by the stock exchange are examined, which may provide some feedback to regulatory authorities, corporations, investors and other stakeholders about the effect of such recommendations on corporate performance. 1.4 Contribution to Knowledge The findings of this study may contribute to current research in several ways. First, it adds to the growing literature on this topic in Australia (e.g., Muth and Donaldson, 1998; Calleja, 1999; Lawrence and Stapledon, 1999; Cotter and Silverster, 2003; Kiel and Nicholson, 2003), and provides updated information on board composition and structure of public companies in this country. Second, the majority of prior studies use agency theory as their underlying theoretical arguments, suggesting that this theory promises a positive impact of board independence on performance (e.g., Fosberg, 1989; Muth and Donaldson, 1998; Cotter and Silverster, 2003; Krivogorsky, 2006; Chan and Li, 2008). Therefore they do not address whether board characteristics are endogenously related to performance. Although several researchers explored this concern, there is little theoretical support in their studies; their workings are best viewed as exploratory data analysis, rather than as testing of formal hypotheses (e.g., Hermalin and Weisbach, 1988; Denis and Sarin, 1999; Bhagat and Black, 2000). In this project an introduction to various theoretical perspectives on boards of directors is given, which may enhance our understanding on the potential relationship between board independence and firm performance. Specifically, testable hypotheses are developed from agency theory, stewardship theory 4 and organizational portfolio theory, which are outlined in the next section, to answer the following research questions: does board independence have any influence on firm performance among Australian listed companies? and does firm performance have any influence on board independence among Australian listed companies? Third, it is found in the literature review that most Australian and overseas studies suffer from a range of research limitations, including: small sample size (e.g., Daily and Dalton, 1992; Lawrence and Stapledon, 1999; Dulewicz and Herbert, 2004; Krivogorsky, 2006); short-term observation of firm performance (e.g., Molz, 1988; Vafeas and Theodorou, 1998; Cotter and Silverster, 2003; Chan and Li, 2008); limited performance measures (e.g., Baysinger and Bulter, 1985; Denis and Sarin, 1999; Kiel and Nicholson, 2003; Luan and Tang, 2007); limited control variables (e.g., Fosberg, 1989; Barnhart and Rosenstein, 1998; Calleja, 1999; Chang and Leng, 2004); and simplistic dichotomy of inside and outside directors as a measure for board independence (e.g., Kesner, 1987; Agrawal and Knoeber, 1996; Muth and Donaldson, 1998; Randoy and Jenssen, 2004). As shown later in Section 1.6, this study is designed to overcome the above limitations to provide improved evidence in the field. 1.5 Conceptual Framework Theoretical support for the belief that independent boards would enhance shareholder returns has been provided by agency theory (e.g., Muth and Donaldson, 1998; Vafeas and Theodorou, 1998; Cotter and Silverster, 2003). A central assumption of the theory is that managers may pursue their own goals rather than seek to maximise shareholder wealth, unless their discretion is kept in check by a vigilant, independent board (e.g., Jensen and Meckling, 1976; Stroh, Brett, Baumann and Reilly, 1996; Daily, McDougall, Covin and Dalton, 2002). By emphasising the potential for divergence of interests between investors and managers, most empirical research in this field assumes that, 5 where board of directors is more independent of management, company performance would be higher. In addition, some agency theorists assert that managers, unlike shareholders, could not readily diversify their employment risks across a range of investments, as a result they tend to be more risk averse than may be in the interests of shareholders (e.g., Fama, 1980; Knoeber, 1986; Prentice, 1993). Developed as an alternative to agency theory, stewardship theory highlights a range of non-financial motives for managerial behaviours, such as the need for achievement, intrinsic satisfaction of successful performance, and respect for authority and work ethics, which have been identified in the organizational literature (e.g., McClelland, 1961; Herzberg, 1966; Etzioni, 1975). Having control empowers managers to maximize corporate profits; the detailed operational knowledge, expertise and commitment to the firm by executive directors would make firms with a management-dominated board more profitable (e.g., Donaldson and Davies, 1991, 1994; Fox and Hamilton, 1994; Davis, Schoorman and Donaldson, 1997). This prediction is tested in a number of papers (e.g., Muth and Donaldson, 1998; Kiel and Nicholson, 2003; Randoy and Jenssen, 2004). According to organizational portfolio theory proposed by Heslin and Donaldson (1999) and Donaldson (2000), an increase in corporate profitability would enhance the perceived integrity and competence of managers, thereby precipitating boards in which managers are increasingly represented. Poor performance would lead to boards that are more independent of management; the risk-averse governance delivered by independent directors would prevent long-term growth and profitability, thus leading to a gradual decline in organizational performance. This theory has received little attention from academics, and therefore yet to be tested. A summary of the relationships between board independence, firm performance and risk, as expected by agency theory, stewardship theory and organizational portfolio theory, is presented in the following table. 6 Table 1.1 Relationships: Board Independence, Firm Performance and Firm Risk Relation Board Independence Past Firm Performance Agency Theory Unknown Stewardship Theory Unknown Organizational Portfolio Theory Negative Subsequent Firm Performance Agency Theory Positive Stewardship Theory Negative Organizational Portfolio Theory Negative Subsequent Firm Risk Agency Theory Positive Stewardship Theory Unknown Organizational Portfolio Theory Negative 1.6 Research Method This study uses an archival research design which is traditionally employed by the literature surrounding this topic. Most Australian studies suffer from the limitation of small sample size. Muth and Donaldson (1998) suggested that a sample size closer to 200 would have been preferable; effects of boards on performance tend to be small which means that more statistical power is needed to detect significant relationships. Calleja (1999) also acknowledged that the small sample size made it difficult to reach any firm conclusions. In this research a sample of 243 firms from the 2003 Australian top 500 is used; the sources of data are available within the public domain. The most popular measurement of board independence in prior research is the proportion of non-executive directors or independent directors on the board. Based on the “best practice recommendations” as outlined earlier, in this thesis five empirical proxies for board independence are adopted, i.e., full board independence represented by the proportion of independent directors on the board, monitoring committee 7 independence measured by the proportion of independent directors on the audit, nomination or remuneration committee, and chairman independence which is a dummy variable to assess whether or not the chairman is an independent director. There are four measures of firm performance – market-based measures of Tobin’s q and shareholder return, and accounting-based measures of return on assets and return on equity; they are the most frequently used performance measures in prior studies, as well as in the field of accounting and financial research (e.g., Hofer, 1983; Shrader, Taylor and Dalton, 1984; Devinney, Richard, Yip and Johnson, 2005). As noted by Hofer (1983), it is common to see several performance indices to be used because organizations legitimately seek to accomplish a variety of objectives, ranging from profitability to effective asset utilization and high shareholder returns. Board characteristics of sample companies are investigated at one point in time - mid2003. Some prior research on this topic suffers from the limitation of short-term observation of firm performance; Shrader et al (1984), in examining the literature on the empirical relationship between strategic planning and organizational performance, found that most studies had used 3- and 5-year periods as measures of long-range planning and performance. Thus the performance figures employed in this project are the three-year averages over the 2000-2003 and 2003-2006 financial years. As the basis for endogeneity testing, firm performance are modelled as both an independent variable (i.e., the effect of performance on board characteristics) and as a dependent variable (i.e., the effect of board characteristics on performance). Bathala and Rao (1995) suggested that the mixed evidence on the link between board composition and firm performance might be attributed to the omission of other variables that affect performance; Schellenger, Wood and Tashakori (1989) argued that the conflicting empirical findings with respect to the existence or non-existence of a board composition effect on financial performance could be due to failure to control risk. To minimize the above concerns some control variables are introduced into the data analysis, including board size, blockholder and managerial shareholdings, dividend payout, diversification, firm age, firm size, leverage and risk. 8 1.7 Major Findings Descriptive statistics, correlation analysis and regressions are conducted for the research variables. The results indicate that, for Australian public companies, there does not appear to be a strong relationship between board independence, and past or subsequent performance; the level of board independence does not affect firm risk. Additional findings include that companies with higher blockholder shareholdings tend to reduce the percentages of independent directors on the board, and audit and remuneration committees; larger firms have relatively more independent outsiders sitting on nomination and remuneration committees. Moreover, larger board or lower managerial shareholdings could lead to poor performance as measured by Tobin’s q; larger firms or firms with lower leverage have better shareholder return. It is found that smaller companies, companies with higher gearing or shareholder return, or companies with lower dividend payout may be riskier. 1.8 Structure of the Thesis The thesis consists of eight chapters, including this introductory chapter. An overview of the remaining chapters is presented below. Chapter 2 - Corporate Governance Reforms: this chapter provides an introduction to the corporate governance models frequently addressed by researchers; within this context the recent developments in corporate governance standards in the U.K., U.S. and Australia are evaluated. Chapter 3 - Empirical Evidence: a review of the literature in Australia and overseas, which gives evidence on whether board characteristics and firm performance are related, is undertaken; the focus is on cross-sectional studies on publicly listed companies. Chapter 4 - Theoretical Development: the evolving perspectives on the roles of board of directors are discussed. The potential relationships between board independence and firm performance, as proposed by different conceptual frameworks, are investigated; as a result six testable hypotheses are constructed. 9 Chapter 5 - Research Method: the chapter gives a description of the research design chosen to test the hypotheses, covering the general research approach, sample selection and data sources, measurement of variables, and analysis procedures. Chapter 6 - Univariate Analysis: this chapter shows descriptive statistics of the data collected; to explore the relationships between board independence, firm performance and risk, the correlation analysis for the research variables during the sample periods of 2000-2003 and 2003-2006 is produced. Chapter 7 - Multivariate Analysis: the results of regression models specified for the effect of firm performance on board independence, and the effect of board independence on performance and risk, are reported. Chapter 8 - Discussion and Conclusions: the findings emanating from Chapter 6 and 7 are further analysed; the analysis responds specifically to the research hypotheses and questions, and leads to the conclusions and recommendations. Limitations in the current study and future research opportunities are identified. 10 Chapter 2. Corporate Governance Reforms 2.1 Introduction As observed by Gillan (2006), the definition of corporate governance differs depending on one’s view of the world. Shleifer and Vishny (1997a, p.737) claimed that “[c]orporate governance deals with the ways in which suppliers of finance to corporations assure themselves of getting a return on their investment.” In Litch (2002), corporate governance is viewed as the rules and structures for wielding power over other people’s interests, including the use and abuse of power. Denis and McConnell (2003, p.2) defined corporate governance “… as the set of mechanisms – both institutional and market-based – that induce the self-interested controllers of a company … to make decisions that maximize the value of the company to its owners …” Becht, Bolton and Roell (2005, p.1) suggested that “[c]orporate governance is concerned with the resolution of collective action problems among dispersed investors and the reconciliation of conflicts of interest between various corporate claimholders”. They identified the following reasons why corporate governance had become such a prominent topic in recent years: the world-wide wave of privatization of the past two decades; the growth in pension fund and active investors; the wave of mergers and takeovers of the 1980s an 1990s; deregulation and integration of capital markets; the 1998 East Asia crisis, and a series of corporate scandals and failures in the U.S. According to Pettigrew (1992), corporate governance lacks any form of coherence, either empirically, methodologically or theoretically, with only piecemeal attempts to understand and explain how the modern corporation is run. Tricker (2000) suggested that corporate governance did not have an accepted theoretical base or commonly accepted paradigm, and the term “corporate governance” was scarcely used until 1980. Similarly, Murphy and Topyan (2005) found that researchers investigated corporate governance less as a planned, systematic inquiry, and more as a response to observed problems in corporations. As a result, corporate governance remains a collection of 11 disparate studies that lack collective coherence; there are significant disagreements in the field of governance research. Nevertheless, the amount of research on this topic has increased dramatically. In 2005, Gillan (2006) carried out a search of Social Science Research Network abstracts containing the term “corporate governance”, resulting in more than 3,500 hits. In July 2006, from the databases of Business Source Complete and Science Direct, 9,549 papers with the key words “corporate governance” in their abstracts were found; a literature survey of recent studies would be a daunting task. The past fifteen years have also witnessed a proliferation of guidelines and codes of “best practice” designed to improve corporate governance of public companies; this heightened international awareness of corporate governance has prompted some stock exchanges to encourage or mandate board independence among listed companies (Gregory, 2001a, b, c). The objective of this chapter is to present an overview of corporate governance systems frequently addressed by researchers in the past two decades; within this context, the recent developments in corporate governance standards in the U.K., U.S. and Australia are introduced in order to provide some background and context information for the current project. In Section 2 of this chapter the major systems in corporate governance research are introduced; the recommendations and listing rules, which have been used to promote board independence in the U.K., U.S. and Australia, are examined in Section 3, followed by a summary in Section 4. 2.2 Corporate Governance Systems Denis and McConnell (2003) noted that, the publication of Jensen and Meckling (1976), in which the authors applied agency theory to corporations and modelled the agency costs of outside equity, had produced voluminous works in the U.S; by the early 1990s, similar research in other countries began to appear. At first, the literature focused on other major world economics, such as Germany, Japan and the U.K; recent years, however, have witnessed an explosion of papers on corporate governance around the world, for both developed and emerging markets. 12 Some researchers pointed out that, in general, two main systems or models of governance had been identified and analysed in the literature (Aoki, 2001; Hall and Soskice, 2001; Denis and McConnell, 2003; Gillan and Starks, 2003; Becht et al, 2005; Aguilera, 2005; Jansson, 2005; Murphy and Topyan, 2005; Buoy, 2005). The first is the “outsider system” (Bouy, 2005), “market-based system” (Becht et al, 2005; Murphy and Topyan, 2005) or “shareholder model” (Jansson, 2005), which has been adopted in the U.K. and U.S. The second is the “insider system” (Bouy, 2005), “long-term large investor system” (Becht et al, 2005; Murphy and Topyan, 2005) or “stakeholder model” (Jansson, 2005) that has been employed by firms in continental Europe, Japan and Korea, among others. Detailed surveys of these models are available in Becht et al (2005), Murphy and Topyan (2005), Bouy (2005) and Jansson (2005), whose propositions are summarized below. The primary objective of the firm in the U.K. and U.S. is to maximize profit, and performance is appreciated by the market value of the firm (Gay, 2002; Murphy and Topyan, 2005; Bouy, 2005). The principal-agent relationship arising from the separation of ownership and decision-making may cause the firm’s behaviour to diverge from the profit-maximizing ideal; since the managers are not the owners of the firm, they can have other objectives rather than maximizing the shareholder wealth (Jensen and Meckling, 1976; Ross, 1987; Quinn and Jones, 1995; Shankmann, 1999). An effective corporate governance framework is needed to minimize agency costs. In an outsider system equities represent a large proportion of financial assets and GDP, and there are developed investment banking and securities markets (Hall and Soskice, 2001; Denis and McConnell, 2003; Becht et al, 2005). The stock market may be dominated by institutional investors, due to the tax incentives to collective schemes, growth of mutual funds, and tendency for firms to issue shares directly to institutional investors; institutional investors may have incentives to monitor management and serve as a control mechanism (Denis and McConnell, 2003; Bouy, 2005; Aguilera, 2005). The outsider systems in the U.K. and U.S. are also characterized by widely dispersed share ownership and high turnover (Gay, 2002; Denis and McConnell, 2003; Becht et al, 2005). As the power of shareholders to select directors and vote on key issues of the company is limited by the fragmentation of ownership, regulatory bodies have to offer adequate shareholder protection and allow investors to assume the risk-reward trade-off 13 with an equal access to information (Perotti and Von Thadden, 2003). Consequently the main concern of corporate governance is the conflict between strong managers and weak dispersed shareholders, and in this spirit the role of directors, stock options, takeovers, minority shareholder protection are frequently investigated by researchers (Becht et al, 2005; Bouy, 2005; Jansson, 2005). As the board of directors is responsible for monitoring managerial performance and preventing conflicts of interests, it must have some degrees of independence from management; however, board independence often poses a problem in reality and the board is regarded as a relatively weak governance device (Hermalin and Weisbach, 2003; Denis and McConnell, 2003; Bouy, 2005). It is the capital markets that play a primary role in corporate governance. When managers fail to maximize the firm’s value, they expose it to the threat of a take-over; the market for corporate control may be a more effective disciplinary device than either the monitoring by institutional investors or board of directors (Holmstrom and Kaplan, 2001; Becht et al, 2005; Murphy and Topyan, 2005). Thus this model is termed as the “market-based” or “market-oriented” system in Becht et al (2005) and Murphy and Topyan (2005); the intensity of mergers and acquisitions in the U.K and U.S. could be justified by rent seeking behaviour, empire building and tax minimization. In an insider system or stakeholder model such as continental Europe, Japan and Korea, corporate ownership is typically concentrated among a stable network of strategically orientated banks and firms, rather than fragmented among individuals and financialorientated institutional investors (Claessens, Djankov, Fan and Lang, 1998; Hansmann and Kraakman, 2001; Hall and Soskice, 2001; Franks and Mayer, 2001; Wojcik, 2001); the market for corporate control is largely non-existent. Instead, banks play the central external governance role through relational financing, commingling debt and equity, providing financial services and monitoring in times of financial distress (McCauley and Zimmer, 1994; Fukao, 1995; Gay, 2002; Becht et al; 2005). Corporations must fulfil wider objectives and have responsibilities to parties other than shareholders; the “best” firms are the ones with committed suppliers, customers and employees, with corporate governance “coalitions” built among banks, long-term investors, employees and management (Bouy, 2005; Jansson, 2005). 14 Employees could exercise voice within corporate governance, for example, through legal rights to co-determination in Germany or extensive use of joint labourmanagement consultation in Japan; this role of employees is reflected in long employment tenures, infrequent use of lay-offs and high investment in firm-specific skills. Top managers tend to be internally promoted, and managerial compensation is much closer to average employees’ schemes and lack of strong shareholder-oriented incentives such as stock options; as a result managers are supposed to be less financeoriented and focus on long-term product strategy (Aoki, 2001; Hall and Soskice, 2001; Jackson, 2001; Jackson and Moerke, 2005). Bouy (2005, p.39) suggested that, for the stakeholder model, “the basic conflict is between ‘strong voting blockholders, weak minority owners’ or ‘weak managers, weak minority owners, strong majority owners’”; although there is little empirical evidence to support this insight. As found by Hansmann and Kraakman (2001), Becht et al (2005) and Murphy and Topyan (2005), which of the two models has been favoured by scholars has varied over time as a function of the relative success of each country’s underlying economy. The German and Japanese stakeholder perspective had been regarded as strengths relative to the Anglo-Saxon shareholder perspective in the 1980s, when Germany and Japan had outperformed the U.S. From the late 1990s, following a decade of recession in Japan and post-unification adjustments in Germany, and an economic and stock market boom in the U.S., the American corporate governance model has been hailed as the path for all to follow. In the 1980s Germany and Japan had a lower cost of capital, which is assumed to be the result of close relationships between corporations and banks and other long-term investors (McCauley and Zimmer, 1994; Fukao, 1995), consequently Japanese firms have higher investment rates than their U.S. counterparts (Prowse, 1990). Another perceived strength in Japanese governance is the long-term relationships between the multiple constituencies in the corporation, which make greater involvement by employees and suppliers possible (Womack, Jones and Roos, 1991); the benefits of these long-term relations are contrasted with the costs of potential “breaches of trust” following “undesirable” takeovers in the U.S. (Shleifer and Summers, 1988). 15 Moreover, Narayanan (1985), Shleifer and Vishny (1989), Porter (1992a, b) and Stein (1988, 1989), among others, asserted that the U.S. managers tended to be obsessed with quarterly performance measures, had excessively “short-termist” perspective and paid too much attention to potential takeover threats. In Porter (1992a, b) the U.S. practices are compared to those in Germany and Japan, where the long-term involvement of investors, especially banks, allow managers to invest for the long run and at the same time monitor their performance. Japanese Keiretsu, a corporate network in which a main bank serves as trade mediator, payment guarantor and information provider to client firms, are also highlighted for their superior ability to resolve financial distress or achieve corporate diversification (Aoki, 1990; Hoshi, Kashyap and Scharfstein, 1990). On the other hand, the financial regulations introduced at the beginning of the last century in the U.S. excessively limit effective monitoring by financial institutions and other large investors (Black, 1990; Grundfest, 1990; Roe, 1990, 1991, 1994). After the meltdown of the Japanese stock market in 1990, the U.S. gained a lower cost of equity, resulting from, as proposed by some commentators, superior minority shareholder protection (e.g., La Porta, Lopez-de-Silanes and Shleifer, 1998), which is one of the reasons why foreign firms increasingly choose to issue shares on the U.S. exchanges (Coffee, 2002). It is also discovered that the low cost of capital in Japan in the 1980s is a sign of excesses leading to overinvestment (Kang and Stulz, 2000). The East Asian crisis is attributed to poor investor protection in relevant countries (Johnson, 2000; Claessens, Djankov, Fan and Lang, 2002; Shinn and Gourevitch, 2002). The ideas that “undesirable” takeovers bring about “breaches of trust” and "shorttermist" behaviour gradually lose their popularity; instead takeovers are viewed as an effective way to break up inefficient conglomerates (Shleifer and Vishny, 1997b). The regulatory constraints in the U.S. that limit intervention by institutional investors offer valuable protection to minority shareholders against expropriation or self-dealing by large shareholders (Bebchuk, 1999, 2000; La Porta, Lopez-de-Silanes, Shleifer and Vishny, 2000). As shown in the review of Holmstrom and Kaplan (2001), there is a vast research on the takeover market in the U.S; it is widely agreed that takeover is the ultimate control mechanism of the Anglo-Saxon market-based model. Nevertheless, as pointed out by 16 Comment and Schwert (1995) and Bebchuk, Coates and Subramanian (2002), the market for corporate control in the U.S. already disappeared, after the introduction of anti-takeover laws and charter amendments at the end of the 1980s; most American firms are extremely well protected against hostile takeovers (Danielson and Karpoff, 1998). The U.K may remain the only country in the Organization for Economic Cooperation and Development (OECD) with an active and open market for corporate control (Short and Keasey, 1999; Becht et al, 2005). Some authors addressed this contradiction. In the opinion of Hansmann and Kraakman (2001), poison pill amendments and other anti-takeover devices are actually an improvement because they eliminate partial bids “of a coercive character.” La Porta, Lopez-de-Silanes and Shleifer (1999) asserted that the market for corporate control in the U.S. was more active than elsewhere, because the U.S. anti-takeover rules were less effective than anti-takeover measures elsewhere. In Holmstrom and Kaplan (2001), it is concluded that hostile takeovers and leveraged buyouts are no longer needed, as the U.S. corporate governance has reinvented itself, and the rest of the world seems to be following the same path. Similarly, Romano (1993) and Coffee (1999) predicted a world-wide convergence of corporate governance practices to the U.S. model. By contrast, Easterbrook and Fischel (1991) and Easterbrook (1997) argued that no global standards of corporate governance would be needed because international differences in corporate governance were attributable more to differences in markets than to differences in law; market forces would automatically create the regulatory underpinnings national systems need. Jackson and Moerke (2005), after an analysis of legal and regulatory reforms, banking and financing, and employment in Germany and Japan, found no evidence to support the arguments for international convergence; the simultaneous continuity and change in corporate governance indicate a potential form of hybridization of national models or renegotiation of stakeholder coalitions in these countries, and there is a growing diversity of firm-level corporate governance practices within national systems. The system of corporate governance in Australia is often described as forming part of the Anglo-Saxon outsider model of ownership and control (Scott, 1997; Weimar and Paper, 1999; Bradley, Schipani, Sundaram and Walsh, 1999; Campbell, 2002). It appears that many large listed companies in this country have relatively dispersed 17 shareholdings, and Australia has almost all the institutions as presented in the U.K. and U.S., such as mature securities market and regulator, takeover panel, disclosure regime and corporate governance codes; there is also a general assumption that, as an Englishspeaking and common law country, Australia must somehow be like the U.K. and U.S. However, questions have recently been raised about the accuracy of this classification. Lamba and Stapledon (2001) examined a sample of 240 listed companies; the high share concentrations at all company size suggest that, at least in terms of ownership structure, Australian listed companies do not conform to the outsider governance model presented in the U.K. and U.S. In Dignam and Galanis (2004), the evidence on share ownership and shareholder voting patterns, institutional investor activism, private rent extraction, market for corporate control and blocks to information flow indicates that Australia does not have an outsider system of corporate governance; rather, there is a system that has more in common with the insider model. According to the authors, Australia is in the process of reforming its corporate governance based on an assumption that it is an outsider model. If that assumption is incorrect, the reforms may have a destabilizing effect; therefore recognizing that the Australian market may have more in common with the insider system would enable a more appropriate response to corporate governance problems (Dignam and Galanis, 2004). 2.3 Corporate Governance Standards Gregory (1999, p.3) noted that, “[i]n the Anglo-Saxon nations – Australia, Canada, the U.K., and the U.S. – maximizing the value of the owners’ investment is considered the primary corporate objective. This objective is reflected in corporate guidelines and codes that emphasize the duty of the board to represent shareholders’ interests and maximize shareholder value.” Thus it appears that the current trend of recognizing that boards have responsibilities separate from management, and describing the practices that best enable directors to carry out these responsibilities, is a manifestation of the Anglo-Saxon shareholder model, in which the main concern of corporate governance is the conflict between strong managers and weak dispersed shareholders, based on agency theory introduced by Jensen and Meckling (1976). 18 This section presents a review on the Cadbury Report (1992) and Combined Code (1998, 2003) endorsed by the London Stock Exchange (LSE), corporate governance listing standards issued by the New York Stock Exchange (NYSE) and Nasdaq Stock Market (Nasdaq) in 2003, and the Guidelines (2003) published by the Australian Stock Exchange (ASX), focusing on the rules with respect of the composition and structure of the board of directors in these documents. In this thesis, “the phrase ‘board composition’ means the make-up of the board in terms of executive and non-executive directors, independent and affiliated non-executive directors, and male and female directors. The phrase ‘board structure’ refers to the structural features of the board, such as the presence or absence of committees (e.g., audit and remuneration committees), and whether the roles of chairperson and chief executive officer (CEO) are performed by one or two persons” (Stapledon and Lawrence, 1996, p.1). 2.3.1 Recommendations in the U.K. The global movement to promote board independence in public companies has rapidly gained momentum in the past fifteen years; Panasian, Prevost and Bhabra (2003) observed that much of this trend was influenced by the publication of the Cadbury Report (1992) in the U.K. The Cadbury Report (1992), sometimes referred to as the Magna Carta of Corporate Governance (Gregory, 2001a), was produced by the Committee on the Financial Aspects of Corporate Governance chaired by Sir Adrian Cadbury. It consists of a formal code and extensive comments and recommendations for publicly held U.K. firms, around the separation of the role of CEO and chairman, balanced composition of the board, selection process for non-executive directors (NEDs), transparency of financial reporting and the need for good internal controls. The report comments that the board must retain full and effective control over the company and monitor the executive management. “[T]he effectiveness with which boards discharge their responsibilities determines Britain’s competitive position. They must be free to drive their companies forward, but exercise that freedom within a framework of effective accountability. This is the essence of any system of good corporate governance” (Cadbury Report, 1992, Report 1.1). 19 The report suggests that every public company should be headed by an effective board which can both lead and control the business; this means a board made up of a combination of executive directors, with their intimate knowledge of the business, and of outside NEDs, who can bring a broader view to the company’s activities. Therefore the board should include NEDs of sufficient calibre and number for their views to carry significant weight in the board’s decisions; their appointment should be a matter for the board as a whole, and there should be a formal selection process, which will enforce the independence of NEDs and make it evident that they have been appointed on merit and not through any form of patronage. As NEDs are supposed to bring an independent judgement to bear on issues of strategy, performance and resources, including key appointments and standards of conduct, it is recommended that the majority of non-executives “… should be independent of management and free from any business or other relationship which could materially interfere with the exercise of their independent judgement, apart from their fees and shareholding” (Cadbury Report, 1992, Code 2.2). It is for the board to decide in particular cases whether this definition of independence is met. Given the importance and particular nature of the chairman’s role, it should be in principle separate from that of the CEO. If the two roles are combined in one person, it will represent a considerable concentration of power; in such situations board members are advised to look to a senior NED, who might be the deputy chairman, as the person to whom they could address any concerns about the combined office of chairman/CEO and its consequences for the effectiveness of the board. It is also recommended that all listed companies establish an audit committee of at least three NEDs with written terms of reference which deal clearly with its authority and duties, and a remuneration committee consisting wholly or mainly of NEDS and chaired by a NED, to recommend to the board the remuneration of the executive directors, drawing on outside advice as necessary. Following the momentum created by the Cadbury Committee, a number of organizations and stock exchanges instituted corporate governance reviews throughout the 1990s, for example, in Australia (Bosch report, 1995; AIMA Report, 1997), Belgium (Cardon Report, 1998), Brazil (IBGC Code, 1999), Canada (Dey Report, 1994), France (Vienot Report I, 1995; Vienot Report II, 1999), Hong Kong (HKSE 20 Guide, 1995), India (Confederation Code, 1998), Japan (Corporate Governance Principles, 1998), Malaysia (Report on Corporate Governance, 1999), Mexico (Code of Corporate Governance, 1999), South Africa (King Report, 1994), South Korea (Code of Best Practice, 1999), Thailand (SET Code, 1997), the Netherlands (Peters Code, 1997), the U.S. (NACD Report, 1996; BRT Report, 1997), Spain (Governance of Spanish Companies, 1998), Sweden (Swedish Academy Report, 1994) and OECD (Millstein Report, 1998; OECD Principles, 1999). Gregory (2001a, b, c) provided detailed reviews on these reports and codes. In the U.K., the Cadbury Report (1992) was followed by the Greenbury Report (1995), Hampel Report (1998) and Combined Code (1998). The Combined Code: Principles of Good Governance and Code of Best Practices (Combined Code), issued by the LSE Committee on Corporate Governance, has been appended to the LSE Listing Rules, covering areas relating to composition, structure and operation of the board, director’s remuneration, accountability and audit, relations with and responsibilities of institutional shareholders. Building on the Cadbury, Greenbury and Hampel reports, the Combine Code (1998) recommends that the board include a balance of executive and NEDs such that no individual or small group of individuals can dominate the board’s decision-making. NEDs should comprise not less than one-third of the board, and the majority of them “… should be independent of management and free from any business or other relationship which could materially interfere with the exercise of their independent judgement” (Combined Code, 1998, Provision A.3.2). NEDs considered by the board to be independent should be identified in the annual report. According to the Code, there are two key tasks at the top of every public company – the running of the board and the executive responsibility for the running of the company’s business; there should be a clear division of responsibilities which ensures a balance of power and authority, such that no one has unfettered powers of decision. Thus a decision to combine the posts of chairman and CEO in one person should be publicly justified. Whether the posts are held by different people or by the same person, there should be a strong and independent non-executive element on the board, with a recognized senior member other than the chairman to whom concerns could be 21 conveyed. The chairman, CEO and senior independent director should be identified in the annual report. The Code also suggests that, unless the board is small, a nomination committee should be established to make recommendations to the board on all new board appointment; a majority of the members of this committee should be NEDs and the chairman may be either the chairman of the board or a NED. The board should also establish an audit committee of at least three directors, all non-executives and a majority of them should be independent, with written terms of reference which deal clearly with its authority and duties. If there is a remuneration committee on the board, the committee should consist exclusively of independent NEDs. The LSE requires that each listed company disclose in its annual report how it has applied the Code principles and whether it has complied with the Code provisions, if not, why not and for what period. Guidance for companies on how this could be approached was needed; this led to the publication of the Turnnbull Report (1999) by the Institute of Chartered Accountants in England and Wales (ICAEW), to define more clearly the accountability of directors and management. In July 2002, Derek Higgs was appointed by the Secretary of State for Trade and Industry and the Chancellor to lead an independent review into the role and effectiveness of NEDs, resulting in the publication of the Higgs Report (2003). Subsequently, the Combined Code (1998) was revised based on the recommendations made in this report. The Combined Code (2003) continues to require the board to identify in its annual report each NED who it considers to be independent; however, the Higgs test for independence has been introduced. “The board should determine whether the director is independent in character and judgment and whether there are relationships or circumstances which are likely to affect, or could appear to affect, the director’s judgement. The board should state its reasons if it determines that a director is independent notwithstanding the existence of relationships or circumstances which may appear relevant to its determination” (Combined Code, 2003, Provision A.3.1); factors relevant to independence include if the director: has been an employee of the company or group within the last five years; 22 has had a material business relationship with the company within the last three years; receives additional remuneration from the company apart from a director’s fee, participates in the company’s share option or a performance-related pay scheme, or is a member of the company’s pension scheme; has close family ties with any of the company’s advisers, directors or senior employees; holds cross-directorships or has significant links with other directors through involvement in other companies or bodies; represents a significant shareholder; and has served on the board for more than nine years from the date of first election. NEDs may serve beyond six years, i.e., two terms, but longer periods of service should be subject to rigorous review. This is a dilution of the Higgs recommendation that NEDs generally be limited to serving for six years and the reasons for any longer period of service explained to shareholders. The revised Code takes the original position concerning board composition a step further by suggesting that at least half of the members of the board, excluding the chairman, should be independent NEDs. The chairman must meet the test of independence on appointment but thereafter is not regarded as independent. According to Higgs (2003), the chairman will be in constant and close contact with the executive directors in carrying out his or her duties; as a result the test for independence is considered to be “neither appropriate nor necessary” once a chairman has been appointed. It should be noted that, in this recommendation, a non-independent NED is regarded in the same category as an executive director, rather than neutral, like the chairman. Higgs (2003) argued that his recommendation that at least half the board comprise independent NEDs should not be interpreted as meaning that non-independent NEDs had no place on company boards. However, for many companies, the presence of non-independent NEDs could make compliance with this recommendation significantly more difficult. The Code states that a chief executive should not go on to become chairman of the same company except in exceptional cases where major shareholders are consulted in 23 advance and the reasons for that appointment set out in the annual report; this is a relaxation of the Higgs recommendation that a chief executive should not be permitted to become chairman of the company. The revised Code provides that every company should establish an audit committee consisting of at least three, or in the case of smaller companies two, independent NEDs, and at least one member should have recent and relevant financial experience. The remuneration committee must consist of at least three, or in the case of smaller companies two, independent NEDs. Some of the more controversial Higgs recommendations were dropped in the Combined Code (2003). For example, Higgs (2003) suggested that the senior independent NED, rather than the chairman, should be responsible for communicating shareholder views to the board. The Code confirms that the chairman should communicate shareholder views to the board and discuss governance and strategy with major shareholders. The senior independent NED is expected to attend sufficient meetings with major shareholders to develop an understanding of their concerns; he or she must also be available to meet with shareholders if they have concerns which have not been resolved by, or are inappropriate to discuss with, the chairman, chief executive or financial director. 2.3.2 Listing Rules in the U.S. In the U.S., in direct response to corporate collapses resulting from accounting irregularities and failures of ethics and controls, i.e., Enron, WorldCom and Arthur Andersen, the Sarbanes-Oxley Act was passed into law by the Congress in July, 2002. The Act sought to enhance corporate governance and disclosure requirements by enforcing a higher level of responsibility, accountability and financial reporting transparency on company executives, directors and auditors. Subsequently, the NYSE and Nasdaq initiated corporate governance reforms and proposed changes to their listing requirements. According to the NYSE (NYSE Press Release, August 16, 2002), the reforms “…focus on giving boards greater independence and investors greater say in the governance of their companies. The tighter corporategovernance standards aim to help win back the trust and confidence of investors”. In November 2003, the Securities and Exchange Commission (SEC) approved the corporate governance listing standards of the NYSE. The rules, which are codified in 24 Section 303A of the NYSE Listed Company Manual, are the culmination of a series of proposals and amendments from the NYSE that had occurred over more than one year. The NYSE rules require that a majority of each listed company’s directors qualify as independent directors. “No director qualifies as ‘independent’ unless the board of directors affirmatively determines that the director has no material relationship with the listed company (either directly or as a partner, shareholder or officer of an organization that has a relationship with the company). Companies must identify which directors are independent and disclose the basis for that determination” (NYSE Manual, 2003, Section 303A.02). In addition to the absence of a material relationship, the NYSE prohibits a finding that a director is independent in the following situation: a director who is an employee, or whose immediate family member is an executive officer, of the company is not independent until three years after the end of such employment relationship; a director who receives, or whose immediate family member receives, more than $100,000 per year in direct compensation from the listed company, other than director and committee fees and pension or other forms of deferred compensation for prior service (provided such compensation is not contingent in any way on continued service), is not independent until three years after ceasing to receive more than $100,000 per year in such compensation; a director who is affiliated with or employed by, or whose immediate family member is affiliated with or employed in a professional capacity by, a present or former internal or external auditor of the company is not independent until three years after the end of the affiliation or the employment or auditing relationship; a director who is employed, or whose immediate family member is employed, as an executive officer of another company where any of the listed company’s present executives serve on that company’s compensation committee is not independent until three years after the end of such services or the employment relationship; or a director who is an executive officer or an employee, or whose immediate family member is an executive officer, of a company that makes payments to, or receives payments from, the listed company for property or services in an amount which, in any single financial year, exceeds the greater of $1 million, or 25 2% of such other company’s consolidated gross revenues, is not independent until three years after falling below such threshold. “Immediate family member” is defined to include a person’s spouse, parents, children, siblings, mothers and fathers-in-law, sons and daughters-in-law, brothers and sisters-inlaw, and anyone, other than domestic employees, who shares such person’s home. The NYSE stated in commentary to the rules that the board should consider the issue not merely from the standpoint of the director, but also from the standpoint of persons or organizations with whom the director had an affiliation. Because the concern is independent from management, the rules do not view ownership of even a significant amount of shares, by itself, as a bar to an independence finding. To facilitate the determination of whether a director has a material relationship with the company, the board may adopt categorical standards, which must be disclosed, and may make a general disclosure if a director meets these standards. Any determination of independence for a director who does not meet the standards must be specifically explained; in the event that a director with a relationship that does not fit within the standards is determined to be independent, a board must disclose the basis for its determination. Under the NYSE rules, the NEDs of a listed company must convene regularly scheduled sessions without members of management in attendance; if the NEDS include persons who are not independent directors as defined under the rules, at least one separate session of independent directors should be convened annually. If a director is chosen to preside at these meetings, that director’s name must be disclosed and, in order that interested parties may be able to make their concerns known to the NEDs, the company must disclose a method for such parties to communicate directly with the presiding director or with the NEDs as a group. Every listed company is required to have a nominating/corporate governance committee and a compensation committee, each of which is comprised entirely of independent directors and has a charter outlining certain minimum duties and responsibilities. Each company must have an internal audit function and an audit committee with a charter addressing the committee’s purpose and certain minimum duties and responsibilities; the audit committee must have a minimum of three members, each of whom qualifies as 26 an independent director under the rules, as well as the independence criteria for audit committee members set forth in the Securities Exchange Act Rule 10A-3(b)(1). According to Rule 10A-3(b)(1), in order to be considered to be independent, a member of an audit committee may not be an affiliated person of the company or any subsidiary of the company, and may not accept any consulting, advisory or other compensatory fee from the company or any subsidiaries thereof, other than in his or her capacity as a member of the board or any board committee. Compensatory fee does not include the receipt of fixed amounts of compensation under a retirement plan for prior service with the company, provided that such compensation is not contingent in any way on continued service. The NYSE asked each company to develop and disclose its corporate governance guidelines and a code of business conduct and ethics for its directors, officers and employee. The CEO must certify on an annual basis that he or she is not aware of any violation by the company of the NYSE corporate governance listing standards, and must promptly notify the NYSE after becoming aware of any material non-compliance with the governance requirements. The NYSE may issue a public reprimand letter to a company that violates a listing standard. The NYSE commented that suspending trading or delisting a company could be harmful to the very shareholders the standards sought to protect; therefore these measures would be used sparingly and judiciously. For companies that repeatedly or flagrantly violate the rules, suspension and delisting remain the ultimate penalties. In November 2003, the SEC also issued an order approving the corporate governance rules of Nasdaq. Perhaps the most important provision of the rules, which are codified as Nasdaq Marketplace Rules 4200 and 4350, is the requirement that a majority of the board of directors of a listed company be independent directors. As defined by Nasdaq Marketplace Rule 4200(a)(15), “[i]ndependent directors means a person other than an officer or employee of the company or its subsidiaries or any other individual having a relationship which, in the opinion of the company’s board of directors, would interfere with the exercise of independent judgement in carrying out the responsibilities of a director. The following persons shall not be considered independent”: 27 a director who is, or at any time during the past three years was, employed by the company or by any parent or subsidiary of the company; a director who accepted or who has a family member who accepted any payments from the company or any parent or subsidiary of the company in excess of $60,000 during any period of twelve consecutive months within the past three years, other than some exceptions listed in the rule; a director who is a family member of an individual who is, or during the past three years was, employed by the company or by any parent or subsidiary of the company as an executive officer; a director who is, or has a family member who is, a parent in, or a controlling shareholder or an executive officer of, any organization to which the company made, or from which the company received, payments, other than those arising solely from investments in the company’s securities or payments under nondiscretionary charitable contribution matching programs, for property or services in the current or any of the past three financial years that exceed 5% of the recipient’s consolidated gross revenues for that year, or $200,000, whichever is more; a director who is, or has a family member who is, employed as an executive officer of another entity where at any time during the past three years any of the executive officers of the listed company serve on the compensation committee of such other entity; or a director who is, or has a family member who is, a current partner of the company’s outside auditor, or was a partner or employee of the company’s outside auditor who worked on the company’s audit at any time during any of the past three years. Nasdaq specified that share ownership alone would not preclude a director from being deemed independent. The term “family member” as used in the test of independence includes a person’s spouse, parents, children and siblings whether by blood, marriage or adoption, or anyone residing in such person’s home. Independent directors must have regularly scheduled meetings, at which only independent directors are present. These sessions, which Nasdaq contemplated would 28 occur at least twice a year in conjunction with regularly scheduled board meetings, are designed to encourage and enhance communication among independent directors. The rules require that the compensation of the CEO and other executive officers of the company must be determined, or recommended to the board for determination, either by a majority of the independent directors, or a compensation committee comprised solely of independent directors. Director nominees must either be selected, or recommended for the board’s selection, either by a majority of the independent directors, or a nominating committee comprised solely of independent directors. Thus Nasdaq, unlike the NYSE, does not mandate a listed company to have a compensation committee and a nominating committee. Each company must establish an audit committee of at least three members, who qualify as independent directors under the rules, and satisfy the independence criteria for audit committee members set forth in the Securities Exchange Act Rule 10A-3(b)(1). In addition, they should not have participated in the preparation of financial statements of the company or any current subsidiary of the company during the past three years, and should be able to read and understand financial statements at the time of appointment to the committee. At least one member must have past employment experience in finance or accounting, requisite professional certification in accounting or any other comparable experience or background which results in the individual’s financial sophistication. It is required that all related party transactions be approved by the audit committee or another independent body of the board of directors. Companies are also required to adopt a code of conduct for their directors, officers and employees, which should be publicly available. Each company must provide Nasdaq with prompt notification after an executive officer becomes aware of any material noncompliance by the company with any of the Nasdaq’s listing requirements. 2.3.3 Australian Guidelines Australia had watched closely as big corporate failures had prompted the U.S. regulatory authorities to launch significant corporate governance reforms, including the Sarbanes-Oxley Act of 2002 and the then proposed amendments to the NYSE and Nasdaq listing standards. In order to promote and restore investor confidence, the ASX convened the ASX Corporate Governance Council in August 2002; its purpose, as 29 announced by the ASX, was to develop recommendations which reflect international best practice. In March 2003, the Council released Principles of Good Corporate Governance and Best Practice Recommendations (Guidelines), which, as indicated by the Council, follows the “if not, why not” approach of the U.K. Combined Code (1998, 2003). It is concluded that, “[t]he best practice recommendations are not prescriptions. They are guidelines, designed to produce an efficient, quality or integrity outcome. This document does not require a ‘one size fits all’ approach to corporate governance. Instead, it states aspirations of best practice for optimising corporate performance and accountability in the interests of shareholders and the broader economy. If a company considers that a recommendation is inappropriate to its particular circumstances, it has the flexibility not to adopt it – a flexibility tempered by the requirement to explain why” (Guidelines, 2003, p.5). In the Guidelines (2003), companies are recommended to formalise and disclose the function reserved to the boards and those delegated to management. All directors should bring an independent judgement to bear in decision-making, and a majority of the board should be independent directors. According to the Council (Guidelines, 2003, p.19), “[a]n independent director is independent of management and free of any business or other relationship that could materially interfere with – or could reasonably be perceived to materially interfere with – the exercise of their unfettered and independent judgement.” It is further defined in Box 2.1 of the Guidelines (2003) that an independent director is a NED and is not a substantial shareholder of the company or an officer of, or otherwise associated directly with, a substantial shareholder of the company; within the last three years has not been employed in an executive capacity by the company or another group member, or been a director after ceasing to hold any such employment; within the last three years has not been a principal of a material professional adviser or a material consultant to the company or another group member, or an employee materially associated with the service provided; 30 is not a material supplier or customer of the company or other group member, or an officer of otherwise associated directly or indirectly with a material supplier or customer; has no material contractual relationship with the company or another group member other than as a director of the company; has not served on the board for a period which could, or could reasonably be perceived to, materially interfere with the director’s ability to act in the best interests of the company; and is free from any interest and any business or other relationship which could, or could reasonably be perceived to, materially interfere with the director’s ability to act in the best interests of the company. The board should state its reasons if it considers a director to be independent notwithstanding the existence of relationships listed above. In this context, it is important for the board to consider materiality thresholds from the perspective of both the company and its directors, and to disclose these. Directors considered by the board to be independent should be identified as such in the annual report. The tenure of each director, which is important to an assessment of independence, should also be disclosed. The board should regularly assess the independence of each director in light of interests disclosed by them; where the independence status of a director is lost, this should be immediately disclosed to the market. The Guidelines (2003) advise that the role of chairman and CEO should not be exercised by the same individual, and the chairman should be an independent director. Where the chairman is not an independent director, it may be beneficial to consider the appointment of a lead independent director. The board is encouraged to establish a nomination committee, a remuneration committee and an audit committee; all the members of the audit committee should be NEDs. Each committee should consist of at least three members, the majority being independent directors, with a formal charter setting out the committee’s role and responsibilities, composition, structure, and membership requirements. The nomination committee should be chaired by the chairman of the board or an independent director; 31 the remuneration committee and the audit committee should be chaired by an independent director. The board or appropriate committee should establish and disclose policies on risk oversight and management, and trading in company securities by directors, officers and employees. The process for performance evaluation of the board, its committees, directors and key executives, should be disclosed; companies may provide disclosure in relation to their remuneration policies to enable investors to understand the costs and benefits of those policies, and the link between remuneration paid to directors and key executives and corporate performance. Companies are also suggested to clearly distinguish the structure of NEDs remuneration from that of executives, and to ensure that payments of equity–based executive remuneration are made in accordance with thresholds set in plans approved by shareholders. Company are advised to adopt a code of conduct to guide their directors, CEOs, chief financial officers and any other key executives as to the practices necessary to maintain confidence in the companies’ integrity, and the responsibility and accountability of individuals for reporting and investigating unethical practices. A code of conduct to guide compliance with legal and other obligations to legitimate stakeholders should also be developed and disclosed. The Guidelines (2003) recommend that written policies and procedures be developed to ensure compliance with the ASX disclosure requirements and to ensure accountability at a senior management level for that compliance. The strategy to promote effective communication with shareholders and encourage effective participation at general meetings should be designed and disclosed; the company should request the external auditor to attend the annual general meeting and answer shareholder questions about the conduct of the audit and the preparation and content of the auditor’s report. Under ASX Listing Rule 4.10.3, from 2004 each company is required to provide a statement in its annual report disclosing the extent to which it has followed the best practice recommendations in the reporting period; where the company has not followed all the recommendations, it must identify the recommendations that have not been followed and give reasons for not following them. 32 In August 2007, the Corporate Governance Council released the second edition of the Guidelines (2007), which contains no significant changes to the recommendations to “[s]tructure the board to add value” as introduced above (Guidelines, 2003, p.19; Guidelines, 2007, p.16), except that the last two points in Box 2.1 have been removed from the definition of independence. “Best practice” has also been removed from the title and text of the document to eliminate any perception that the principles are prescriptive and so not to discourage companies from adopting alternative practices and “if not, why not” reporting where appropriate. 2.4 Summary Recent years have witnessed an explosion of research on corporate governance around the world, and there are, in general, two major governance systems that have been identified in the past two decades. The first is the outsider system adopted in the U.K. and U.S., and the second is the insider system in Germany and Japan. It is noted that researchers tended to favour the insider system during the 1980s, when the economies of Germany and Japan outperformed others, and tended to favour the outsider system in the 1990s, when the economies of the U.S. looked better. Australia’s system of corporate governance has been described as forming part of the Anglo-Saxon outsider model of ownership and control. However, some researchers raised questions about this classification, and argued that the Australian system might have more in common with the insider system. The calls for greater board independence become increasingly popular after the publication of the Cadbury Report (1992); it appears that this movement is based on the Anglo-Saxon shareholder model, which assumes that the basic conflict is between managers and shareholders. From 2003, listed companies in the U.K., U.S. and Australia have been subject to new governance code, rules or guidelines, which sought to improve board independence in public companies. There are some differences in the general approaches of governance endorsed by the stock exchanges. In the U.S., the NYSE and Nasdaq take a mandatory approach in which every company must comply with every standard, in order to be listed on the stock exchanges. On the other hand, the ASX and LSE follow a voluntary approach in which “[i]f a company considers that a recommendation is inappropriate to its particular 33 circumstances, it has the flexibility not to adopt it – a flexibility tempered by the requirement to explain why” (Guidelines, 2003, p.5). It is agreed that a majority of each listed company’s directors, or in the case of the LSE at least half of the members, should qualify as independent directors. Although the definitions of “independence” vary, it appears that all of them are based on the statement in the Cadbury Report (1992, Code 2.2) - an independent director “… should be independent of management and free from any business or other relationship which could materially interfere with the exercise of their independent judgement, apart from their fees and shareholding”. Although it is for the board to decide in particular cases whether the definition of independence is met, there are lists of the persons who should not be considered independent. For a company listed on the ASX and LSE, the board may state its reasons if it determines that a director is independent notwithstanding the existence of relationships or circumstances included in the lists. In contrast, the lists presented by the NYSE and Nasdaq are coercive, and more prescriptive. According to the ASX and LSE, the roles of chairman and CEO should not be exercised by the same individual, and the chairman should be an independent director or meet the independence test on appointment; if the two roles are combined in one person or the chairman is not independent, companies are suggested to appoint a lead or senior independent director. The NYSE and Nasdaq do not have any opinion on the concern over CEO duality. All the stock exchanges support listed companies to establish an audit committee. For firms listed on the ASX and LSE, the committee should be comprised entirely of NEDs, the majority being independent directors; in addition, the ASX recommends that the committee be chaired by an independent director. The NYSE and Nasdaq require that all the members of the committee qualify as independent directors under their rules, as well as the independence criteria for audit committee members set forth in the Securities Exchange Act. The ASX, LSE and NYSE believe it would be necessary for each company to have a nomination committee and a remuneration committee. The ASX recommends that a majority of each committee’s members be independent directors; the nomination 34 committee should be chaired by the chairman of the board or an independent director, and the remuneration committee should be chaired by an independent director. The LSE suggests that a majority of the members of the nomination committee should be NEDs, and the committee should be chaired by the chairman of the board or a NED; the remuneration committee should consist exclusively of independent directors. The NYSE and Nasdaq require that both committees be comprised entirely of independent directors. From the above review, it could be concluded that, in the long-run, these recommendations and listing rules are likely to result in a shift in the overall power of boards of directors in favour of independent directors, and away from executive management, among public companies in the U.K., U.S. and Australia. 35 Chapter 3. Empirical Evidence 3.1 Introduction The empirical research on board of directors, matching the trend towards greater board independence, is growing. In general, researchers have taken two approaches to study the impact of independent or outside directors on firm performance (Bathala and Rao, 1995; Lawrence and Stapledon, 1999; Bhagat and Black, 1999, 2000; Panasian et al, 2003). The first approach is based on relating board composition and structure to certain corporate events, such as executive turnover (e.g., Weisbach, 1988; Borokhovich, Parrino and Trapani, 1996; Mikkelson and Partch, 1997; Suchard, Singh and Barr, 2001) and remuneration (e.g., Newman and Wright, 1995; Sridharan, 1996; Brick, Palmon and Wald, 2006), financial reporting (e.g., Wright, 1996; Beasley, 1996; Dechow, Sloan and Sweeney, 1996; Peasnell, Pope and Young, 1998; Beasley and Petroni, 1998), making, or defending against, a takeover bid (e.g., Byrd and Hickman, 1992; Brickley, Coles and Terry, 1994), management buyout (e.g., Chun, Rosenstein, Rangan and Davidson, 1992) and shareholder litigation (e.g., Romano, 1991; Ferris, Lawless and Makhija, 2001; Ferris, Jagannathan and Pritchard, 2003; Helland and Sykuta, 2005). Some scholars, for example Lawrence and Stapledon (1999) and Bhagat and Black (1999) have reviewed this research; although these studies offer some insights into how different boards behave on particular jobs, “[t]he principle weakness of this approach is that it cannot tell us how board composition affects overall firm performance. Firms with majority-independent boards could perform better on particular tasks, such as replacing the CEO, yet worse on other tasks, leading to no net advantage in overall performance” (Bhagat and Black, 1999, p.3). The second approach involves investigating directly the correlation between board characteristics and financial performance, i.e., the “bottom line” of firm performance. As indicated by Bathala and Rao (1995), Lawrence and Stapledon (1999), Bhagat and Black (1999, 2000), and Panasian et al (2003), it may avoid the weakness inherent in the first group of studies. 36 In this chapter an up-to-date review of this stream of literature in Australia and overseas, which gives evidence on whether board characteristics and firm performance are related, is produced; its focus is on cross-sectional studies on publicly listed companies. The remainder of this chapter is organized as follows. Section 2 provides a survey of Australian research in this field. In Section 3 some U.S. studies on the board demographics-firm performance link, and three event studies on the stock-market reactions to changes in board composition, are examined. Section 4 provides an introduction to empirical evidence from other regions. A summary of the findings and identified limitations in these studies is then presented in Section 5. 3.2 Australian Evidence Kiel and Nicholson (2003) conducted a brief survey of Australian studies on board of directors, and noted that there were only three papers addressing the relationship between board composition and firm performance. They concluded that “[t]he Australian literature on corporate governance has been primarily descriptive, with an emphasis on describing the size and composition of boards and the extent to which board interlocks occur” (Kiel and Nicholson, 2003, p.191). In Muth and Donaldson (1998), the authors investigated the validity of agency theory and stewardship theory, which make different predictions about the impact of board independence on firm performance. By examining a sample of 145 large ASX-listed companies, it was found that a higher board independence factor, which is composed of CEO duality, board size, proportion of NEDs, interest alignment with owners and average age across all directors, leads to both lower subsequent shareholder wealth (dividends and share price appreciation) and sales growth; the level of board independence, however, appears to have no significant effect on profit performance factor made up of return on assets (ROA), return on equity (ROE) and profit margin. Muth and Donaldson (1998) concluded that agency theory predictions relating to board independence and firm performance were not upheld while those of stewardship theory were supported. Calleja (1999) investigated 83 of the top 100 Australian companies ranked by market capitalization, listed on the ASX at 31 December 1997; her performance measure is the 37 one year adjusted shareholder return for each company at 31 December 1997. Regressions are carried out for the number of board committees, board size and the proportion of NEDs, on shareholder return; no statistically significant relationship is found. Similarly, Lawrence and Stapledon (1999) attempted to determine whether there was an association between board composition and corporate performance in the top 100 companies listed on the ASX at the end of 1995. They discovered that the proportion of independent directors and the proportion of executive directors were insignificant explanatory variables for share price performance, i.e., the mean percentage change in share price, during the 1985-1995 and 1990-1995 periods. For accounting performance measure, during the 1987-1991 and 1991-1995 sample periods, the proportion of independent directors is significantly negatively related to the ratio of revenue to assets. However, there is no relation between board composition and other accounting measures such as net profit, earnings before interest and tax (EBIT), ratio of revenue to assets, and net profit to revenue. Lawrence and Stapledon (1999) suggested that, as far as Australian largest listed companies were concerned, independent directors did not appear to have added value over the 1985 to 1995 period. In Cotter and Silverster (2003), the authors focused on the independence of board of directors and its audit and compensation committees; they assumed that “[f]or firms whose boards use a committee structure, much of the monitoring responsibility of the board is expected to rest with the independent committee members” (Cotter and Silverster, 2003, p.211). Moreover, they argued that the more direct indicator of effective monitoring of management was firm value rather than performance; “[w]hile performance and value are linked, firm value more directly captures predictions emanating from agency theory” (Cotter and Silverster, 2003, p.216). Market value of equity is used as the measure for firm value. It appears that the year of interest in their study is 1997. Their analysis of 109 large Australian companies show a negative relationship between leverage and audit committee independence, regardless of whether this is measured as the proportion of independent directors or the absence of the CEO from this committee, indicating that when debt-holders have lower incentives to monitor the financial 38 reporting and audit functions, the independence of committee level monitoring becomes more important. For the full board, managerial ownership and substantial shareholders are negatively associated with independence; that is, a more independent board is used for monitoring when there is low managerial ownership and an absence of substantial shareholders. However, neither board nor committee independence is significantly associated with firm value. Kiel and Nicholson (2003) is the first large-scale investigation of the Australian experience concerning board characteristics and corporate performance; their sample includes 348 companies out of the top 500 companies trading on the ASX in 1996. It is reported that larger companies have larger, more interlocked boards, a greater proportion of outside directors and more likely to separate the roles of chairman and CEO. Turning to correlation of board demographics and firm performance, different results occur depending on whether the market-based measure (Tobin’s q) or the accounting– based measure (ROA) of firm performance is used. It appears that the market rewards large boards and also boards with a relatively lower proportion of outside directors, although no such relationship is found with respect to the accounting-based performance measure. In Bonn, Yoshikawa and Phan (2004), the authors addressed the concern over the influence of national differences in corporate governance on firm performance. Their Australian sample consists of 104 manufacturing firms from the top 500 companies listed on the ASX, and the Japanese sample consists of 160 manufacturing firms from the Nikkei 300 Index. With the Australian firms, they found a positive association between outside director ratio and ROA, and between the ratio of female directors and market-to-book ratio (MBT). For Japanese firms, board size and the average age of directors are both negatively associated with MBT. Bonn et al (2004) asserted that their findings confirmed the importance of national context on corporate governance practices. It is recommended that when applying existing theory in different countries, or developing new theory, researchers need to take the national context into consideration; there may be different agency relationships in 39 different countries, and it is necessary to be cautious when interpreting and generalizing the results across national boundaries. Balatbat, Taylor and Walter (2004) examined ownership structure and board attributes of 313 Australian initial public offerings (IPOs) between 1976 and 1993, and their relations with up to 5 years of post-listing operating performance. From their sample of 1,316 firm-year observations, the authors identified a positive association between managerial ownership and firm performance in terms of operating return (EBIT deflated by total assets) in the fourth and fifth years after the IPO; there is some evidence of a positive relationship between institutional ownership and performance, although this is not consistent across post-listing years. There is no evidence that board composition is associated with variation in performance, although firms with dual leadership are found to perform better than those with a unitary leadership structure. Balatbat et al (2004, p.327) commented that “[t]he circumstances faced by IPO firms may frequently render more traditional corporate governance practices irrelevant, especially as they relate to monitoring managers who provide highly firm-specific human capital as a key component of the firm’s value.” Hutchinson and Gul (2004) explored whether some corporate governance variables, including the ratio of non-executive to executive directors on the board, would moderate the relationship between investment opportunities and firm performance. The sample they used includes 310 companies listed on the ASX in 1998-1999. Their regression indicates that NEDs present a positive effect on ROE; this effect, however, disappears in the sensitivity test. 3.3 Evidence from the U.S. Bhagat and Black (2000) searched the literature on whether board demographics affect firm performance or vice versa. Based on their review of twelve U.S. studies, they concluded that “[p]rior research does not establish a clear correlation between board independence and firm performance” (Bhagat and Black, 2000, p.5). The literature to date confirms that the empirical research around this topic has been most developed in the U.S; of the thirty-seven overseas papers for which complete texts 40 were obtained in 2008, twenty-four are from this country, including three event studies. In this section an overview of these U.S. papers is provided. 3.3.1 Cross-Sectional Studies In an early study, Pfeffer (1972) proposed that the percentage of inside or outside directors on the board might be an important indicator of the extent to which the organization was externally or internally orientated. If, either directors matter, or the extent of the organization’s inside-outside orientation matters, then some consequences should be evident for those organizations that do or do not match well with environmental requirements. Specifically, it was hypothesized that “organizations that deviate relatively more from a preferred inside-outside director orientation should be relatively less successful when compared to industry standards than those that deviate less from a preferred board composition” (Pfeffer, 1972, p.225). This hypothesis was tested using a random sample of 80 large U.S. corporations with relevant data in 1969. The author developed an equation to represent the pooled experience of the sample companies with respect of inside-outside director orientation proxied by the percentage of inside directors. According to Pfeffer (1972), the equation gives an optimal insider-outsider relationship for each company. The findings indicate that firms that deviated more from the optimal equation are likely to perform more poorly compared to industry standards, in terms of net income to sales ratio and net income to stockholders’ investment ratio. Baysinger and Bulter (1985, p.115), in order to “… examine performance differences across corporations as a function of differences in board independence and changes in independence occurring between 1970 and 1980”, constructed a sample of 266 major U.S. firms. They classified the directors of sample companies into three components. The executive component includes corporate officers and retirees, and other insiders; the instrumental component includes financiers, consultants, legal counsel and interdependent decision-makers. The monitoring component is made up of public director, professional directors, private investors and independent decision-makers; it is asserted that only these directors in the monitoring group “… meet the popular standards of independence” (Baysinger and Bulter, 1985, p.113). 41 The measure of financial performance selected in this paper, relative financial performance, is calculated by dividing the firm’s ROE by the average ROE for all the firms in its primary industry, including those not in the sample. Baysinger and Bulter (1985) reported that firms that had invited relatively more independent directors onto their boards in 1970 enjoyed relatively better records of financial performance in 1980; however, changes in the proportion of independent directors did not produce significant changes in performance over time. To offer some insights into the relationship between financial involvement of directors and firm performance, Kesner (1987) randomly chose 250 of the 1983 Fortune 500 companies. They found that insiders owned a far greater amount of stock than their outside counterparts, and the proportion of insiders was positively related to current firm performance (1983), in terms of profit margin and ROA, and future performance (1984-1985) measured by total return to investors. According to the writer, “[t]hese findings suggest that higher inside representation is associated with greater profitability and higher asset utilization” (Kesner, 1987, p.504). In addition, it appears that in low growth industries director ownership does not affect current or future performance, but the reverse is true for companies in high growth industries on most performance measures. Kesner (1987, p.505) concluded that “[a] firm in a rapid growth industry tends to perform better when directors have a high personal financial stake in the company”. Hermalin and Weisbach (1988), to investigate whether firm performance, CEO tenure and changes in market structure would lead to changes in board composition, assembled a database on the directors of 142 NYSE-traded companies between 1971 and 1983, obtaining 1,521 firm-year observations in their sample. Full-time employees of the firm are designated as insiders. Directors who are closely associated with the firm, but are not full-time employees, are designated as “greys”; “grey” directors are either related to an officer of the company or have extensive business dealings with the company, which make their independence from management questionable. The remaining board members are identified as outsiders. Their results suggest that, poor performance, measured by stock return, leads to the resignations of insiders, although there is not an analogous effect for earnings change. Outsiders are added after poor performance measured by both stock return and earnings 42 change. Decreases in the number of industries in which firms operate increase the departures of insiders, and firms tend to replace departing insiders with outsiders. In addition, insiders are more likely to be added to the board when a CEO nears retirement; a new CEO, who tends to add outside directors to the board, leads to the departures of insiders. According to Molz (1988), two extreme forms of corporate board organization can be characterized as managerial dominated and pluralistic. “A managerial dominated board is one made up primarily of inside directors (officers of the firm). A pluralist board is one made up of more diverse directors, and the managers of the corporation are not the dominant group. Other types of boards identified here and elsewhere generally fall somewhere between these two extremes. A better empirical understanding of these extremes will facilitate further study of variations within them” (Molz, 1988, p.236). Molz (1988) aggregated multiple measures of managerial or pluralistic control, including joint chairman/CEO, outside-dominated social responsibility committee, inside versus outside directors, frequency of board meetings, salary ratio of the highest paid to the second highest-paid executive, stockholdings of inside and outside directors, representation of woman and identifiable minority groups, and tenure of the chairman/CEO, into one single scale, using a sample of 50 firms selected at random from the 1983 Fortune 500 Industrial list. Molz (1988) reported that there was no significant relationship between the degree of managerial control on the board and financial performance as measured by ROA, ROE and total return to shareholders. In Fosberg (1989, p.32), it is noted that “[a]gency theory assigns to outside directors the role of monitoring the firm’s management to make certain that it performs its duties in a manner consistent with the best interests of shareholders”; thus increasing the percentage of outside directors on the board may enhance management performance. To test this managerial monitoring hypothesis, a paired sample methodology is used where firms in the same industry of comparable size and capital structure and with a significantly different proportion of outside directors on the board are paired, producing 127 pairs of firms over the five years of 1979 – 1983. Outside directors are defined to be those directors who are not current or former members of the firm’s management or their relatives. 43 Fosberg (1989) found that agency theory could not be confirmed by his analysis. Specifically, no relationship is found between the proportion of outside directors and the various variables used to gauge managerial performance, including ROE, sales, expenses, number of employees, sales to total assets, expenses to total assets, and number of employees to total assets ratio. Fosberg (1989, p.32) argued that “[t]here are at least two feasible explanations for this finding. First, management may succeed in getting outside directors elected to the board who are either incapable or unwilling to properly discipline management. In this eventuality, outside directors would not be providing the monitoring services contracted for by the shareholders. A second explanation is that the other mechanisms for controlling the agency costs associated with the separation of ownership and control, such as the market for corporate control, effectively motivate and discipline management, thereby leaving little for the board to do in this regard.” Schellenger, Wood and Tashakori (1989) believed that the conflicting empirical evidence with respect to the existence or non-existence of a board composition effect on financial performance might be due to failure to control risk. “Differences in performance may reflect nothing more than differences in risk. Conversely, observed equal performance may ignore differences in risk. Difference in financial performance must be evaluated relative to the risk associated with that performance. Consequently, financial performance is best measured by risk-adjusted shareholder wealth” (Schellenger et al, 1989, p.458). After testing a sample of 526 U.S. firms with complete data for the year 1986, they located a positive association between the percentage of outside directors and performance as measured by ROA and risk-adjusted shareholder return. Consequently Schellenger et al (1989, p.465) suggested that “[t]his study provides support for advocates of outsider representation on the boards of corporations.” In addition, it is reported that the percentage of outsiders and standard deviation of returns, a proxy for total risk, are negatively correlated, and the proportion of outside directors and beta, a measure of systematic risk, are positively correlated. According to the researchers, “[w]hat this suggests is that firms with a greater percentage of outsiders on the board have greater systematic risk, but less total risk. The difference implies that 44 unsystematic risk is less for firms with a greater percentage of outsiders on the board” (Schellenger et al, 1989, p.463). Using the same sample and database in Hermalin and Weisbach (1988), Hermalin and Weisbach (1991, p.102) attempted “… to measure differences in firm performance caused by board composition and ownership structure. These two variables are intended to measure the direct incentives and monitoring faced by top management.” Like their previous study, the directors of sample companies are classified as insiders, “greys” and outsiders. According to their results, at low levels of management shareholdings (less than one percent), performance, measured by both q and EBIT, improves with increases in ownership; beyond one percent, performance declines with ownership. There does not appear to be a relation between board composition and corporate performance, although firms with longer median tenures of outside directors tend to have higher performance measured as q. As observed by Pearce II and Zahra (1992, p.414), “[d]espite the wide recognition of the consequences of board composition for company survival and performance, very few empirical studies have been undertaken to explain its determinants.” To reduce this gap in the literature, they collected data through a mailed questionnaire directed to the CEO (or president) of the Fortune 500 corporations; 119 responses were received. Four measures of corporate performance are employed in Pearce II and Zahra (1992) – ROA, ROE, earnings per share (EPS) and net profit margin. Each is operationalized as the 1983-1985 average for the three years preceding, and the same criteria are measured for the three years following 1987-1989. Outside directors are classified into two groups - affiliated and non-affiliated directors. The scholars acknowledged that their study focused on composition variables at a given point in time; however, it is unclear what the given point is. They reported that effective past performance, in terms of ROA, ROE and EPS, was associated with larger boards and lower representation of outsiders; large boards and high representation of outsiders are positively associated with future performance measured by the three performance criteria. It appears that the distinction between 45 affiliated and non-affiliated outsiders is not as important to performance as one would expect based on the normative literature. Pearce II and Zahra (1992, p.432) asserted that “the study highlights the dual role of corporate performance in the context of research on board composition” and “the results provide compelling evidence of an important relationship between board composition and firm performance” (Pearce II and Zahra, 1992, p.434). Daily and Dalton (1992) argued that empirical works in the area of corporate governance had concentrated on the largest of firms, and the equivocal results might be attributable to the reliance upon the large scale firms. Their study, then, focuses on successful, but modestly sized firms, because “[i]t is in these organizations that the impact of individual actions may be more salient” (Daily and Dalton, 1992, p.376). Their sample is composed of the 100 fastest-growing small publicly held companies in the U.S. in 1990. Although no significant performance differences were found when CEOs elected the dual versus the independent structure, Daily and Dalton (1992) located a positive relation between total numbers and proportion of outside directors and performance indicated by price/earnings ratio. One year later, they produced another paper in which a sample of 186 small corporations with relevant data in 1990 is tested. Adopting the same board and performance variables as in Daily and Dalton (1992), Daily and Dalton (1993) reported similar findings to those in Daily and Dalton (1992). It is confirmed that there are no significant performance differences when CEOs elected the dual versus the independent structure, and there is a positive relationship between total numbers and proportion of outside directors and financial performance. Thus “[i]t seems that firms adhering to suggested board reforms realize performance advantages” (Daily and Dalton, 1993, p.75). In Yermack (1996, p.189), the author decided to test the hypothesis that “…firm value depends on the quality of monitoring and decision-making by the board of directors, and that the board’s size represents an important determinant of its performance”, using a sample of 3,438 annual observations for 452 companies between 1984 and 1991. He identified an inverse association between firm value measured as Tobin’s q and board size; all the three financial ratios used in his work, i.e., ROA, sales over assets 46 and return on sales, have negative associations with the board-size log. Yermack (1996, p.194) concluded that the findings were “… consistent with an interpretation that coordination, communication, and decision-making problems increasingly hinder board performance when the number of directors increases.” However, the influence of board composition on firm value is ambiguous. In the OLS model there is a negative association between the percentage of outside directors and q, where in the fixed-effect model a positive association is found. It is also found that firm value is positively related to firm size and insider ownership, and firms are valued more highly when the CEO and chairman positions are separated. To investigate the relationships among the mechanisms to control agency problems between managers and shareholders and with firm performance measured by Tobin’s q, Agrawal and Knoeber (1996) collected data on the Forbes 800 firms in 1988; a sample of 383 firms with complete record in 1987 was obtained. The only consistent findings presented in their paper are that fewer outside directors may lead to improved firm performance or, better performance may lead to fewer outsiders on the board. Agrawal and Knoeber (1996, p.393) commented: “[t]he persistent effect of board composition on firm performance presents a puzzle. The fraction of outsiders on the board of directors is an internal decision, and so we expect it to be made to maximize firm value. Our results indicate otherwise. The negative effect of outsiders on the board on firm performance suggests that firms tend to have too many outside directors. We do not have a ready explanation for this finding.” The objective in Barnhart and Rosenstein (1998) is to test the sensitivity of simultaneous equations techniques used in corporate governance literature. They noted that “[a] number of recent empirical papers have used simultaneous equations methods, such as two- and three-stage least squares to model the relations between corporate governance variables and firm valuation … However, current theory provides little guidance in the specification of corporate governance models, and the econometric literature points out that misspecification of any of the equations in a system may result in serious bias in all of the equations. In fact, ordinary least squares (OLS) tends to be less sensitive to misspecification error …” (Barnhart and Rosenstein, 1998, p.2). 47 Their sample contains 321 U.S. firms from the 1990 Standard & Poor’s (S&P) 500. As the authors did not disclose the time frame of their research, it is unclear for which year(s) the data is analysed. Based on the mixed results of several OLS and 3SLS models, the authors suggested that their paper provided support for a curvilinear relationship between the proportion of independent directors or managerial ownership and Tobin’s q; there is stronger support to the presumption that board composition, managerial ownership and firm performance are jointly determined, in that governance changes over time to allow for value maximization. According to Barnhart and Rosenstein (1998, p.14-15), “[t]he empirical results are strongly dependent on the specification of the overall model and of the first-stage regressions. Relatively minor changes in either have profound effects on overall results … This leads to the conclusion that results using simultaneous equations methods must be interpreted cautiously, that ordinary least squares estimates should not be casually dismissed, and that sensitivity analysis is essential when estimating an empirical model whose structure is uncertain.” Dalton, Daily, Ellstrand and Johnson (1998), in order to conduct meta-analyses for both board composition and board leadership structure and their relationships to financial performance, examined empirical research surrounding this topic. It is not necessary that these relationships be the focus of an article to be included for the meta-analyses; it is only necessary that a Pearson correlation between these variables be available in the piece, or derivable from it. Therefore whether a given variable is a dependent, independent, or a control variable is not an issue. They identified 54 studies with 159 usable samples for the board composition/financial performance analysis, and located 31 studies with 69 usable samples addressing board leadership/financial performance relationships. The results for all samples suggest that particular operationalizations of board composition, including the percentages of inside, outside, affiliate and independent/interdependent directors, have virtually no effect on accounting and market performance indicators; there is no evidence of a systematic relationship between board leadership structure, i.e., joint CEO/chairman or CEO duality, and firm performance indices. Dalton et al (1998, p.284) recommended that future research should pay attention to board committees, as “… many of the critical processes and decisions of boards of 48 directors do not derive from the board-at-large, but rather in subcommittees.” It is also suggested that “[c]onsideration of multiple theories in evaluating the performance advantages of suggested corporate governance reforms may lead to a more complete understanding of the subtleties which characterize the relationships between board composition, board leadership structure and firm performance” (Dalton et al, 1998, p. 285). Denis and Sarin (1999) carried out a time-series analysis of equity ownership structure and board composition in a random sample of 583 U.S. firms with 4,563 firm-years over the 10-year period 1983-1992. The researchers labelled directors as insiders if they were employees of the firm, as affiliated outsiders if they had substantial business relations with the firm, were related to insiders, or were former employees, and as independent outsiders if they were neither insiders nor affiliated outsiders. Their results indicate that ownership and board characteristics are interrelated; specifically, insider ownership is negatively related to the fraction of independent outsiders, and board size is positively related to the fraction of independent outsiders. Large changes, both increase and decrease, in insider ownership, outside representation and board size are strongly associated with CEO replacements and corporate control threats. Changes in insider ownership are negatively related to prior stock price performance, and market-adjusted stock return appears to be higher among those firms that subsequently increase the fraction of independent outsiders, and those firms that subsequently increase board size. The authors therefore concluded that “… the predominant factors associated with ownership and control changes appear to be top executive changes, prior stock price performance, and corporate control threats” (Denis and Sarin, 1999, p.210). Bhagat and Black (2000) is the largest sample study of large U.S. corporations. The researchers constructed a “1991 sample” of 934 firms and a sub-sample of 205 firms, with board composition data in early 1991 and 1988, respectively. For the 1991 sample, during the “retrospective” period (1988-1990), board independence, proxied by the proportion of independent directors minus proportion of inside directors, correlates negatively with all the performance measures employed, including Tobin’s q, ROA, market-adjusted stock price return, and ratio of sales to 49 assets. During the “prospective” period (1991–1993), the correlation remains negative for q. There is a negative relationship between future industry sale growth and board independence; it is acknowledged that there is no good explanation for this correlation. For the 1988 sample, prior performance (1985-1987) correlates negatively with board independence in early 1988 for q and ROA, and there is a negative relationship between current performance (1988-1991) and changes in board independence over the same period for q and ROA. Thus Bhagat and Black (2000) concluded that there was a reasonably strong correlation between poor performance and subsequent increase in board independence; however, there is no evidence that greater board independence leads to improved firm performance. Coles, McWilliams and Sen (2001, p.23) argued that “… previous work has generally focused on examining subsets of governance mechanisms, typically studying one or two governance variables in any one study. Our view is that the most critical issue still to examine, is the ability of firms to choose among a number of different governance mechanisms in order to create the appropriate structure for that firm, given the environment in which it operates.” Coles et al (2001) then located a sample of 144 firms and searched for the impact of board composition, leadership structure, CEO compensation and tenure, and ownership structure on market value added (MVA), a market measure of performance, and economic value added (EVA), an accounting measure of performance, over a five-year time frame from 1984 to 1988. Contrary to their expectations supported by agency theory, the researchers found negative influences of independent director representation and CEO salary sensitivity on MVA; they also identified a positive contribution of CEO duality to EVA. Coles et al (2001) suggested that future research could investigate the roles of firm risk and diversification on performance, which might provide some insights into their findings. Singh and Davidson III (2003), in their analysis of the relations between managerial ownership, blockholder ownership, and agency costs measured in terms of asset turnover and selling, general and administrative (SG&A) expenses scaled by sales, controlled for the effects of the size and composition of board of directors on the level of agency costs. The authors classified board members as insiders, affiliated outsiders 50 and independent outsiders; their sample includes 236 observations of 118 large U.S. corporations for the years 1992 and 1994. Singh and Davidson III (2003) reported that managerial shareholdings were positively related to asset utilization, but did not serve as a significant deterrent to excessive discretionary expenses. Board size is negatively related to asset turnover, but unrelated to discretionary expenditures; board composition, however, does not seem to significantly influence agency costs. Anderson and Reeb (2004), rather than concentrating on the board’s role in alleviating agency problems between managers and shareholders, explored the board’s possible role in mitigating conflicts between opposing shareholder groups. The researchers used the 1992 S&P 500 firms for the period from 1992 to 1999 to develop their sample, obtaining 2,686 firm-year observations of 403 firms. Each firm is classified as either a family or non-family firm, using the fractional equity ownership of the founding family and the presence of family members on the board as benchmarks. They found that, in general, there was no significant relationship between board independence measured by the fraction of independent directors and firm performance measured as Tobin’s q; only when delineating the sample based on the presence of founding families did they identify a positive association between the fraction of independent directors and q. It is documented that family firms, on average, perform better than non-family firms. This finding, however, appears to be primarily driven by family firms with greater degrees of board independence relative to family firms with few independent directors. In firms with family ownership, when family control of the board exceeds independent director control, the firm’s performance is poorer; when family control is less than independent directors’, performance is better. Additional tests indicate a negative association between family member presence on the nominating committee and independent director representation on the full board; there is a positive association between the presence of institutional holdings and independent director representation on the board. The results are interpreted to imply that families often seek to minimize the presence of independent directors, while outside shareholders call for independent directors to minimize founding-family opportunism. 51 Anderson and Reeb (2004, p.209) suggested that “[t]hese findings highlight the importance of independent directors in mitigating conflicts between shareholder groups and imply that the interests of minority investors are best protected when, through independent directors, they have power relative to family shareholders.” Chan and Li (2008) noted that, with the demise of Enron and debatable accounting practices of Global Crossing, many in the U.S. had been furious with the monitoring provided by directors, especially those in the audit committee. Although the independence of audit committee is widely accepted as a must for good governance and internal control for assessing risks, little is know about the “quality” of these independent directors sitting on the committee. Thus, using a sample of Fortune 200 companies in the year 2000 and defining top executives of other publicly traded firms among independent outsiders as expertindependent directors, Chan and Li (2008) tested the influence of audit committee independence, which is measured by the percentage of independent directors on the committee, on Tobin’s q, with two thresholds of independence for boards, 50% and 35% or more expert-independent director membership, respectively. Their findings indicate that the independence of audit committee results in higher firm value when a majority of expert-independent directors serve on the board. It is also reported that directors with finance-trained backgrounds serving on the audit committee enhance firm value, when expert-independent directors are a majority in the committee; however, the presence of chief executives who are also chairman of the board is related to negative q. 3.3.2 Event Studies The studies as introduced above, using different variables, measures and models, evaluate whether board composition and structure correlates directly with performance. Some scholars, such as Rosenstein and Wyatt (1990, 1997) and Klein (1998), have suggested that an alternative way to address the concerns of whether board composition is important and whether outside or inside directors increase firm value is to examine the stock-market reactions to changes in board composition. In Rosenstein and Wyatt (1990), the writers obtained a sample of 1,251 announcements of the appointment of only one outside director and no inside directors, over the 198152 1985 period; an outside director is defined as a director who is not a present or former employee of the firm and whose only formal connection with the firm is his or her duties as a director. They categorized announcements by the director’s primary occupation as either a financial outsider (officer of any potential supplier of capital), corporate outsider (officer or employee of other corporation), and neutral outsider (outside director with any other affiliations). In the study, it is reported that the appointments of an outside director are accompanied by positive excess returns, even though most boards are numerically dominated by outsiders before the appointments. There is no clear evidence that outside directors of any particular occupation are perceived more or less valuable than others by the stockmarket. Rosenstein and Wyatt (1990, p.190) concluded that “… the addition of an outside director increases firm value. The empirical results are consistent with the hypothesis that outside directors are selected, on average, in the interests of shareholders.” Later on, the same authors argued that “[t]hough Rosenstein and Wyatt (1990) found a positive stock-market reaction to the appointment of an outside director, their evidence does not imply that the appointment of an insider is harmful to shareholders. If corporations adjust board composition to respond, in the interests of shareholders, to new challenges, then inside director appointments may also increase shareholder wealth” (Rosenstein and Wyatt, 1997, p.230). Thus, by sampling director announcements over the same period as in Rosenstein and Wyatt (1990), Rosenstein and Wyatt (1997) investigated the stock-market reactions to appointments of inside managers to corporate boards. Rosenstein and Wyatt (1997) categorized low (less than 5%), moderate (5%-25%), and high (greater than 25%) levels of insider ownership; they expected that the stock-price effects of insider appointments might vary across levels of insider ownership. Their analysis shows that, for the full sample of 170 insider appointments, the average stock-market reaction is close to zero; however, the market reaction is negative when inside directors own less than 5% of the firm’s stock, positive when their ownership level is between 5% and 25%, and insignificantly different from zero when ownership exceeds 25%. According to Rosenstein and Wyatt (1997, p.229), “[t]hese results suggest that the expected benefits of an inside director’s expert knowledge clearly 53 outweigh the expected costs of managerial entrenchment only when managerial and outside shareholder interests are closely aligned.” In Klein (1998), the author investigated market reactions to changes in board committee composition between 1992 and 1993 around the 1993 proxy mailing date, for a sample of 461 firms listed on the S&P 500 for both years. Directors are classified as insiders, outsiders and affiliates. Insiders are presently employed by the firm. Outsiders have no affiliation with the firm beyond being a number of the firm’s board. Affiliates are former employees, relatives of the CEO, or those who have significant transactions and/or business relationships with the firm. Klein (1998) reported that firms that significantly increased inside director representation on finance and investment committees experienced significantly higher abnormal returns on the proxy mailing date than firms decreasing the percentage of inside directors on these two committees. To complement the results, Klein (1998) conducted regressions of firm performance measures, including ROA, Jensen Productivity and market returns, on board and committee composition, using a sample of 485 firms listed on the 1992 S&P 500. No significant relationship is found between performance measures and the percentage of inside directors on the board as a whole; however, there are significantly positive linkages between the percentages of inside directors on finance and investment committees and performance measures. Klein (1998, p.275) thereby claimed that “[t]hese findings are consistent with Fama and Jensen’s assertion that inside directors provide valuable information to boards about the firms’ long-term investment decisions.” 3.4 Evidence from Other Regions In this section, some studies which attempt to uncover the board characteristics-firm performance relations in Belgium (Dehaene, Vuyst and Ooghe, 2001), Canada (Panasian et al, 2003), China (Peng, 2004), continental Europe (Krivogorsky, 2006), Hong Kong (Chen, Cheung, Stouraitis and Wong, 2005), Malaysia (Chang and Leng, 2004), New Zealand (Hossain, Prevost and Rao, 2001; Chin, Vos and Casey, 2004), South Korea (Choi, Park and Yoo, 2007), Sweden (Randoy and Jenssen, 2004), Taiwan 54 (Luan and Tang, 2007), and U.K. (Vafeas and Theodorou, 1998; Dulewicz and Herbert, 2004), are introduced. Vafeas and Theodorou (1998) noted that most empirical research studying the relationships between board characteristics and performance used U.S data; “[w]hile the assumption of a utility-maximizing agent is universal, each country’s regulatory and economic environment, the strength of capital markets, and current governance practices are different. As a result, the importance and value of various governance structures should be separately examined in each country” (Vafeas and Theodorou, 1998, p.384). Their sample includes 250 public U. K. firms with complete data for the 1994 financial year; four performance measures, i.e., MBT (equity capitalisation plus total liabilities divided by total assets), market to book value of equity, stock return and ROA, are used. The explanatory variables for performance are the percentages of independent and “grey” directors on the board, percentages of equity ownership by independent, “grey” and executive directors, percentages of non-executive directors serving in the audit, remuneration and nomination committees, and chairman independence. Their tests do not discern a significant link between board characteristics, director ownership and firm performance. Vafeas and Theodorou (1998, p.383) asserted that “[t]hese results are consistent with governance needs varying across firms, and contrast the notion that uniform board structures should be mandated.” Similarly, Dehaene et al (2001) found that most of the empirical studies on corporate governance were made in an Anglo-Saxon context; the legal framework and structural context in continental Europe differ strongly from those in the Anglo-Saxon business world, “[t]herefore, the corporate governance discussion in continental Europe countries ought to be based on country-specific research” (Dehaene et al, 2001, p.394). They analysed board characteristics in a sample of 122 large Belgian companies, with complete accounting and board information in 1995, to assess the impact of board composition and structure on firm performance, as measured by industry-adjusted ROA and ROE. Dehaene et al (2001) identified a positive relationship between the number of external directors and ROE; however, where the functions of CEO and chairman are combined, 55 ROA appears significantly higher. They suggested that “[d]istinct corporate governance models for companies exist because they operate in different business context. Comparing these models in isolation can lead to futile conclusions” (Dehaene et al, 2001, p.394). In Hossain et al (2001), the authors assessed the efficacy of monitoring by directors in New Zealand; specifically, they investigated whether the impact of board composition and structure on firm performance was influenced by the Companies Act of 1993 and related legislation. “The passage of the 1993 Companies Act and accompanying legislation provides a unique environment for studying the monitoring ability by the board of directors because its direct purpose is to increase and enhance monitoring and firm performance. The study has implications for whether corporate governance can be legislated. The findings should be of interest in light of attempts by some quarters to get the government through its regulatory bodies involved in setting guidelines for corporate governance” (Hossain et al, 2001, p.120-121). Data for a sample of firms listed on the New Zealand Stock Exchange (NZSE) was collected for the years 1991 through 1997, resulting in 633 firm-years of observations. Directors are placed into one of three categories: inside directors, affiliated directors or independent directors; independent directors are defined as individuals who are not an active or retired employee of the firm, do not have close business ties with the firm, and are not a representative of or appointed by a majority shareholder of the firm. Their findings suggest that while the proportion of independent directors has a positive influence on firm performance measured as Tobin’s q, the Companies Act of 1993 with its obvious implications for the corporate governance role of the board has not resulted in an increased sensitivity of firm performance to the proportion of independent outsiders on the board. Three years later, another New Zealand study, Chin et al (2004) reports contrasting results. The initial dataset in Chin et al (2004) includes all firms listed on the NZSE from 1996 to 2001; the final sample comprises a total of 426 annual observations over the five-year period. The value of firm is also measured by Tobin’s q, which is supposed to be explained by director ownership, board size and percentage of outside directors; outside directors are 56 defined as those who are not current or former employees of the firm. According to their analysis, there does not seem to be a rise-fall relationship in performance relating to ownership structure, nor to the percentage of outside directors, nor to board size. The authors believed that this might be due to endogenous factors or due to the small size of the New Zealand pool of corporate directors. In 1993, the Toronto Stock Exchange (TSE) nominated a committee under the leadership of Peter Dey to assess the corporate governance practices of publicly held Canadian corporations. The Dey Committee ultimately proposed a set of guidelines for improved board performance, highlighting the importance of independent directors. Two years later, the TSE adopted the recommendations as a listing requirement whereby companies had to specify their governance practices with respect to each of the guidelines and, where company practices differed from the guidelines, an explanation for the difference. To investigate the consequences of the TSE’s adoption of the Dey Committee guidelines on firm performance as measured by Tobin’s q, Panasian et al (2003) developed a sample of 274 firms with a complete record of board and ownership characteristics for the five years 1993-1997, from the companies comprising the 1995 TSE 300 index. Directors are classified as either related or unrelated as per the Dey Committee’s definition of an unrelated director - “… a director who is independent of management and is free from any interest and any business or any other relationship which could, or could reasonably be perceived to materially interfere with the director’s ability to act with a view to the best interests of the corporation, other than interests and relationship arising from shareholdings” ( Dey Report, 1994, Guideline 2). Panasian et al (2003) found that the adoption of the recommendation that boards comprise a majority of unrelated directors improved performance among companies which became compliant during the period following the Dey Report, and companies which were always compliant and increased their proportion of outsiders thereafter; however, this improvement appears to be limited to firms that had average levels of q less than one for the two years prior to introduction of the guidelines. For the entire sample, the relationship between q and the percentage of unrelated directors is insignificant, although the TSE’s adoption of the Dey guidelines might enhance performance for all the companies. For the subset of firms with pre-Dey 57 average q less than one, there is a positive relationship between the proportion of outsiders and performance. As Tobin’s q has been used by some scholars to estimate the extent of agency problems with the firm, Panasian et al (2003, p.30) concluded that “… increasing outsiders is most beneficial for firms that are most likely to have agency problems.” Peng (2004, p.453) noted that “[a]lthough the debate on the link between board composition and firm performance is hardly resolved in developed economies, appointing outsiders to corporate boards has become an increasingly widespread practice in emerging economies going through institutional transitions such as China, which provides an interesting ‘research laboratory’.” Based on a database covering 405 Chinese listed firms and 1211 company-years of observations during 1992-1996, the analysis of Peng (2004) indicates that outside directors do made a difference in firm performance, if such performance is measured by sales growth; outside directors, however, have little impact on financial performance measured as ROE. CEO duality displays a positive effect on ROE and sales growth; a possible explanation is that “… CEO duality may be more important for firms in turbulent environments, as China’s transitions may be argued to be” (Peng, 2004, p.463). Peng (2004) also documented a bandwagon effect behind the diffusion of the practice of appointing outsiders to corporate boards. In Dulewicz and Herbert (2004), board characteristics and evaluations about how boards actually worked in 1997 are assessed in relation to firm performance over the ensuring three years. The authors sent questionnaires to the chairmen of U.K. listed companies in 1997; 134 responses were received. In the questionnaire, there are 117 indicators of current board practice relating to 16 key tasks of the board and ratings of “potential for improvement” on the same indicators. Performance data on these companies were sought, resulting in a sample of 86 companies. They reported that the number of executive directors is positively associated with cash flow return on total assets (CFROTA). The number and proportion of NEDs are negatively related to sales, which in turn is positively associated with the percentage of equity held by directors. In addition, a significant correlation is found between 58 CFROTA and the chairman independence factor, i.e., companies with a chairman who is not the CEO or is a NED may perform better. The evidence provided by the correlations between current practice and potential ratings on the 16 board tasks and firm performance is confusing. Dulewicz and Herbert (2004) acknowledged that board practice examined in their study was generally not linked to performance. Randoy and Jenssen (2004) explored how the attractiveness of board independence was contingent on the level of competition in the product market; they obtained 294 firmyear observations from 98 randomly selected companies listed and headquartered in Sweden, over the 1996-1998 period. Board independence is measured by the percentage of outside directors; an outside director is defined as someone who is not, and has not been employed by the firm. Product market competition is proxied by a two-year moving average profit margin for the 19 industry groups assigned by the Stockholm Stock Exchange. The sample set is then split into three subsets based on the industry’s level of product market competition; it is acknowledged that the partition points are not theoretically motivated. Randoy and Jenssen (2004) reported a negative effect on firm performance by board independence among firms that faced high levels of product market competition; this is the case for both Tobin’s q and ROE. They also identified a positive effect on performance measured as q by board independence among firms that faced low levels of product market competition. They concluded that “… board independence is less relevant or even redundant in highly competitive industries, where the firm is already ‘monitored’ by a competitive product market … on the other hand, board independence enhances firm performance among companies facing less competitive product market” (Randoy and Jenssen, 2004, p.281). Chang and Leng (2004) observed that poor governance standards in both private and government-owned firms were blamed in part for the Asian financial crisis of 19971998, demonstrating the importance of effective corporate governance in developing countries. To establish corporate governance factors that significantly influence the financial performance of Malaysian firms, they developed a sample of 77 randomly selected public companies in Malaysia, over the period 1996-1999, with each firm-year representing a separate observation. 59 The independent variables used in their regressions are factors that may affect firm performance as measured by ROE and dividend payout ratio, including firm size measure by sales, leverage measured by total debt to total capital, ownership concentration measured by the proportion of shares owned by the largest shareholder, proportion of shares owned by institutional investors, proportion of NEDs, and dummy variables for CEO duality and NED chairman on the audit committee. There are three variables which are found to be significant in influencing ROE, i.e., firm size, leverage and institutional ownership. However, several corporate governance variables, i.e., ownership concentration, proportion of NEDs, CEO duality, and NED chairman on the audit committee, do not have any impact on corporate performance. In Chen et al (2005), the researchers attempted to address three research questions. Does concentrated family ownership affect firm operating performance and value? Does it affect dividend policy? What is the impact of corporate governance (CEO duality, board composition and audit committee) on performance, value and dividend payouts in family controlled firms? According to Chen et al (2005), Hong Kong is an appropriate market for conducting their project because it is characterized by widespread family control of publicly listed companies, while at the same time having a common law legal system and corporate governance influenced by recent developments in the U.K. Their initial dataset consists of all companies listed on the Hong Kong Stock Exchange (HKSE) in 1995, which they followed through 1998; the final sample contains a balanced panel of 1,648 firm-years for 412 firms. Their empirical analysis does not establish a relation between family ownership and proxies for firm performance - ROA, ROE and MBT. Only for small market capitalisation firms there is a negative association between dividend payout ratio and family ownership up to 10% of the company’s stock and a positive association for family ownership between 10 and 35%; these firms also exhibit low sensitivity of dividend payouts to performance. Board characteristics, i.e., the proportion of independent NEDs, outsider-dominated board and presence of audit committee, have 60 little impact on firm performance and dividend policy, although there is a negative relation between CEO duality and MBT. Chen et al (2005, p.448) recommended that “… corporate governance in Hong Kong needs to be strengthened. In addition, more effort is needed in order to ensure the true independence of non-executive directors and that they are able to perform an adequate monitoring function … Our findings have relevance for other East Asian countries too, which are characterized by even lower standards of corporate governance and investor protection compared to Hong Kong.” Using the same measures for firm performance as in Chen et al (2005), Krivogorsky (2006, p.177) investigated “… the empirical validity of the claims that the composition of the board and ownership structure affects a firm’s profitability after considering the mechanisms by which a European company is directed and controlled (as described in European Corporate Governance Codes).” The 87 firms used in his analysis are continental European companies listed on the NYSE as foreign registrants during 2000 and 2001. Board characteristics tested include the percentages of inside directors, independent directors, and scholars on the board, and CEO duality. It is disclosed that “independence” involves an absence of close family ties or business relationships with management and controlling shareholder(s); however, little is known about the standards of scholars endorsed in this paper. It is documented that the percentage of independent directors has a positive correlation with ROA and ROE. The authors also identified a positive relation between the level of institutional ownership and ROA, ROE and MBT. Consequently Krivogorsky (2006, p.191) confirmed that “… the theoretical predictions of agency theory on a positive relationship between outside (independent) director and firm performance … are also applicable in the European environment as well.” Luan and Tang (2007) pointed out that, theoretically and empirically, the linkage between outside directors and firm performance was not conclusive in previous studies; they suspected that the mixed results were due to the failure to meet the requirements of the independence of outside directors. As it is argued that in Taiwan there is a rigorous definition of outside director independence, a dataset from the island is employed to examine the impact of independent director assignment on a firm’s performance. There 61 are 259 firms having been tested in their model, in which ROE in 2002 is the dependent variable; independent director assignment, a dummy variable to assess whether independent directors were present in 2002, is used as an independent variable. Their findings suggest that, after controlling for past performance, independent director appointments have a positive effect on ROE. Luan and Tang (2007, p.640) believed that “[t]his may be the first paper to show that independent outside directors may be beneficial to the firm.” Choi et al (2007) introduced that, despite the inconclusive empirical results in the U.S. and elsewhere, the idea of a monitoring board was vigorously imported and implemented by South Korean authorities in the aftermath of the Asian financial crisis. As weak corporate governance was viewed as one of the factors that had contributed to the crisis, the government instituted a series of regulatory changes in 1998 and for listed firms required at least 25% of the board composed of non-executive directors. Since outside directors were uncommon in this country prior to 1997, according to Choi et al (2007), post-crisis Korea presents a natural laboratory for testing the effect of board independence enforced by the authorities. Using a sample of 1,834 firm-year observations from 1999 to 2002, the authors found that the proportion of outside directors had a positive effect on firm performance as measured by Tobin’s q; the effect is stronger for independent directors than for “grey” directors who may have professional ties with the firm. Choi et al (2007, p.942) concluded that “… the presence of independent outsiders is critical in an emerging market that is subject to external shocks and that may lack sufficient liquidity as well as indigenous institutional infrastructure.” However, they also argued that, given the still significant influence of insiders in Korean firms, their results should not be interpreted as evidence supporting American-style superindependent boards; rather, it may support the notion that an injection of independent directors into insider-dominated boards could enhance performance. 3.5 Summary and Limitations As introduced in Section 2, there are several papers which provide evidence on the relationship between board characteristics and the “bottom line” performance of Australian public companies; Appendix 1 exhibits a summary of their samples, research 62 variables and method, and major findings. In Section 3 and 4 some papers from the U.S. and other regions are reviewed; a summary of these studies could be found in Appendix 2. 3.5.1 Australian Studies Seven studies, i.e., Muth and Donaldson (1998), Calleja (1999), Lawrence and Stapledon (1999), Cotter and Silverster (2003), Kiel and Nicholson (2003), Balatbat et al (2004) and Hutchinson and Gul (2004), with mixed evidence, conclude that board independence may not add value to Australian corporations. Only one study, Bonn et al (2004), indicates that greater board independence may enhance performance. According to Muth and Donaldson (1998), the level of board independence, constructed by CEO duality, board size, proportion of NEDs, interest alignment with owners and average age across all directors, is negatively associated with shareholder wealth and sales growth. Lawrence and Stapledon (1999) asserted that the proportion of independent directors was negatively related to the ratio of revenue to assets. Kiel and Nicholson (2003) discovered a negative correlation between the proportion of outside directors and the market-based performance measure of Tobin’s q. Calleja (1999) and Hutchinson and Gul (2004) could not find a relation between shareholder return or ROE, and the proportion of NEDs. Cotter and Silverster (2003) reported that neither board nor committee independence was significantly associated with firm value. Balatbat et al (2004) concluded that, for IPO firms, board composition was not related to variation in operating performance, although firms with dual leadership tended to perform better. At last, Bonn et al (2004) documented a positive association between outside director ratio and ROA, for their Australian sample firms. As illustrated in Appendix 1, all these studies apart from Kiel and Nicholson (2003), Balatbat et al (2004) and Hutchinson and Gul (2004) suffer from the limitation of small sample size. Muth and Donaldson (1998) suggested that a sample size closer to 200 would have been preferable; effects of boards on performance tend to be small which means that more statistical power is needed to detect significant relationships. Calleja (1999) also acknowledged that the small sample size in her study made it difficult to reach any firm conclusions. 63 It appears that all Australian studies concentrate on the concern whether board characteristics and subsequent performance are related. As suggested by Bhagat and Black (2000), board composition and structure may affect firm performance, and firm performance may also affect board demographics; thus the projects could have been designed to address both of the two research questions, i.e., does board independence have any influence on firm performance, and does firm performance have any influence on the firm’s board independence? In Muth and Donaldson (1998), Calleja (1999), Kiel and Nicholson (2003), Bonn et al (2004) and Hutchinson and Gul (2004), the researchers used the simplistic executive (inside), non-executive (outside) director dichotomy as an indicator of board independence. Kiel and Nicholson (2003) acknowledged that this did not allow them to calculate the presence of “grey” directors as identified by Baysinger and Bulter (1985). In a review of the directors of 100 Australian companies, randomly selected from the top 500 companies listed on the ASX, Clifford and Evans (1997) found that 35% of NEDs in their sample were involved in transactions with their companies which potentially threatened their independence. They noted that the combination of insider and “grey” directors would constitute a majority of the board, and this could lead to companies appearing to comply with the recommendations for board independence through possessing a non-executive majority on the board, whilst in fact being controlled by internal management. Thus, in examining the relationships between board independence and firm performance, it would be appropriate to classify the NEDs as independent or affiliated. Calleja (1999) and Kiel and Nicholson (2003) also suffer from the limitation that they are essentially cross sectional without controls. Lawrence and Stapledon (1999) controlled for firm size and board size; Cotter and Silverster (2003) used firm size, and Bonn et al (2004) used firm age, as the only control variable in their works. There are many factors which could contribute to a company’s performance; it would be appropriate to add more controls to cover some important contributing factors. The performance measures employed by Calleja (1999), Kiel and Nicholson (2003), Bonn et al (2004), Balatbat et al (2004) and Hutchinson and Gul (2004) are quite limited; on the other hand, the measures used in Lawrence and Stapledon (1999) may be “noisy”, that is, too many variables without sufficient explanation about their choice. 64 For example, it is unclear why Lawrence and Stapledon (1999) used the numbers of employees as an accounting performance measure. In addition, it appears that most Australian researchers did not examine the consistency between their performance measures, to verify the results of their tests using these measures. The one year performance in Calleja (1999), Cotter and Silverster (2003), Bonn et al (2004) and Hutchinson and Gul (2004) may be too short for the purpose of their research. Bonn et al (2004) recognized that generalizability of their findings would have been enhanced if they had used data from a longer period, and the one-year lag might not be sufficient to examine the influence of board composition on firm performance. In Balatbat et al (2004), due to the combination of cross-sectional and time-series data (firm-year observations), the OLS and 2SLS regressions conducted may be unsuitable for the purposes of their project (Leamer, 1978). As recommended by Chang and Leng (2004), the appropriate method of analysis would involve panel data regression technique. There are two frequently used estimation techniques for panel data regression, i.e., the fixed-effect model and random-effect model (Gujarati, 2003); the Hausman test, a model of specification test, can be used to decide between the two models (Hausman, 1978). Taking into consideration of the limitations in prior research as discussed above, and the increased pressure on public companies for greater board independence as introduced in Chapter 2, there is perhaps a need for an in-depth investigation of the relationships between board independence and corporate performance in Australia. 3.5.2 Overseas Studies Some studies give evidence that greater board independence may improve performance or firm value. Baysinger and Butler (1985) reported that firms that had invited relatively more independent directors onto their boards in 1970 enjoyed relatively better records of industry-adjusted ROE in 1980, although changes in the proportion of independent directors did not produce significant changes in performance over time. Schellenger et al (1989) identified a positive correlation between ROA, risk-adjusted shareholder return and the proportion of outside directors; moreover, firm risk is lower for firms with a greater percentage of outsiders. 65 Pearce II and Zahra (1992) found that high representation of outsiders was associated positively with future performance criteria such as ROA, ROE and EPS; it appears that the distinction between affiliated and non-affiliated outsiders is not as important to performance as one would expect based on the normative literature. Daily and Dalton (1992, 1993) reported a positive relationship between total numbers and proportion of outside directors and firm performance measured by price/earnings ratio, although no significant performance differences were found when CEOs elected the dual versus the independent structure. Dehaene et al (2001) documented a positive relationship between the number of external directors and industry-adjusted ROE; industry-adjusted; ROA appears higher where the functions of CEO and chairman are combined. Hossain et al (2001) and Choi et al (2007) found that the proportion of independent directors had a positive influence on Tobin’s q, and Krivogorsky (2006) discerned a positive correlation between ROA, ROE and the fraction of independent directors on the board. The evidence in Luan and Tang (2007) suggests that firms that choose to appoint independent directors tend to have a higher ROE. It is concluded in Chan and Li (2008) that the independence of audit committee enhances firm value when a majority of expert-independent directors serve on the board; the presence of CEO duality leads to negative q. In an event study Rosenstein and Wyatt (1990) discovered that the appointment of an outside director was accompanied by positive abnormal return, even though most boards had been dominated by outsiders before the appointments. Stock price neither increases nor decreases on average when an insider is added to the board; the stockprice effects of insider appointments vary significantly across levels of insider ownership (Rosenstein and Wyatt, 1997). In contrast, a number of papers provide results indicating that firms with a greater percentage of outside or independent directors may perform worse, or firms with a greater percentage of inside directors may perform better. Kesner (1987) noted that the proportion of insiders was positively related to current firm performance, in terms of profit margin and ROA, and future performance measured by total return to investors. Agrawal and Knoeber (1996) identified a negative correlation between the proportion of outside directors and Tobin's q. 66 According to Klein (1998), firms increasing inside director representation on finance and investment committees experience higher abnormal returns than firms decreasing the percentage of inside directors on these committees, and there are positive linkages between the percentages of inside directors on finance and investment committees, and ROA and market return. Coles et al (2001) identified a negative influence of independent directors on market performance, and a positive contribution of CEO duality to accounting performance. In Dulewicz and Herbert (2004), the number of executive directors is positively associated with CFROTA, and the number and proportion of NEDs are negatively related to sales turnover, although companies with a chairman who is not the CEO or is a NED may perform better. Several researchers addressed the concern whether board composition is endogenously related to firm performance, resulting in inconsistent findings. According to Hermalin and Weisbach (1988), independent outsiders are added following poor stock return and earnings change, and poor stock return leads to the resignations of insiders. Pearce II and Zahra (1992) found that effective past performance in terms of ROA, ROE and EPS was associated with lower representation of outsiders. However, in Denis and Sarin (1999), market-adjusted stock return appears to be higher among those firms that subsequently increase their fraction of independent outsiders. In the largest sample study of large American firms, Bhagat and Black (2000) concluded that there was a reasonably strong correlation between poor performance measured as ROA, q, market-adjusted stock price return and ratio of sales to assets, and subsequent increase in board independence, proxied by the proportion of independent directors minus proportion of inside directors. There is no evidence that greater board independence leads to improved firm performance. There are also some papers which fail to identify a clear correlation between board characteristics and performance, or produce ambiguous evidence. In Molz (1988), Fosberg (1989), Hermalin and Weisbach (1991), Dalton et al (1998), Vafeas and Theodorou (1998), Singh and Davidson III (2003), Chin et al (2004), Chang and Leng (2004) and Chen et al (2005), no significant link is found between board characteristics and corporate performance, except a negative relationship between CEO duality and MBT as documented in Chen et al (2005). 67 Yermack (1996) discerned a negative association between the percentage of outside directors and q in the OLS model, and a positive association in the fixed-effects model; firms are valued more highly when the CEO and chairman positions are separated. Peng (2004) demonstrated that, although outside directors made a difference in sales growth, they had little impact on ROE; consistent with Dehaene et al (2001) and Coles et al (2001) but in contrast to Yermack (1996), Dulewicz and Herbert (2004) and Chen et al (2005), CEO duality gives a positive contribution to performance. A number of studies attempt to offer different insights into the board compositionperformance link. As noted by Pfeffer (1972), firms that deviate more from an optimal inside-outside director orientation are less successful when compared to industry standards than those that deviate less from an optimal board composition. Barnhart and Rosenstein (1998) presented some support for a curvilinear relationship between the proportion of independent directors and q-value; their findings indicate that board composition, managerial ownership and firm performance may be jointly determined. Panasian et al (2003) found that, for the entire sample in their study, the relationship between Tobin’s q and the percentage of unrelated directors was insignificant; however, for the subset of firms with average levels of q less than one in the past two years, there was a positive relationship between the proportion of independent directors and performance. It is therefore proposed that independent directors are more beneficial for firms that are more likely to have agency problems. Similarly, Anderson and Reeb (2004) observed that, in general, there was no significant relationship between board independence and firm performance. However, in firms with founding-family ownership, there is a positive association between the fraction of independent directors and Tobin’s q, suggesting that there are performance premiums for family firms with greater degrees of board independence relative to non-family firms or family firms with insider-dominated boards. Randoy and Jenssen (2004) identified a negative effect on performance by outside directors among firms that faced high levels of product market competition for q-value and ROE, and a positive effect on performance measured as q by outside directors among firms that faced low levels of product market competition. They concluded that board independence was less relevant or even redundant in highly competitive industries where the firm was already “monitored” by a competitive product market, 68 while board independence might enhance performance among companies facing less competitive product market. From Appendix 2, it appears that most overseas papers suffer from the same limitations as identified in Australian research: small sample size (Pfeffer, 1972; Molz, 1988; Pearce II and Zahra, 1992; Daily and Dalton, 1992; Coles et al, 2001; Dehaene et al, 2001; Dulewicz and Herbert, 2004; Krivogorsky, 2006); short-term observation of firm performance (Pfeffer, 1972; Molz, 1988; Schellenger et al, 1989; Daily and Dalton, 1992, 1993; Agrawal and Knoeber, 1996; Klein, 1998; Vafeas and Theodorou, 1998; Dehaene et al, 2001; Luan and Tang, 2007; Chan and Li, 2008); limited performance measures (Baysinger and Bulter, 1985; Agrawal and Knoeber, 1996; Barnhart and Rosenstein, 1998; Denis and Sarin, 1999; Hossain et al, 2001; Peng, 2004; Chin et al, 2004; Chang and Leng, 2004; Luan and Tang, 2007; Choi et al, 2007; Chan and Li, 2008); limited control variables (Pfeffer, 1972; Baysinger and Bulter, 1985; Kesner, 1987; Molz, 1988; Fosberg, 1989; Schellenger et al, 1989; Pearce II and Zahra, 1992; Daily and Dalton, 1992; Dalton et al, 1998; Dehaene et al, 2001; Dulewicz and Herbert, 2004; Chin et al, 2004; Chang and Leng, 2004); simplistic dichotomy of inside and outside directors as a measure for board independence (Pfeffer, 1972; Kesner, 1987; Molz, 1988; Fosberg, 1989; Schellenger et al, 1989; Daily and Dalton, 1992, 1993; Agrawal and Knoeber, 1996; Dehaene et al, 2001; Dulewicz and Herbert, 2004; Randoy and Jenssen, 2004; Chin et al, 2004; Chang and Leng, 2004); and failure to examine whether board characteristics are endogenously related to performance (Pfeffer, 1972; Baysinger and Bulter, 1985; Kenser, 1987; Molz, 1988; Fosberg, 1989; Schellenger et al, 1989; Hermalin and Weisbach, 1991; Daily and Dalton, 1992, 1993; Yermack, 1996; Klein, 1998; Vafeas and Theodorou, 1998; Coles et al, 2001; Dehaene et al, 2001; Hossain et al, 2001; Singh and Davidson III, 2003; Anderson and Reeb, 2004; Dulewicz and Herbert, 2004; Randoy and Jenssen, 2004; Chin et al, 2004; Chang and Leng, 2004; Chen 69 et al, 2005; Krivogorsky, 2006; Luan and Tang, 2007; Choi et al, 2007; Chan and Li, 2008). Moreover, the positive relationship between the percentage of independent directors on the board in 1970 and relative firm profitability in 1980, identified in Baysinger and Butler (1985), may not be taken as strong support for board independence; the authors did not offer explanation of why it took such a long time (10 years) for the benefits of independent director monitoring to manifest themselves in firm profitability. Similar to the Australian study of Balatbat et al (2004), in Hermalin and Weisbach (1991), Denis and Sarin (1999), Hossain et al (2001), Randoy and Jenssen (2004), Chin et al (2004) and Krivogorsky (2006), due to the use of firm-year observations, the OLS regressions performed by these researchers may not be suitable for the purposes of their analyses. In the circumstances, they could have employed panel data regression technique, such as the fixed-effect models in Yermack (1996), Anderson and Reeb (2004), Chang and Leng (2004) and Chen et al (2005), or the random-effect models in Choi et al (2007). This chapter presents a review of the literature testing board composition and financial performance relationships; from the review it appears that such empirical studies have become a popular topic, although their results have been inconsistent. Zahra and Pearce II (1989) identified some sources of these conflicting findings, including differences in time-frames, samples, performance measures, and the operational definitions employed with regard to the outsiders’ variable. Beyond these methodological concerns, however, like most scholars in this field Zahra and Pearce II (1989) expected a positive association between board outsiders’ proportion and company performance, based on the theory of firms developed by Jensen and Meckling in their 1976 publication. In addition to the quantitative research as introduced above, there are some works which use the qualitative methods to investigate the links between directors and firm performance. For example, by employing a pattern matching analysis of five cases, Kiel and Nicholson (2007) tested the hypothesized impact of board demography on performance as predicted by agency theory, stewardship theory and resource dependence theory. They concluded that while each theory could explain a particular case, no single model could explain the general pattern of results. 70 Chapter 4. Theoretical Development 4.1 Introduction As noted by Dalton et al (1998), Cotter and Silverster (2003) and Krivogorsky (2006), most researchers in this field have used agency theory, which promises a positive influence of board independence on corporate performance, as their underlying theoretical arguments. However, as shown in Chapter 3, “[p]rior research does not establish a clear correlation between board independence and firm performance” (Bhagat and Black, 2000, p.5). Consequently, some authors, for example, Dalton et al (1998, p.285), suggest that “consideration of multiple theories (beyond agency theory) … may lead to a more complete understanding.” There are five sections in this chapter. Section 2 presents an examination of the evolving perspectives on the contribution of board of directors. The potential relationships between board independence and firm performance, as proposed by these theories, are then summarized in Section 3, to provide a theoretical framework for this study. As a result, five research hypotheses are developed and stated in Section 4. 4.2 Theories of Board of Directors Juran and Louden (1966) remarked that it was an astonishing fact that the job of board of directors was, in proportion to its intrinsic importance, one of the least studied in the entire spectrum of industrial activities. Thirty years later, Johnson, Daily and Ellstrand (1996) observed that there was no convergence in the understanding of the role of the boards of directors. According to Stiles and Taylor (2001, p.4), “[t]he description of what a board of directors is for varies depending on which particular theoretical approach scholars take …” The roles of boards of directors, as suggested by legalistic theory, agency theory, stewardship theory, resource and strategy theories, and organizational portfolio theory, are introduced as fellows. 71 4.2.1 Legalistic View Regarded as the earliest theory on corporate governance, the legalistic approach is grounded in the Companies’ Act and common law (Ong and Lee, 2000). Chaganti, Mahajan and Sharma (1985) pointed out that almost every country’s law decreed that the business affairs of a corporation be managed under the jurisdiction of its board, and boards could contribute to the performance of their firms by carrying out their legally mandated responsibilities. While the board may not necessarily need to have technical expertise related to the firm’s services or products, it has legal power to ensure that the company reaches or maintains a certain level of acceptable performance. Therefore a board may see its primary function as controlling performance (Chaganti et al, 1985), and directors are required to undertake the duty in “good faith” and “due care” for the benefit of the company and shareholders (Pennington, 1986). Since a firm is a legal entity separated from its owners to “furnish immortality”, the directors have an obligation to ensure its survival (Koontz, 1967; Henn, 1974; Kosnik, 1987); the ultimate responsibility for corporate performance rests with the board (Tricker, 1994). According to Juran and Louden (1966), for directors to achieve “control before results”, they should set the course of the business so that the general directions are determined in advance and they should preview up-coming plans developed by the management. The board may achieve some of these activities through a thorough study of available leading indicators, a broad-based judgment of what lies ahead, a periodic re-evaluation of the company’s plans, and developing alternative plans to circumvent unexpected events (Juran and Louden, 1966). For the control role of directors to be effective, a board which is independent from management would be preferred, as the board would be more likely to get peer status with the CEO and serve as the decision-makers in the organization (Vance, 1978, 1983). However, the legalistic view does not propose a positive relationship between board independence and corporate performance, although it is suggested that an independent board, with effective control over the company, may ensure that the company reaches or 72 maintains an acceptable or satisfactory level of performance (Chaganti et al, 1985; Pennington, 1986; Patton and Baker, 1987). There is no precise definition of acceptable or satisfactory performance. The performance considered being acceptable or satisfactory by directors might be different from that of the managers, or shareholders. According to Parkinson (1993), the duty of care says that directors must act in accordance with the standard of the “ordinary prudent man”, which provides a low standard and one that has been accused of being devoid of content. The fiduciary duty is framed in terms of subjective intentions, and “any plausible assertion that a course of action is designed to increase the company’s financial well-being will be enough to protect it from attack” (Parkinson, 1993, p.96). Daily and Dalton (1994) provided evidence that non-executive directors would tend to prevent performance from dropping to the level that causes bankruptcy. There is a large literature around the topic that directors do not always perform their legal functions (e.g., Juran and Louden, 1966; Koontz, 1967; Mace, 1971; Bacon, 1973; Louden, 1982; Epstein, 1986; Patton and Baker, 1987; Lorsch and MacIver, 1989; Zahra, 1990; Pearce II and Zahra, 1991). In theory directors hire executives to run the corporation, in practice the reverse occurs; executives typically nominate directors (Epstein, 1986). Directors frequently fail to perform their legal mandate because they are “creatures of the CEO” (Patton and Baker, 1987) or simply “rubber stampers” (Zahra, 1990). Patton and Baker (1987) interviewed a number of lawyers, bankers, accountants, and recently-retired chief executives about the practice of CEO duality; many felt that the joint authority would reduce the check and balance allowed by the board. As the chairman is also the CEO, he or she is the chief protector of shareholders as well as chief manager; conflict of interests may occur. When CEOs dominate boards, power is vested in these executives; they could select, reward or replace directors (Zahra, 1990). Since directors are principally selected and retained by the CEO, the board may pay little attention to its control function; board power is thus limited to the ceremonial approval of managerial decisions, which falsely gives a form of legitimacy to the managers’ decisions (Pearce II and Zahra, 1991). When companies do not perform well, boards are blamed for not controlling the managers’ performance (Loevinger, 1986; Fleischer, Hazard and Klipper, 1988). 73 Rosenstein (1987) indicated that directors might not be equipped to handle changing organizational complexity. As the CEOs are the most important persons in companies, they may not want a capable board that could challenge their power and authority; therefore directors tend to be obtrusive and they may delegate the control authority to the CEO (Rosenstein, 1987). Vance (1978, 1983) reported that delegation of power was most visible among the small, new, low-technology, and closely-held firms where the founders were still working in the firm. As noted by Stiles and Taylor (2001), consideration of the legal duties of directors takes us only a small way towards understanding the work of directors and boards, for at least two main reasons. First, though boards may have de jure control of the organization, the de facto control may rest with management. Secondly, the legal notion of boards has also been attacked because it pays insufficient attention to the interests of stakeholders other than shareholders. 4.2.2 Agency Theory Denis and McConnell (2003) found that the publication of Jensen and Meckling (1976), in which the authors applied agency theory to corporations and modelled the agency costs of outside equity, spawned a voluminous body of research on corporate governance in general, and boards of directors in particular, around the world. According to Jensen and Meckling (1976), an agency relationship is a contract under which one party (the principal) engages another party (the agent) to perform some services on the principal’s behalf. Under the contract, the principal delegates some decision-making authority to the agent; in the situation, where both the principal and the agent are utility maximisers; there is no a priori reason to believe that the agent will always act in the principal’s best interests. The agency problem that arises is the problem of inducing an agent to behave as if he or she were maximising the principal’s welfare; the agency problem is also characterized by asymmetric information, as the principal has more restricted information set than the agent. This problem, in turn, gives rise to agency costs. At the most general level, agency costs could be measured as the dollar equivalent of the reduction in welfare experienced by the principal due to the divergence of the principal’s and the agent’s 74 interests. Jensen and Meckling (1976) further divided agency costs into monitoring costs, bonding costs and residual loss. In investor-owned firms, shareholders act as the principals with interests in deriving maximum utility from the action of managers, serving as the agents. Under debt contracts, the principals are debtholders, or lenders; the agents are managers acting on behalf of shareholders (Godfrey, Hodgson and Holmes, 2003). An early study by Berle and Means (1932) shows a divergence of interests between investors and managers. Investors have three interests; the first is that the company should be able to earn the maximum profits under an acceptable risk; the second is that they want as large a proportion of profits to be distributed to them as possible, and third, the company’s stock should remain marketable at a fair price. Managers have only one major aim - to run the company for their “personal profits”. Williamson (1975) termed the phenomenon as “opportunism” whereby people act with self-interest and guile in pursuing their own goals. To reduce agency costs, firms should align the interests of managers with those of shareholders; suggested measures include: the separation of CEO and chairman because the CEO cannot both represent the shareholders and management due to conflict of interests (Rechner and Dalton, 1991); equity ownership by firm‘s management to tie the management’s compensation to firm performance (Jensen and Meckling, 1976); strengthening the governance structure of organizations whereby boards of directors keep potentially self-serving managers in check by performing audits and performance evaluations (Fama and Jensen, 1983); managerial labour market in which a poor performing manager limits his or her career opportunities (Fama, 1980); market for corporate control in which a poor performing firm risks being acquired by another and the consequence may be the replacement of management of the acquired firm (Grossman and Hart, 1980); and inclusion of at least some outside directors to monitor the performance of the CEO and other managers (Baysinger and Hoskisson, 1990). 75 Agency theory perceives the board’s effectiveness in monitoring management to be crucial (Stroh, Brett, Baumann and Reilly, 1996). Corporations could benefit from the external oversight a board can offer by shielding the invested stakes of equity holders as well as debtholders, from potential managerial self-interest, as well as from the risk that the CEO may mix personal and business goals (Castaldi and Wortmann, 1984; Daily, Johnson and Dalton, 2002). Bacon and Brown (1975) suggested that the monitoring efforts would help to reduce agency costs and ensure compliance of managers to focus on established procedures and goals. Board of directors can function as an important information system for external stakeholders to monitor firm performance and reduce asymmetric information (Bacon and Brown, 1975). If monitoring fails, the more expensive external measures, such as acquisitions, divestitures and ownership amendments, would arise. As external controls could be detrimental to the principals, monitoring is generally preferred (Walsh and Steward, 1990). Thus from an agency perspective, directors are valued for their ability to be independent when overseeing operating matters, protecting the assets of the firm, and holding managers accountable to the firm’s stakeholders to ensure the success of the enterprise. It is asserted that monitoring practices that align shareholders’ and managers’ interests and prevent managers from pursuing self-serving objectives would be positively associated with firm performance; the results would be the maximization of company’s profitability and shareholders’ wealth (Fama, 1980; Scott, 1983). Consequently, proponents of agency theory expect that where board of directors is more independent of management, company performance would be higher. Regarding the difference between agency theory and the legalistic approach, it is indicated that legalistic theory views directors’ power coming from state law where agency theory indicates that directors’ power arise from shareholders; while control is the dominant function of directors under legalistic theory, monitoring is its counterpart under agency theory (Ong and Lee, 2000). Although agency theory has received widespread acceptance among researchers, it is argued in some papers that the fundamental assumption of the theory that all action is driven by a desire to maximise wealth is too simplistic and rather negative, which does not allow other motivations for managerial behaviours to be examined (e.g., Tinker, 76 Merino and Niemark, 1982; Christenson, 1983; Williams, 1989; Sterling, 1990; Chambers, 1993; Gray, Owen and Adams, 1996; Howieson, 1996). Jensen and Meckling (1994) criticized the theory as being a simplification for mathematical modelling and an unrealistic description of human behaviour. Doucouliagos (1994) commented that labelling all motivation as self-serving could not explain the complexity of human action. Frank (1994) suggested that this model of man would not suit the demand of a social existence; man takes care of both his personal and social needs at the same time, and self-interest can be sacrificed for the sake of organization. In reality it is possible that as managers contribute their human capital to the organization, they could be equally, if not more, concerned with the success of the company as shareholders (Frank, 1994). 4.2.3 Stewardship Theory The stewardship model has been developed as an alternative to agency theory (Donaldson and Davis, 1991, 1994; Donaldson, 1990; Fox and Hamilton, 1994; Davis, Schoorman and Donaldson, 1997). Agency model, rooted in the field of economics and finance, examines the structures of capitalism, finds only self-interested behaviour and assumes “this is human nature … and neglects the enormous amount of neutral and other-regarding behavior that exist … and the structures that might increase it” (Perrow, 1986, P.234). According to Donaldson (1990), the assumptions of agency theory are extreme, which fail to recognize a range of non-financial motives for managerial behaviours, including the need for achievement and recognition, the intrinsic satisfaction of successful performance, respect for authority and work ethic. These concepts are well supported in the organizational literature (McClelland, 1961; Argyris, 1964; Herzberg, 1966; Etzioni, 1975). Donaldson and Davis (1991) found that managers could be viewed as interested in achieving high performance and capable of using a high level of discretion to act for the benefit of shareholders; they are good stewards of corporate assets, loyal to the company, when confronted with a course of action seen as personally unrewarding, may comply based on a sense of duty and identification with the organization. There is also 77 evidence that close supervision often causes people to shift their focus from performing their roles to actively resisting attempts to control them (Argyris, 1964). Similarly, it is reported that managers whose behaviours are pro-organizational and collectivistic have higher utility than individualistic, self-serving behaviours (Donaldson and Davis, 1991, 1994; Fox and Hamilton, 1994; Davis, Schoorman and Donaldson, 1997). Where the interests of managers and principals are not aligned, managers may place higher value on co-operation than defection. Since managers perceive greater utility in co-operative behaviours and behave accordingly, their behaviours could be considered rational (Fox and Hamilton, 1994). The belief that managers have a wide range of motives beyond self-interest is the rationale for arguing that goal conflict may not be inherent in the separation of ownership from control. Davis et al (1997) claimed that the reallocation of corporate control from owners to professional managers might be a positive development toward managing the complexity of modern corporations. Consequently, insider dominated boards are favoured by stewardship theorists for their depth of knowledge, access to current operating information, technical expertise and commitment to the firm (Muth and Donaldson, 1998; Kiel and Nicholson, 2003; Anderson and Reeb, 2004; Dulewicz and Herbert, 2004; Randoy and Jenssen, 2004). Boards should play a greater role in facilitating and empowering managers instead of monitoring and control; managers would protect and maximize shareholders’ wealth through firm performance because, by doing so, their utility functions would be maximized (Donaldson and Davis, 1991, 1994; Fox and Hamilton, 1994). Boards should play an indirect function, and empower the managers who would directly be responsible for the performance of their companies. For CEOs who are stewards, their pro-organizational actions may be best facilitated when the corporate governance structures give them high authority and discretion (Donaldson and Davis, 1991); structurally, this situation is attained more readily if the CEO is also chair of the board of director. The CEO-chair who is unambiguously responsible for the fate of the corporation would have the power to determine strategy without fear of countermand by an outside chair of the board. 78 4.2.4 Resource and Strategy Theories From the 1980s, three theories founded on the organizational literature have been increasingly used by academics in investigating corporate governance issues, i.e., resource based theory, resource dependence theory and strategic choice theory. The resource based approach generally argues that a firm's internal environment, in terms of its resources and capabilities, is critical for creating sustainable competitive advantage (Prahalad and Hamel, 1990; Teece, Pisano and Shuen, 1997). Being aware of, improving, and protecting the unique resources of the firm would then reinforce firm strengths and rearrange firm weaknesses, and thereby improve a firm’s competitive position. However, firms are generally characterized by a lack of internal resources, and in-house knowledge may in many cases be scarce or non-existing (Storey, 1994). It has in this respect been considered important to have a board with outside members in order to overcome this internal lack of resources and complement the management with experience, knowledge and skills (Castaldi and Wortmann, 1984). The board of directors, and especially outside directors, may be considered as a bundle of strategic resources to be used by and within the firm as they can provide advice and counsel to the CEO and the management in areas where in-firm knowledge is limited or lacking. The resource-based view consequently recognizes that outside directors can be a valuable source of competitive advantage through their professional and personal qualifications. From a different point of view, resource dependence theory notes that the long-term survival and success of a firm is dependent on its abilities to link the firm with its external environment (Pfeffer, 1972; Pfeffer and Salancik, 1978). A basic argument in this theory is that firms must constantly interact with its external environments either to purchase resources, or to distribute its finished products. Firms should seek to gain control over its environment to create more stable flows of resources and lessen the effects of environmental uncertainty. Directors, as boundary spanners, could form links with the external environment, which may be useful for the managers in the achievement of the various goals of the organization (Pfeffer and Salancik, 1978; Zahra and Pearce, 1989; Pearce and Zahra, 1992; Borch and Huse, 1993). 79 Specifically, outside directors may help firms initiate and maintain control over critical relationships, assets and contacts in the external environment. The firm may also co-opt representatives from important organizations as board members in order to achieve organizational goals and manage environmental contingencies. Directors who are prestigious in their professions and communities can be a source of timely information for executives. They become involved in supporting the organization by influencing their other constituencies on behalf of the focal organization (Pfeffer, 1972; Pfeffer and Salancik, 1978). Pfeffer and Salancik (1978, p.163) found that “[w]hen an organization appoints an individual to a board, it expects the individual will come to support the organization, will concern himself with its problems, will favorably present it to others, and will try to aid it”. According to the strategic choice theory, strategy is the primary link between organization and its environment, and boards should actively participate in the process of strategy formulation and implementation, because: directors, as boundary spanners, may have the necessary resources to collect information about competitiveness and industry changes, which could be important for strategic actions; directors who are managers in other companies would be in a advantageous position to provide information in strategic decision-making process; increasing shareholders’ expectations has led to directors paying more attention to strategic activities; the strategy formulation process is generally complicated, directors are urged to join in to provide inputs; and as the results of the complex competitive environment, boards should offer CEO guidance in strategic actions (Zahra, 1990). Proponents of this model recognize that strategic management is under the direct responsibility of top management, as they are ultimately responsible for integrating the functional and divisional areas of the firm, and balancing the short, medium and longterm planning and control cycles for the firm. It is suggested that CEOs and boards must understand the possible areas of conflict between them, and allow for mutual collaboration. These conflicts should be clearly stated and agreed upon (Kreiken, 1985; 80 Zahra, 1990). Directors could add value to the process of strategy formulation and implementation by: making strategy a regular topic on the agenda; knowing the origins of the plans; taking a broad analysis of environmental factors; considering strategic alternatives and three theories founded on the organizational literature; and checking on progress and adaptation of strategic plans and actions (Kreiken, 1985). Companies with better strategies would achieve higher levels of performance (Hambrick and Mason, 1984). With board involvement, CEOs and managers could be more effective in strategic management (Andrews, 1986). From the above introduction, it appears that all the three theories agree that support is the most important function of boards of directors. The resource based approach notes that the board of directors could support the management in areas where in-firm knowledge is limited or lacking. The resource dependence model suggests that the board of directors could be used as a mechanism to form links with the external environment in order to support the management in the achievement of organizational goals. The strategic choice perspective recommends that the board of directors should support the management by actively participating in the process of strategic formulation and implementation. However, the scholars in these schools do not provide any opinion on board independence and its relationship with firm performance, although outside directors are valued in their models with different rationales. Outside directors comprise independent directors and affiliated directors, or “grey directors” as termed by Baysinger and Bulter (1985). The above models are more interested in affiliated directors and their experience, knowledge and linkages with other organizations (Pfeffer and Salancik, 1978; Castaldi and Wortmann, 1984; Zahra, 1990). 4.2.5 Organizational Portfolio Theory Heslin and Donaldson (1999) and Donaldson (2000) proposed another theory of organizational change and success, namely, organizational portfolio theory; however, 81 unlike resource and strategy theories, it views the central function of boards of directors as a mechanism to adjust firm performance variance, i.e., firm risk. Since firms typically need to experience a crisis of poor performance before they make required structural changes (Chandler, 1962; Williamson, 1971; Child, 1972; Donaldson, 1987; Hamilton and Shergill, 1992), it is postulated that a combination of portfolio factors which causes poor performance is the usual trigger of organizational adaptations (Heslin and Donaldson, 1999; Donaldson, 2000). According to Heslin and Donaldson (1999, p.81), “[o]ne purpose of this theory is to provide a fresh perspective on the determinants and consequences of board composition”. The authors suggested that instances of poor firm performance led to executive directors on the board being replaced by more independent non-executive directors. Such changes decrease the amount of trust and discretion granted to executives, based on the assumption that the decline in profitability is a result of their poor management of the firm. In turn, an increase in organizational profitability enhances the perceived integrity and competence of managers, thereby precipitating boards in which managers are increasingly represented. In Heslin and Donaldson (1999), three factors are identified that are likely to prevent instances of poor performance and so forestall calls for a tougher and more independent board. Diversification generally lowers corporate risk as unrelated product-lines tend to be on different business cycles and thus exhibit negative correlated performance; the high performances of some product-lines often occur simultaneously with the poor performances of other product-lines. The overall effect would be moderate risk so that firm performance remains above the level at which board change may occur. Divisionalization follows diversification and further reduces risk; divisional performances are decoupled by the discretion of divisional managers to make autonomous decisions and divisional managers’ incentives to smooth the fluctuations in performance they report, thereby diminishing firm-level risk. Finally, divestments are often undertaken during periods of low performance and result in an injection of cash flow which forestalls the immediate likelihood of poor performance. 82 There are also three factors that could contribute to poor performance and lead to the appointment of more non-executives as board members or chair (Heslin and Donaldson, 1999). Depressed business cycles have a detrimental influence on company profitability. Competition also has an adverse impact on profitability, particularly if competitors are well organized. Debt raises the level of profitability that is regarded as satisfactory as it requires large, regular interest payments, thus increases the chance of sufficiently poor performance occurring to diminish confidence in management Based on the evidence provided by Baysinger, Kosnik and Turk (1991) and Hill and Snell (1998), Heslin and Donaldson (1999) assumed that, in general, executive directors would raise risk and non-executives would reduce risk. Executives are more likely to approve risky proposals such as increasing expenditures on R & D (Baysinger et al, 1991). This probably reflects their intimate knowledge of the business and resulting confidence that anticipated benefits will flow from their proposed investments (Lorsch and MacIver, 1989). In contrast, independent directors may have less first-hand familiarity with the business; as they are often appointed in order to curb the ostensibly radical excesses of management, they would tend to be risk-averse (Heslin and Donaldson, 1999). The pressure of strong public criticism and the threat of legal action for failure in their fiduciary responsibility for poor corporate performance may reinforce the risk-aversion of independent directors and chairpersons (Davis and Thompson, 1994). Therefore, according to Heslin and Donaldson (1999), during periods when executives constitute a large proportion of the board, risk tends to increase. When peaks from the high risk strategy co-occur with favourable combinations of the other portfolio factors, outstanding performance is likely to result. This would reinforce confidence in the integrity and competence of the largely non-independent corporate governance structure, thus bolstering the position of executives on the board. When the troughs in the high risk strategy occur simultaneously with other performance-depressing portfolio factors, the particularly low firm performance may trigger the installation of an independent chair and a higher proportion of independent directors on the board; the resulting risk-averse governance would tend to reduce firm performance variance (Heslin and Donaldson, 1999). 83 It is considered that reducing firm risk may be a means of increasing short-term economic value (Brealey and Myers, 1996). However, Heslin and Donaldson (1999) and Donaldson (2000) indicated that low risk could prevent the performance crises needed to trigger required structural adaptation; high economic value achieved by lowering risk is thereby prone to inhibiting long-term growth and profitability. By reducing risk, independent directors would not only prevent long-term success, but indeed foster a gradual decline in organizational performance. Thus shareholders interested in the medium to long-term profitability of a firm should resist joining demands for board independence. “Prematurely instituting independent directors to closely monitor and discipline executive management may unnecessarily forfeit the advantages of non-independent boards. Paramount among these is a willingness to take the risks necessary to precipitate both instances of outstanding shortterm performance and the structural changes needed for long-term organizational growth and prosperity” (Heslin and Donaldson, 1999, p.86). Although organizational portfolio theory is built on the empirically supported premise that low performance is required to trigger adaptive organizational changes, it is acknowledged that “… the theory is at present an extended conjecture following from that premise … It is a series of propositions waiting for empirical testing. Only after it has received such empirical confirmation would the policy implications sketched here become valid prescriptions” (Donaldson, 2000, p.395). 4.3 Board Independence and Firm Performance As discussed in the last section, although outside directors are valued in legalistic theory, resource theories and strategy choice theory with different rationales, these models do not make any explicit predictions regarding the relationship between board independence and firm performance. Three theories, namely, agency theory, stewardship theory and organizational portfolio theory, provide some guidelines on the potential correlation between board independence and corporate performance. As noted by some authors, theoretical support for the conventional wisdom that independent boards would enhance shareholder returns has been provided by agency theory (e.g., Muth and Donaldson, 1998; Dalton et al, 1998; Vafeas and Theodorou, 1998; Cotter and Silverster, 2003). A central assumption of the theory is that managers 84 may pursue their own goals rather than seek to maximise shareholder wealth, unless their discretion is kept in check by a vigilant, independent board (Castaldi and Wortmann, 1984; Daily et al, 2002). By emphasising the potential for divergence of interests between investors and managers, agency theory predicts that where board of directors is more independent of management, company performance would be higher (Fama, 1980; Scott, 1983). In contrast, stewardship theory highlights a range of non-financial motives for managerial behaviours, such as the need for achievement, the intrinsic satisfaction of successful performance, and respect for authority and the work ethics (Donaldson and Davis, 1991, 1994; Donaldson, 1990; Fox and Hamilton, 1994; Davis et al, 1997). Having control empowers managers to maximize corporate profits; the detailed operation knowledge, expertise and commitment to the firm by executive directors, would make firms with a management-dominated board more profitable (Donaldson and Davis, 1991, 1994; Fox and Hamilton, 1994; Davis et al, 1997). Organizational portfolio theory asserts that firm performance determined by a combination of portfolio factors drives organizational changes (Heslin and Donaldson, 1999; Donaldson, 2000). An increase in corporate profitability would enhance the perceived integrity and competence of managers, thereby precipitating boards in which managers are increasingly represented; poor performance would lead to boards that are more independent of management. During periods when executives constitute a large proportion of the board, risk tends to increase; when peaks from the high risk strategy co-occur with favourable combinations of the other portfolio factors, outstanding performance is likely to result. On the other hand, the risk-averse governance delivered by independent directors would inhibit longterm growth and profitability (Heslin and Donaldson, 1999). In Chapter 3, some papers which provide empirical evidence on the relations between board characteristics and the “bottom line” performance of public listed companies, in Australia and overseas, are examined. Appendix 3 exhibits a summary of their conceptual frameworks, hypotheses and conclusions, which shows that the agency model is the most frequently used theoretical framework in the literature (Baysinger and Butler, 1985; Schellenger et al, 1989; Hermalin and Weisbach, 1991; Agrawal and Knoeber, 1996; Rosenstein and Wyatt, 1997; Muth and Donaldson, 1998; Vafeas and 85 Theodorou, 1998; Lawrence and Stapledon, 1999; Coles et al, 2001; Dehaene et al, 2001; Hossain et al, 2001; Cotter and Silverster, 2003; Kiel and Nicholson, 2003; Singh and Davidson III, 2003; Panasian et al, 2003; Bonn et al, 2004; Balatbat et al, 2004; Anderson and Reeb, 2004; Peng, 2004; Dulewicz and Herbert, 2004; Randoy and Jenssen, 2004; Chin et al, 2004; Chang and Leng, 2004; Chen et al, 2005; Krivogorsky, 2006; Luan and Tang, 2007; Choi et al, 2007; Chan and Li, 2008). The prediction proposed by stewardship theory, with respect to the relationship between board independence and firm performance, is also tested in a number of papers (Muth and Donaldson, 1998; Dalton et al, 1998; Coles et al, 2001; Kiel and Nicholson, 2003; Anderson and Reeb, 2004; Dulewicz and Herbert, 2004; Randoy and Jenssen, 2004; Luan and Tang, 2007). Several researchers explored the concern whether board characteristics are endogenously related to performance; however, there is little theoretical support in their studies (Hermalin and Weisbach, 1988; Pearce II and Zahra, 1992; Denis and Sarin, 1999; Bhagat and Black, 2000). 4.4 Testable Hypotheses Based on the preceding discussion, the potential relations between board independence and firm performance as suggested by agency theory, stewardship theory and organizational portfolio theory could be illustrated as follows: Figure 4.1 Theoretical Development: High Board Independence Agency Theory Low Firm Performance High Board Independence High Firm Performance Stewardship Theory & Organizational Portfolio Theory Low Firm Performance Organizational Portfolio Theory 86 Agency theory suggests that where board of directors is more independent of management, company performance would be higher. As an alternative to agency theory, stewardship model predicts that high board independence may lead to low performance. Organizational portfolio theory proposes that poor performance may trigger the installation of an independent chair and a higher proportion of independent directors on the board; however, the resulting risk-averse governance would foster a gradual decline in performance. Figure 4.2 Theoretical Development: Low Board Independence High Firm Performance Stewardship Theory & Organizational Portfolio Theory High Firm Performance Low Board Independence Agency Theory Low Firm Performance Organizational Portfolio Theory Agency theory indicates that low board independence would lead to low performance. Stewardship theory expects that, as shareholders would maximize their returns if the organization structure facilitates effective control by management, board of directors with a lower level of independence may lead to higher company performance. According to organizational portfolio theory, an increase in organizational profitability would enhance the perceived integrity and competence of managers, thereby precipitating boards in which managers are increasingly represented. During periods when executives constitute a large proportion of the board, risk tends to increase; when the high risk strategy co-occur with favourable combinations of the other portfolio factors, outstanding performance is likely to result. Therefore, with respect of the relation between board independence and firm performance, the following research hypotheses are developed: 87 H1: There is a negative relationship between board independence and past firm performance (organizational portfolio theory); H2 : There is a negative relationship between board independence and subsequent firm performance (stewardship theory, organizational portfolio theory); and H3: There is positive relationship between board independence and subsequent firm performance (agency theory). Stewardship theory and organizational portfolio theory make the same prediction about the relation between board independence and subsequent firm performance. To differentiate between these two theories, a further hypothesis which could be tested is: H4 : There is a negative relationship between board independence and subsequent firm risk (organizational portfolio theory). As introduced earlier, supported by some empirical evidence (Lorsch and MacIver, 1989; Baysinger et al, 1991; Davis and Thompson, 1994; Hill and Snell, 1998), Heslin and Donaldson (1999) argued that, in general, executive directors would raise risk and non-executives would reduce risk. Thus, according to organizational portfolio theory, there may be a negative relationship between board independence and firm risk. However, some agency theorists claim that managers, unlike shareholders, could not readily diversify their employment risks across a range of investments, as a result they tend to be more risk averse than may be in the interests of shareholders (e.g., Fama, 1980; Knoeber, 1986; Prentice, 1993; Vafeas and Theodorou, 1998; Coles et al, 2001; Godfrey et al, 2003). From the point of view of agency theory: H5: There is a positive relationship between board independence and subsequent firm risk (agency theory). 4.5 Summary The description of what a board of directors is for varies depending on which theoretical approach scholars take. According to the legalistic view, boards are legally required to act in the best interests of shareholders and control is the dominant function of directors. Agency theory perceives the board’s effectiveness in monitoring management to be crucial; firms could benefit from the external oversight an independent board can offer 88 by shielding the invested stakes of equity holders as well as debtholders, from potential managerial self-interest. Stewardship model shows that managers have a wide range of motives beyond self-interest, and boards should play a greater role in facilitating and empowering managers instead of monitoring and control. There are three organizational models indicating that support is the most important function of directors. The resource based approach notes that board of directors could support the management in areas where in-firm knowledge is limited or lacking. The resource dependency model suggests that board of directors could be used as a mechanism to form links with the external environment in order to support the management in the achievement of organizational goals. According to the strategic choice perspective, board of directors should support the management by actively participating in the process of strategic formulation and implementation. Organizational portfolio theory, which is built on the premise that low performance is required to trigger adaptive organizational changes, views the central function of boards of directors as a mechanism to adjust firm risk. It is proposed that poor firm performance would lead to executive directors on the board being replaced by independent directors; however, by reducing risk, independent directors may indeed foster a gradual decline in organizational performance. Regarding the correlation between board independence and corporate performance, agency theory, stewardship theory and organizational portfolio theory offer different expectations. From an agency theory perspective, where the board of directors is more independent of management, company performance would be higher. According to stewardship theory, shareholders would maximize their returns if the corporate structure facilitates effective control by management, thus board of directors with a lower level of independence may lead to higher company performance. As a new theory waiting for empirical testing, organizational portfolio model suggests that board independence may be negatively associated with past and subsequent performance. 89 Chapter 5. Research Method 5.1 Introduction As shown in earlier chapters, the issues of board composition and structure generally and independent directors specifically, have become a fertile area of interest and research. This study intends to: test the applicability of several theories which make different predictions about the effect of board independence on firm performance and vice versa; address some of the limitations of prior studies, including small sample size, short-term observation of firm performance, limited control variables and performance measures, and simplistic dichotomy of inside and outside directors as an empirical proxy for board independence; and shed some light on the potential influence of a recently altered regulatory environment with respect to corporate governance mechanisms, particularly those relating to the requirements for a majority of independent directors on the board and board committees, on firm performance. The focus, as indicated in Chapter 3, is on the empirical correlation between board independence and the “bottom line” of firm performance, rather than the relationship between board characteristics and certain corporate events, or the relationship between the performance of boards of directors or individual directors and firm performance. There are two broad research questions to be investigated in this project. Does board independence have any influence on firm performance among Australian listed companies? Does firm performance have any influence on board independence among Australian listed companies? According to Gill and Johnson (2002), the use of a particular methodology for a research project depends on the scope, purpose and target population of the study as well as the resources available to the researcher. For researchers to achieve their objectives, they need to adopt the right methodology and select the right data collection techniques through which they could obtain appropriate data within their available resources. 90 This chapter provides a discussion of the methods used in this research. In Section 2, an introduction of the general research approaches and justification of the approach chosen in this study are provided, which is followed by a description of the sample and data sources in Section 3. Section 4 defines the research variables tested in the project, including measures of board independence and financial performance, and controls. The analysis tools selected are then discussed in Section 5. The last section presents a summary of this chapter. 5.2 Research Approach There are two general research approaches widely recognized in social science, i.e., the quantitative and qualitative approaches. The goal of the quantitative approach is to determine whether the predictive generalizations of a theory hold true (Ryan, Scapens and Theobald, 2002; Veal, 2005; Frankfort-Nachmias and Leon-Guerrero, 2006). As noted by Frankfort-Nachmias and Leon-Guerrero (2006), the quantitative approach is deductive, researchers in this school deal directly with the operationalisation, manipulation, prediction, and testing of empirical variables; therefore they place great emphasis on methodology, procedures and statistical measures of validity. It is suggested that quantitative research should be organised to show a clear progression from theory to operationalisation of concepts, from choice of methodology and procedures to data collection, and from statistical tests to findings and ultimately conclusions (Ryan et al, 2002; Veal, 2005; Frankfort-Nachmias and Leon-Guerrero, 2006). Smith (2003) and Veal (2005) identified four types of quantitative methods. Experimental method: this method is characterized by random assignment of subjects to experimental conditions and use of experimental controls; Quasi-experimental method: this method shares almost all the features of experimental design except that it involves non-randomized assignment of subjects to experimental conditions; Survey method: this method uses questionnaires or interviews for data collection with the intent of estimating the characteristics of a large population of interest based on a smaller sample from that population; and Archival method: archival research is based on historical (secondary) data and uses cross-sectional and/or time-serial data to investigate a problem. 91 According to Hussey and Hussey (1997), in a quantitative study, data are collected in numerical form; one of the advantages of this approach is the relative ease and speed with which the research can be conducted. In contrast, the qualitative approach is inductive in nature. Qualitative researchers use field research methods, primary case studies and participant observation within natural settings; their reports present much more descriptive materials and show how the observations prompt the researchers to analyse and isolate variables (induction) and how, in turn, these variables may be developed into a theoretical framework (FrankfortNachmias and Leon-Guerrero, 2006). Mason (1996) found that there were three types of qualitative studies, namely, case study, ethnographic study and phenomenological study. Case study: the researchers explore a single entity or phenomenon bounded by time and activity and collect detailed information through a variety of data collection procedures over a sustained period of time; Ethnographic study: the researchers investigate an intact cultural group in a natural setting over a specific period of time; and Phenomenological study: the researchers examine human experiences through a detailed description of the people being studied. Thus it appears that, in most cases, qualitative data are concerned with qualities and non-numerical characteristics. Hussey and Hussey (1997) commented that qualitative research might present problems relating to rigour and subjectivity. This study uses an archival research design which is traditionally employed by the literature surrounding this topic. It follows such a design because one of its purposes is to test the predictive generalizations of several theories with respect to the relationship between board independence and firm performance. Moreover, the nature of the data required to conduct this research on ASX-listed companies indicates that such data would be publicly available; it is a time and cost efficient decision to use secondary data. Consequently, this project is organized to show a clear progression from theories to hypotheses, from choice of methodology and procedures to data collection, and from statistical tests to findings and conclusions. As noted by Novak and Gowin (1984), such a consistency would ensure that the knowledge 92 claims made as a result of the research program are logically linked to the philosophical base of the theories and concepts used. 5.3 Sample and Data Collection For any empirical study, it is essential to clearly define the population of interest, and to ensure that the sample selected provides an accurate representation of that population (Thomas, 1996; Weisberg, Krosnick and Bowen, 1996; Ryan et al, 2002; Smith, 2003; Veal, 2005). In Chapter 3, it is revealed that most Australian studies and some overseas papers in this field suffer from the limitation of small sample size. Muth and Donaldson (1998), for example, suggested that a sample size closer to 200 would have been preferable; effects of boards on performance tend to be small which means that more statistical power is needed to detect significant relationships. Calleja (1999) also acknowledged that the small sample size in her study made it difficult to reach any firm conclusions. This project uses the top 500 companies listed on the ASX in 2003, ranked by market capitalisation, as the initial dataset. Each year the ASX collects information on these companies to calculate its All Ordinaries Index, which is the primary indicator of the Australian equity market. At December 31, 2003, the top 500 companies represent 95% of the total market capitalisation of the ASX-listed companies (S&P, 2004). Thus, this dataset offers a reasonable coverage for the population of interest, i.e., Australian publicly listed corporations. The top 500 companies change over time; it is necessary to select a specific year to determine the list of the firms which presents the initial sample for this research. 2003, which is the midpoint of the period for which the data is collected and tested (20002006), is chosen to reduce the chance of missing data due to delisting or suspension of sample firms. The data used in this study was collected in October 2006. Due to lack of comparable performance data in the financial institutions section, Muth and Donaldson (1998) had to reduce their sample of top Australian firms. Kiel and Nicholson (2003) also removed banks from their analysis because the recorded assets of financial institutions consist of loans which represent the use of depositors’ funds. Calleja (1999) and Cotter and Silverster (2003, p.217) noted that “[t]rusts have unique characteristics which impact on their corporate governance practices. The trust manager 93 and the trustee are jointly responsible for governance matters but have a fundamental separation of responsibilities and powers between them”; consequently trusts were excluded from their samples of large ASX-listed companies. There are 503 firms in the 2003 list of top 500 companies provided by Huntleys’ Shareholder (2003). Following the local studies as shown above, financial institutions including property trusts and investment funds are removed from this list; as a result, an initial sample of 384 companies listed on the ASX is obtained. The sources of data required to construct the research variables, which are described in the next section, are available within the public domain. Data on firm performance and risk, dividend payout, firm size and financial leverage is collected from Fin Analysis database. The data on board independence and size, and blockholder and managerial ownership is developed from the corporate reports provided by Connect 4 database. The information on diversification and firm age is obtained from Huntleys’ Shareholder (2003). The sample is further reduced to 243 firms after the following are excluded: 62 companies which had been delisted or suspended from the ASX during the three years between 2003 and 2006; 32 companies without complete market performance data on Fin Analysis for the periods of 2000-2003; most of them have been listed on the ASX after 2000; 12 companies whose annual reports could not be obtained from Connect 4, or the company’s website if there is one; and 35 companies without sufficient information on board composition and structure, and/or managerial ownership in their 2003 reports. 5.4 Research Variables In order to examine the relationship between board independence and firm performance, the level of board independence, firm performance and some control variables used in the research have to be measured. 5.4.1 Measurement of Board Independence As shown in Appendix 1 and 2 listed at the end of this thesis, the most popular measure of board independence in prior research is the proportion of outside or independent directors on the board (e.g., Baysinger and Butler, 1985; Fosberg, 1989; Schellenger et 94 al, 1989; Hermalin and Weisbach, 1991; Pearce II and Zahra, 1992; Yermack, 1996; Agrawal and Knoeber, 1996; Barnhart and Rosenstein, 1998; Calleja, 1999; Lawrence and Stapledon, 1999; Denis and Sarin, 1999; Coles et al, 2001; Singh and Davidson III, 2003; Bonn et al, 2004; Anderson and Reeb, 2004; Peng, 2004; Randoy and Jenssen, 2004; Chin et al, 2004; Krivogorsky, 2006; Choi et al, 2007). The definitions of an outside or independent director embodied in these studies may be slightly different. Some authors, for example, Baysinger and Bulter (1985), Lawrence and Stapledon (1999), Kiel and Nicholson (2003), Peng (2004) and Luan and Tang (2007) argued that a limitation of some previous studies was that NEDs in the samples were not classified as independent or affiliated. The combination of insider and “grey” directors may constitute a majority of the board, and this could lead to companies appearing to comply with the recommendations for board independence through comprising a non-executive majority on the board, whilst in fact being controlled by internal management (Clifford and Evans, 1997). From the recommendations and listing rules as outlined in Chapter 2, it appears that, in the U.K., U.S. and Australia, the stock exchanges encourage or mandate a shift in the overall power of boards of directors in favour of independent directors, and away from executive management. For companies listed on the ASX, it is recommended in the Guidelines (2003) that: a majority of each company’s directors should qualify as independent directors; each company should establish audit, nomination and remuneration committees dominated by independent directors; and the roles of chairman and CEO should not be exercised by the same individual, and the chairman should be an independent director. Thus, in this study five empirical proxies for board independence are adopted, i.e., full board independence represented by the proportion of independent directors on the board, monitoring committee independence measured by the percentages of independent directors on the audit, nomination and remuneration committees, and chairman independence which is a binary variable to assess whether or not the chairman is an independent director. 95 If the sources of information only divide directors into executive directors and NEDs, it would be necessary to divide NEDs into independent directors and affiliated directors, using the definition of independence proposed by the ASX Corporate Governance Council as a benchmark. According to the Guidelines (2003, p.19), “[a]n independent director is independent of management and free of any business or other relationship that could materially interfere with – or could reasonably be perceived to materially interfere with – the exercise of their unfettered and independent judgement”. It is further defined in Box 2.1 of the Guidelines (2003) that an independent director is a NED and is not a substantial shareholder of the company or an officer of, or otherwise associated directly with, a substantial shareholder of the company; within the last three years has not been employed in an executive capacity by the company or another group member, or been a director after ceasing to hold any such employment; within the last three years has not been a principal of a material professional adviser or a material consultant to the company or another group member, or an employee materially associated with the service provided; is not a material supplier or customer of the company or other group member, or an officer of otherwise associated directly or indirectly with a material supplier or customer; has no material contractual relationship with the company or another group member other than as a director of the company; has not served on the board for a period which could, or could reasonably be perceived to, materially interfere with the director’s ability to act in the best interests of the company; and is free from any interest and any business or other relationship which could, or could reasonably be perceived to, materially interfere with the director’s ability to act in the best interests of the company. The Guidelines (2003) suggests that family ties and cross-directorships may be relevant in considering interests and relationships which may compromise independence; however, it is unclear how long an independent director could serve on the same board. This research follows the U.K. Higgs Report (2003) which nominates ten years in relation to director tenure consideration. 96 Following the approach supported by prior studies (e.g., Pfeffer, 1972; Kesner, 1987; Molz, 1988; Schellenger et al, 1989; Pearce II and Zahra, 1992; Daily and Dalton, 1992, 1993; Agrawal and Knoeber, 1996; Muth and Donaldson, 1998; Vafeas and Theodorou, 1998; Calleja, 1999; Lawrence and Stapledon, 1999; Dehaene et al, 2001; Cotter and Silverster, 2003; Kiel and Nicholson, 2003; Bonn et al, 2004; Dulewicz and Herbert, 2004; Chan and Li, 2008), the levels of board independence among sample companies are assessed at one point in time, i.e., mid-2003, using the above criteria. Director independence is evaluated from disclosures in the company’s 2003 report. Specifically, the director’s report or board of directors section of the annual report, which gives an introduction to each board member, is used to ascertain whether the director is a substantial shareholder or is an officer of a substantial shareholder (substantial shareholders of the company are disclosed in shareholder information section of the annual report), whether he/she has been employed in an executive capacity by the company or another group member (group members of the company, including associated entities and controlled entities, are disclosed in the notes to the financial statements), and whether he/she has served on the board for more than ten years. Moreover, from the corporate governance section and related party notes to the financial statements it is assessed whether the board member has been a principal or employee of a material adviser or consultant to the company, whether he/she is a material supplier or customer of the company, or an officer of a supplier or customer, and whether he/she has any material contractual relationship with the company other than as a director of the company. AASB 1031 provides guidance in relation to a quantitative assessment of materiality. An item is presumed to be material if it is equal to or greater than 10% of the appropriate base amount; according to the ASX (ASX, 2006), this would seem a reasonable position for consideration by the board in determining materiality. Thus in this study the materiality threshold is set at 10% of net assets or operating result before tax, for balance sheet or profit and loss items respectively. If a close analysis of the information could not provide an objective basis for determining director independence, the company is excluded from the analysis. It is introduced in Chapter 2 that the ASX asks each company to include a statement in its annual report from 2004 onwards disclosing the extent to which it has followed the 97 best practice recommendations; the company should identify the recommendations that have not been followed and give reasons for not following them. Therefore, from mid2003 Australian listed companies have been subject to the corporate governance principles of the Guidelines (2003); the results of this study may indicate the potential consequences of these recommendations on the performance of Australian firms. However, it should be noted that, as the Guidelines (2003) became effective in the financial year ending 30 June 2004, it would be unlikely that the board members would have changed radically from 2003 to 2004. Weisbach (1988) and Lawrence and Stapledon (1999) believed that board composition, in terms of executive, affiliated and independent directors, would only change slowly over time. This belief is supported by the ASX analysis of corporate governance disclosures (ASX, 2005, 2008), which shows that the compliance statistics for the best practice recommendations had been improved slowly but surely over 2003-2007. Although there may be an extraneous effect of anticipated regulatory change for the Guidelines (2003) to become effective in 2004, the effect, as noted by some authors (e.g., Larcker, Richardson and Tuna, 2007), could not create significant biases in the results, as it would affect all listed firms. 5.4.2 Performance Measures In Chapter 3, it is suggested that the performance measures in some papers are quite limited, for example, Baysinger and Bulter (1985), Agrawal and Knoeber (1996), Barnhart and Rosenstein (1998), Calleja (1999), Denis and Sarin (1999), Hossain et al (2001), Kiel and Nicholson (2003), Bonn et al (2004), Balatbat et al (2004), Peng (2004), Chin et al (2004), Chang and Leng (2004), Luan and Tang (2007), Choi et al (2007) and Chan and Li (2008). Currently, there appears to be no consensus concerning the selection of an appropriate set of measures which account for corporate financial performance (Chakravarthy, 1986); it is unlikely, however, that any one corporate performance indicator could sufficiently capture this performance dimension (Daily and Dalton, 1992). It is common to see several indices used because organizations legitimately seek to accomplish a variety of objectives, ranging from financial profitability to effective asset utilization and high stockholder returns (Hofer, 1983). 98 There are two broad groups of performance measures - “accounting measures drawn from the accounting systems used by firms to track their internal affairs and financial market measures relating to the share prices and dividend streams observed in the operation of financial markets” (Devinney, Richard, Yip and Johnson, 2005, p.15). Accounting measures of performance are historical and therefore experience a backward and inward looking focus. Developed as a reporting mechanism, they represent the impact of many factors, including the past successes of advice given from the board to the management team; they are the traditional mainstay of corporate performance factors (Kiel and Nicholson, 2003). However, accounting measures are “distortable”; this distortion arises from such sources as accounting procedures and policies, government policies towards specific activities, human error and purposeful deception (Devinney et al, 2005). Nevertheless, ROA and ROE are included in this study as measures of corporate performance; as noted by Muth and Donaldson (1998), ROA and ROE have been used extensively in research on board composition. In contrast, market-based measures are forward looking indicators that reflect current plans and strategies, in theory representing the discounted present value of future cash flows (Fisher and McGowan, 1983). Related to the value placed on the firm by the market, market measures are not susceptible to the impact of accounting policy changes or mere timing effects; they are objective in the sense that they exist outside of the influence of individuals (Devinney et al, 2005). Examples of market measures frequently endorsed by the authors in the field of corporate governance include shareholder return (e.g., Kenser, 1987; Hermalin and Weisbach, 1988; Molz, 1988; Schellenger et al, 1989; Muth and Donaldson, 1998; Vafeas and Theodorou, 1998; Calleja, 1999) and Tobin’s q (e.g., Hermalin and Weisbach, 1991; Yermack, 1996; Agrawal and Knoeber, 1996; Barnhart and Rosenstein, 1998; Bhagat and Black, 2000; Hossain et al, 2001; Kiel and Nicholson, 2003; Panasian et al, 2003; Anderson and Reeb, 2004; Randoy and Jenssen, 2004; Chin et al, 2004; Choi et al, 2007; Chan and Li, 2008), which are used in this study. The acceptance of shareholder return as a performance measure is also encouraged by the Australian Institute of Company Directors (AICD), Australian Employee Ownership Association (AEOA) and Australian Shareholders’ Association (ASA) (AICD, AEOA and ASA, 2007). Table 5.1 presents the definitions of the performance measures adopted in this research. 99 Table 5.1 Measures of Firm Performance Measure Definition ROA Ratio of EBIT to book value of total assets ROE Ratio of profit after interest and tax to book value of equity Shareholder return Realised annual rate of return incorporating capital gains and dividend payments Tobin’s q Ratio of market value to book value of total assets; market value of total assets is computed as market value of common stocks plus book value of preferred stocks and long-term debt The formulas for ROA and ROE are taken from Devinney et al (2005) and Management Accounting – Financial Strategy issued by the Chartered Institute of Management Accountants (CIMA, 2006). The definition of shareholder return is essentially the same as the ones used in prior studies, although it is termed shareholder wealth in Muth and Donaldson (1998), total return to shareholders in Molz (1988), or stock return in Hermalin and Weisbach (1988) and Vafeas and Theodorou (1998), among others. The unavailability of many of the variables comprising the theoretical Tobin’s Q in Lindenberg and Ross (1981) and Morck, Shleifer and Vishny (1988) prevent similar calculations being conducted. Instead, like the prior studies (e.g., Yermack, 1996; Agrawal and Knoeber, 1996; Barnhart and Rosenstein, 1998; Bhagat and Black, 2000; Hossain et al, 2001; Kiel and Nicholson, 2003; Panasian et al, 2003; Anderson and Reeb, 2004; Randoy and Jenssen, 2004; Chin et al, 2004; Choi et al, 2007; Chan and Li, 2008), the alternative formula for approximating Tobin’s q in Chung and Pruitt (1994) is followed. As discussed in Chapter 3, some prior studies also suffer from the limitation of shortterm observation of firm performance (e.g., Pfeffer, 1972; Molz, 1988; Schellenger et al, 1989; Daily and Dalton, 1992, 1993; Agrawal and Knoeber, 1996; Klein, 1998; Vafeas and Theodorou, 1998; Calleja, 1999; Dehaene et al, 2001; Cotter and Silverster, 2003; Bonn et al, 2004; Luan and Tang, 2007; Chan and Li, 2008); generalizability of their findings would have been enhanced if they had used data from a longer period (Bonn et al, 2004). 100 Shrader et al (1984), in examining the literature on the empirical relationship between strategic planning and organizational performance, found that most studies surrounding the topic had chosen 3- or 5-year periods as their time frames, as suggested to be appropriate for a given strategic planning intervention to take effect. Therefore, to reduce the influence of short-term fluctuations, the performance figures in the study are the three-year averages over the 2000-2003 and 2003-2006 financial years. 5.4.3 Control Variables From Appendix 1 and 2 which provide a summary of major research in Australia and overseas, it appears that some of the papers are cross sectional without the benefit of control variables (e.g., Pfeffer, 1972; Baysinger and Bulter, 1985; Kesner, 1987; Molz, 1988; Fosberg, 1989; Schellenger et al, 1989; Daily and Dalton, 1992; Dalton et al, 1998; Calleja, 1999; Dehaene et al, 2001; Kiel and Nicholson, 2003; Dulewicz and Herbert, 2004; Chin et al, 2004; Chang and Leng, 2004). In Bathala and Rao (1995) and Coles et al (2001), it is suggested that the mixed evidence on the correlation between board composition and firm performance may be attributed to the omission of other variables that affect firm performance. Schellenger et al (1989) argued that the conflicting findings with respect to the existence or nonexistence of a board composition effect on financial performance could be due to failure to control firm risk. In order to identify the specific effect of board independence on firm performance, and the effect of firm performance on board independence, a number of covariates are introduced into the models in this study to control for confounding influences on performance and independence. As pointed out by Bathala and Rao (1995), while the agency literature recognizes the importance of board of directors in monitoring of management decisions, this is only one of the mechanisms used to control agency conflicts. The literature identifies a few other devices which ensure that managers’ interests are aligned with those of shareholders, such as managerial ownership, dividend payout and leverage; the theoretical underpinnings for each of these are outlined below. Jensen and Meckling (1976) asserted that increasing managerial ownership could mitigate agency conflicts; the higher the proportion of equity owned by managers, the greater the alignment between managers and shareholder interests. The empirical 101 evidence supporting this view could be found in Morck et al (1988), Kim, Lee and Francis (1988) and Hudson, Jahera and Lloyd (1992). In relation to leverage and dividend payout, Jensen (1986) argued that the payment of dividends and the contractual obligations associated with debt reduced the amount of discretionary funds available to management, thereby reducing their incentive to engage in non-optimal activities. Similarly, Grossman and Hart (1982) suggested that increased debt would cause managers to become more efficient in order to lessen the probability of bankruptcy, and loss of control and reputation. According to Easterbrook (1984), the regular payment of dividends would force firms to go to the capital markets for investment funding; scrutiny of firms accessing the markets would act as a deterrent to opportunistic behaviours by managers. Harris and Raviv (1991) confirmed that the empirical evidence was broadly consistent with the proposition that debt could mitigate agency conflicts. In addition to the control mechanisms as discussed in Bathala and Rao (1995), the analysis in this research includes several control variables, which capture the firm characteristics likely to be associated with board composition and firm performance, drawing on the empirical models identified in prior studies into the determinants of board composition or firm performance. These variables include: board size (Barnhart and Rosenstein, 1998; Lawrence and Stapledon, 1999; Bhagat and Black, 2000; Singh and Davidson III, 2003; Randoy and Jenssen, 2004; Choi et al, 2007; Chan and Li, 2008); blockholder ownership (Bhagat and Black, 2000; Coles et al, 2001; Singh and Davidson III, 2003; Randoy and Jenssen, 2004); diversification (Hermalin and Weisbach, 1988; Yermack, 1996; Agrawal and Knoeber, 1996; Hossain et al, 2001); firm age (Daily and Dalton, 1993; Denis and Sarin, 1999; Bonn et al, 2004; Balatbat et al, 2004; Anderson and Reeb, 2004; Peng, 2004; Randoy and Jenssen, 2004; Krivogorsky, 2006; Choi et al, 2007); firm size (Hermalin and Weisbach, 1988, 1991; Pearce II and Zahra, 1992; Yermack, 1996; Agrawal and Knoeber, 1996; Muth and Donaldson, 1998; Barnhart and Rosenstein, 1998; Lawrence and Stapledon, 1999; Denis and Sarin, 1999; Bhagat and Black, 2000; Coles et al, 2001; Hossain et al, 2001; Cotter and 102 Silverster, 2003; Singh and Davidson III, 2003; Panasian et al, 2003; Anderson and Reeb, 2004; Peng, 2004; Randoy and Jenssen, 2004; Chen et al, 2005; Krivogorsky, 2006; Luan and Tang, 2007; Choi et al, 2007; Chan and Li, 2008); and risk (Muth and Donaldson, 1998; Klein, 1998; Denis and Sarin, 1999; Anderson and Reeb, 2004; Choi et al, 2007). The measures selected for board size, blockholder ownership, diversification, dividend payout, firm age, firm size, leverage, management shareholdings and risk are described below. Table 5.2 Measures of Control Variables Control Measure Definition Board size Board size Number of directors on the board Blockholder ownership Blockholder shareholdings The percentage of common stocks held by the top 20 shareholders Diversification Diversification Number of industrial and geographical segments Dividend payout Dividend ratio Ratio of dividend payments to profit after interest and tax Firm age Firm age Number of years listed on the ASX and one of the stock exchanges which were amalgamated to form the ASX in 1987 Firm size Market capitalisation Market value of common stocks Leverage Gearing ratio Ratio of short-term and long-term debt to book value of equity Managerial ownership Executive director The percentage of equity, including shareholdings options as well as common stocks, held by executive directors Firm risk Standard deviation of returns The measure of a firm’s total risk, i.e., the volatility of expected returns The measures for dividend payout, firm size and leverage are taken from Huntleys’ Shareholder (2003). These measures, along with the indicators for board size, 103 diversification, firm age and managerial ownership as listed in the table, have been typically used in the research surrounding this topic. Like Coles et al (2001) and Randoy and Jenssen (2004), in which blockholder ownership is defined as the proportion of equity owned by major shareholders, in this study the variable blockholder ownership represents the cumulative percentage of the company’s issued equity held by the top 20 shareholders; ASX Listing Rule 4.10.9 requires each company to include in its annual report the names of its 20 largest shareholders and the percentage of issued capital each hold. Firm risk could be measured in a number of ways (Baird and Thomas, 1990; Beatty and Zajac, 1994). Baird and Thomas (1990) reviewed how risk had been conceptualized in different disciplines of management, finance, marketing and psychology, and concluded that researchers in the area of strategic management typically defined risk as unpredictability of business outcome variables, e.g., variability of accounting or market return. Finance literature suggests that total risk, i.e., the uncertainty of expected returns, consists of systematic risk - the risk of the market, and unsystematic risk – the risk unique to the firm (e.g., Brigham, 1985; Reilly, 1985; Ross and Westerfield, 1988; Ross, Westerfield and Jaffe, 2005; Reilly and Brown, 2006). Schellenger et al (1989) asserted that board of directors could influence both the systematic and unsystematic risk of the firm. According to Hill and Snell (1988), Lorsch and MacIver (1989), Baysinger et al (1991), Davis and Thompson (1994), Heslin and Donaldson (1999) and Donaldson (2000), directors may raise or reduce total risk of the firm. Thus, it is decided to choose a measure of total risk, namely, the standard deviation of stock returns. Consistent with the performance figures, the figures for dividend payout, firm size and leverage are the three-year averages of 2000-2003 and 2003-2006; firm risk is also calculated for the periods of 2000-2003 and 2003-2006. Like the measures of board independence, data on board size, blockholder and executive director shareholdings, diversification and firm age are collected for the 2003 financial year. 5.5 Data Analysis In addition to descriptive statistics and correlation analysis, in this project, OLS and logit regressions are conducted for the research variables as described in the last section. 104 It is noted that a number of scholars have used simultaneous equations, such as two- and three-stage least squares to model the relationships between board variables and performance (e.g., Agrawal and Knoeber, 1996; Bhagat and Black, 2000; Balatbat et al, 2004; Anderson and Reeb, 2004; Chan and Li, 2008). However, in examining the sensitivity of simultaneous equation techniques in corporate governance research, Barnhart and Rosenstein (1998, p.2) found that “current theory provides little guidance in the specification of corporate governance models, and the econometric literature points out that misspecification of any of the equations in a system may result in serious bias in all of the equations. In fact, ordinary least squares (OLS) tends to be less sensitive to misspecification error …” Their arguments are summarised as follows. It is well known that OLS estimation of simultaneous equations models yields estimators that are biased and inconsistent (Judge, Hill, Griffiths, Lutkepohl and Lee, 1982). Although the 2SLS and 3SLS estimators are consistent, these estimators are biased and their exact distributions are known only for special cases, leading to questionable point estimates and statistical tests; for correctly specified models, the choice of instruments involves tradeoffs between bias and efficiency (Phillips, 1980). In Pindyck and Rubinfeld (1991, p.315), it is noted that a serious specification error in one equation could affect the parameter estimates in the other equations; thus systems estimation “involves a trade-off between the gain in efficiency and the potential costs of specification error.” Rhodes and Westbrook (1981), after an investigation of the exact density functions of OLS and 2SLS estimators when exogenous variables are wrongly excluded, concluded that under misspecification, OLS might be the superior estimation technique. In their study on the combined influence of ownership structure and board composition on corporate performance, Barnhart and Rosenstein (1998) reported that the empirical results were strongly dependent on the specification of the overall model and of the first-stage regressions; relatively minor changes in either have profound effects on overall results. It is therefore suggested that “sensitivity analysis is essential in most corporate governance research, where no formal structural model has been developed and a 105 variety of models exist that are similar in concept but different in specification, functional form, and control variables. In situations such as this, where the structure of empirical models is uncertain, systems estimation results should be interpreted cautiously, sensitivity analysis should be conducted, and OLS results should not be casually dismissed” (Barnhart and Rosenstein, 1998, p.2). As introduced in Chapter 3, some researchers, for example, Yermack (1996), Anderson and Reeb (2004), Chang and Leng (2004), Chen et al (2005) and Choi et al (2007), have performed panel regression analysis using either fixed-effect or random-effect models. These models are not tested in this study, because, if panel data is used, the sample would be further reduced. OLS and logit regressions are employed to overcome some of the limitations identified in prior research, including short-term observation of firm performance and failure to examine whether board characteristics are endogenously related to performance. In Chapter 4, based on agency theory, stewardship theory and organizational portfolio theory, five testable hypotheses are developed: H1: There is a negative relationship between board independence and past firm performance (organizational portfolio theory); H2 : There is a negative relationship between board independence and subsequent firm performance (stewardship theory, organizational portfolio theory); H3: There is positive relationship between board independence and subsequent firm performance (agency theory); H4 : There is a negative relationship between board independence and subsequent firm risk (organizational portfolio theory); and H5: There is a positive relationship between board independence and subsequent firm risk (agency theory). In the regressions to test H 1 , board independence serves as the dependent variable; the independent variables include performance, board size, blockholder and managerial shareholdings, diversification, dividend payout, firm age, firm size, leverage, and risk. The measures of board independence, board size, blockholder and managerial ownership, diversification, firm age use the 2003 data; the figures for firm performance, 106 dividend payout, firm size, leverage and risk are for the years 2000-2003. An algebraic statement of the models is as follows: Yi 1 ( Performance) i 2 ( BoardSize) i 3 ( BlockholderOwnership) i 4 ( Diversification) i 5 ( DividendPayout ) i 6 ( FirmAge) i 7 ( FirmSize) i 8 ( Leverage) i 9 ( ManagerialOwnership) i 10 ( Risk ) i i Where, for the i th company Y = Full board independence, monitoring committee independence or chairman independence = Constant of the equation = Coefficient of the variable Performance = ROA, ROE, shareholder return or Tobin’s q Board Size = Number of directors on the board Blockholder Ownership = Blockholder shareholdings Diversification = Number of industrial and geographical segments Dividend Payout = Dividend ratio Firm Age = Number of years listed on the ASX Firm Size = Natural logarithm of market capitalisation Leverage = Gearing ratio Managerial Ownership = Executive director shareholdings Firm Risk = Standard deviation of return = Error term In the models to test H 2 and H 3 , firm performance is the dependent variable, and the independent variables consist of board independence and other controls. The measures for performance, dividend payout, firm size, leverage and risk use the 2003-2006 figures; the rests use the 2003 data. The models can be described as: Yi 1 ( Independence) i 2 ( BoardSize) i 3 ( BlockholderOwnership) i 4 ( Diversification) i 5 ( DividendPayout ) i 6 ( FirmAge) i 7 ( FirmSize) i 8 ( Leverage) i 9 ( ManagerialOwnership ) i 10 ( Risk ) i i Where, for the i th company Y = ROA, ROE, shareholder return or Tobin’s q 107 = Constant of the equation = Coefficient of the variable Independence = Full board independence, monitoring committee independence or chairman independence Board Size = Number of directors on the board Blockholder Ownership = Blockholder shareholdings Diversification = Number of industrial and geographical segments Dividend Payout = Dividend ratio Firm Age = Number of years listed on the ASX Firm Size = Natural logarithm of market capitalisation Leverage = Gearing ratio Managerial Ownership = Executive director shareholdings Firm Risk = Standard deviation of return = Error term In the models to test H 4 and H 5 , the dependent variable is firm risk; the independent factors include board independence, performance and other control variables. Firm risk, performance, firm size, leverage and dividend payout are measured for the period 20032006; other variables use the 2003 data. Yi 1 ( Independence) i 2 ( BoardSize) i 3 ( BlockholderOwnership) i 4 ( Diversification) i 5 ( DividendPayout ) i 6 ( FirmAge) i 7 ( FirmSize) i 8 ( Leverage) i 9 ( ManagerialOwnership ) i 10 ( Performance) i i Where, for the i th company Y = Standard deviation of return = Constant of the equation = Coefficient of the variable Independence = Full board independence, monitoring committee independence or chairman independence Board Size = Number of directors on the board Blockholder Ownership = Blockholder shareholdings Diversification = Number of industrial and geographical segments Dividend Payout = Dividend ratio 108 Firm Age = Number of years listed on the ASX Firm Size = Natural logarithm of market capitalisation Leverage = Gearing ratio Managerial Ownership = Executive director shareholdings Performance = Shareholder return = Error term As discussed earlier in this chapter, the agency literature gives some guidelines in relation to the effects of dividend payout, leverage and managerial ownership on corporate performance (Bathala and Rao, 1995). In Coles et al (2001) and Singh and Davidson III (2003), it is argued that large block-holding shareholders may have greater incentives to monitor management than small investors as they have more at stake. According to resource dependence theory, increased board size may yield benefits by creating a network with the external environment and securing a broader resource base (Pfeffer, 1972; Pfeffer and Salancik, 1978; Zahra and Pearce II, 1989; Pearce II and Zahra, 1992). In addition, diversification has been shown to be value destroying by some authors (e.g., Berger and Ofek, 1996; Servaes, 1996; Denis, Denis and Sarin, 1997). However, the potential effects of firm age and firm size on performance are unclear in the literature. It seems that no formal structure has been developed for the influences of board size, blockholder and managerial shareholdings, diversification, dividend payout, firm age, firm size, leverage and risk on board independence, although these variables have been extensively used as controls in the research seeking to uncover the correlation between board composition and firm performance. In Bhagat and Black (2000), it is acknowledged that the factors that determine board composition are not well understood. As it is the case that “… the structure of empirical models is uncertain” (Barnhart and Rosenstein (1998, p.2), sensitivity tests using different performance measures, i.e., ROA, ROE, shareholder return and Tobin’s q, are provided to assess the robustness of the regression results for H 1 ; additional tests on the models for H 1 , H 2 , H 3 , H 4 and H 5 without firm size control are also performed over the sample periods. In statistics there is no specific rule of thumb for parsimony. As shown in Appendix 2, there are ten independent variables in Yermack (1996) and Hutchinson and Gul (2004), and eleven in Hermalin and Weisbach (1988) and Choi et al (2007). Since additional 109 tests without firm size are preformed, it appears that the concern that the regression models may not be parsimonious could not be justified. 5.6 Summary To identify the specific effect of board independence on firm performance, and the effect of firm performance on board independence, this study uses an archival research design, which is traditionally employed by the literature surrounding this topic. Most Australian studies and some overseas papers suffer from the limitation of small sample size. It is decided to use the top 500 companies listed on the ASX in 2003, ranked by market capitalisation, as the initial dataset. The sources of data required to conduct this research are available within the public domain. The most popular measurement of board independence is the proportion of outside or independent directors on the board. Drawing on the recommendations of the ASX in its Guidelines, five empirical proxies are developed in the study to represent full board independence, monitoring committee independence and chairman independence, among Australian public companies. There are four measures of firm performance – market-based measures of Tobin’s q and shareholder return, and accounting-based measures of ROA and ROE; they are the frequently used performance indicators in the field of corporate governance research. Following the approach endorsed by some researchers, board characteristics of sample companies are examined at one point in time: mid-2003. To correct the limitation of short-term observation of firm performance identified in some papers, the performance figures use the three-year averages over the 2000-2003 and 2003-2006 financial years. Some control variables are introduced into the data analysis, including board size, blockholder ownership, diversification, dividend payout, firm age, firm size, leverage, managerial ownership and firm risk. OLS and logit models in which board independence, firm performance and risk serve as the dependent variables are developed; the robustness of the findings is investigated through a series of tests using different measures for firm performance and/or without firm size control. 110 Chapter 6. Univariate Analysis 6.1 Introduction As disclosed in Chapter 5, after removing financial institutions from the 2003 list of top 500 companies, a sample of 384 companies is obtained, which is further reduced to 243 firms due to missing data. Individual directors are assessed in terms of their independence from management, based on the information included in the 2003 annual reports, with the definition of independence espoused by the ASX being adopted. These data are hand collected, making the evaluation of directors a time-intensive process. It is found that, among the 243 sample firms, 116 (47.74%) had an independent board chairman; 102 (41.98%), 163 (67.08%), 68 (28.40%) or 122 (50.21%) had a majority of independent directors sitting on the board, audit committee, nomination committee or remuneration committee, respectively. In 2003 some of the sample companies had undertaken reviews of their corporate governance practices in light of the recently released ASX Guidelines; some companies proposed that they would conduct such reviews in the next financial year. There are some companies which asserted that they had already been in compliance with the ASX recommendations; however, such announcements should be read with caution. For example, one company declared that “[a]ll current members of the board are ‘independent’ within the ASX definition, to the extent that the components of that definition can be objectively assessed”, although there was a managing director, who was also the chief executive of the company, on the board. This chapter presents the descriptive statistics and correlation analysis of research variables; the abbreviations for these variables adopted in the analysis are listed below. The remainder of this chapter is organized as follows. Section 2 shows preliminary statistics of the dataset. In Section 3, in order to explore the relationship between board independence and past performance, Pearson correlations for the sample period 20002003 are examined. Section 4 provides a correlation analysis for the period 2003-2006, to identify the influence of board independence on subsequent performance and risk. A summary of the findings is then produced in the last section. 111 Table 6.1 Abbreviations of Research Variables Abbreviation Variable FIND Percentage of independent directors on the board ACIND Percentage of independent directors on the audit committee NCIND Percentage of independent directors on the nomination committee RCIND Percentage of independent directors on the remuneration committee CMIND A binary variable to assess whether or not the chairman is an independent director ROA1 Return on assets for the period 2000-2003 ROA2 Return on assets for the period 2003-2006 ROE1 Return on equity for the period 2000-2003 ROE2 Return on equity for the period 2003-2006 SHRET1 Shareholder return for the period 2000-2003 SHRET2 Shareholder return for the period 2003-2006 TOBQ1 Tobin’s q for the period 2000-2003 TOBQ2 Tobin’s q for the period 2003-2006 SIZE Number of directors on the board BLOCK Percentage of equity held by the largest 20 shareholders SEGMT Number of industrial and geographical segments DIVR1 Dividend ratio for the period 2000-2003 DIVR2 Dividend ratio for the period 2003-2006 AGE Number of years listed on the ASX and one of the stock exchanges which were amalgamated to form the ASX in 1987 MCAP1 Market capitalisation (in $million) for the period 2000-2003 MCAP2 Market capitalisation (in $million) for the period 2003-2006 Log(MCAP) Natural logarithms of market capitalisation GEAR1 Gearing ratio for the period 2000-2003 GEAR2 Gearing ratio for the period 2003-2006 EQED Percentage of equity held by executive directors 112 RISK1 Standard deviation of shareholder return for the period 2000-2003 RISK2 Standard deviation of shareholder return for the period 2003-2006 6.2 Preliminary Statistics Schmidt (2005, p.24) observed that “[i]t is customary to include a table in your paper showing the means and standard deviations of every variable in the data set so that the reader can get a sense of what the data look like.” Table 6.2 gives a description of board characteristics for the 243 sample firms in 2003. Table 6.2 Descriptive Statistics: Boards of Directors Sample Period: 2003 Included Observations: 243 * Variable Mean Median Maximum Minimum Std. Dev Skewness Kurtosis FIND 41.65% 40.00% 100% 0% 0.22 0.03 2.56 ACIND* 54.57% 60.00% 100% 0% 0.34 -0.25 2.06 NCIND* 23.29% 0% 100% 0% 0.33 1.07 2.78 RCIND* 41.15% 50.00% 100% 0% 0.36 0.24 1.78 SIZE 6.33 6.00 15.00 3.00 2.05 1.02 4.53 For a firm without audit, nomination or remuneration committee, its ACIND, NCIND or RCIND is deemed to be 0% Kurtosis included in the table measures the peakedness or flatness of the distribution of the data. The kurtosis of the normal distribution is 3. If the kurtosis exceeds 3, the distribution is peaked relative to the normal; if the kurtosis is less than 3, the distribution is flat relative to the normal. Casual observation of Table 6.2 reveals that the sample contains a wide range of firms. The proportion of independent directors on the board, audit committee, nomination committee or remuneration committee varies between 0% and 100%, with a mean of 41.65%, 54.57%, 23.29% or 41.15%, respectively. The total number of directors on the board ranges from a low of 3 to a high of 15, with an average of just over 6. Based on the mean, median and standard deviation of the percentage of independent directors on the nomination committee, it could be concluded that Australian public 113 companies, in general, had not been in compliance with the recommendations that each company should establish a nomination committee and a majority of its members should be independent directors. Most firms, however, had an audit committee which had been dominated by independent members. The above findings may be explained by the changes in the listing requirements of the ASX. In January 2003, the ASX introduced Listing Rule 12.7 which provides that the top 500 companies must have an audit committee and that the composition of the audit committee must comply with the best practice recommendations of the ASX Corporate Governance Council; however, it is not mandatory for these companies to have a nomination committee and a remuneration committee. The firm characteristics for the sample in 2003 are summarized below. Table 6.3 Descriptive Statistics: Other Research Variables Sample Period: 2003 Included Observations: 243 Variable Mean Median Maximum Minimum Std. Dev Skewness Kurtosis BLOCK 65.10% 67.09% 99.86% 13.60% 0.18 -0.42 2.74 SEGMT 4.46 4.00 11.00 1.00 2.23 0.84 3.19 AGE 16.90 11.00 132.00 3.00 17.81 2.90 15.39 EQED 11.84% 2.21% 80.99% 0% 0.18 1.70 4.89 It is confirmed in the table above that the sample is comprised of firms with widely differing attributes. The number of years listed on the stock exchange varies from 3 to 132, and number of industrial and geographical segments ranges from 1 to 11. The proportion of equity held by the top 20 shareholders ranges from 13.60% to 99.86%, and proportion of equity held by executive directors varies between 0% and 80.99%. Startz (2007, p.190) noted that “[t]here are relatively few places in econometrics where normality of the data is important. In particular, there is no requirement that the variables in a regression be normally distributed. I don’t know where this myth comes from.” Nevertheless, the Jarque-Bera tests for normality are performed; the findings are displayed in Table 6.4, as part of the preliminary statistics. Unless otherwise indicated, the levels of significance reported in this paper are for two-tailed tests. 114 Table 6.4 Jarque-Bera Statistics Sample Period: 2003 Included Observations: 243 Variable Statistic Probability FIND 1.98 0.3710 ACIND 11.62 0.0030 NCIND 46.78 0 RCIND 17.35 0.0001 SIZE 66.06 0 BLOCK 7.84 0.0198 SEGMT 28.88 0 AGE 1895.10 0 EQED 152.71 0 The Jarque-Bera statistic measures the difference of the skewness and kurtosis of the data with those from the normal distribution. The reported p-value is the probability that a Jarque-Bera statistic exceeds the observed value under the null hypothesis of a normal distribution - a small p-value leads to the rejection of the null hypothesis. As shown in the table, the hypothesis of normal distribution is rejected, for every variable other than the percentage of independent directors on the board, at the 1% or 5% significance level. Thus, in this sample most data is not normally distributed. 6.3 Correlations: The Sample Period 2000-2003 Table 6.5 gives Pearson product-moment correlations among the measures for board independence and past performance, and shows several significant correlation coefficients. 115 Table 6.5 Pearson Correlations: 2000-2003 Sample Period: 2000-2003 Included Observations: 243 Correlation t-Statistic FIND FIND 1.000 ACIND NCIND RCIND CMIND ROA1 ROE1 SHRET1 TOBQ1 ----ACIND 0.752 1.000 17.731** ----NCIND RCIND CMIND ROA1 ROE1 SHRET1 TOBQ1 * 0.441 0.368 1.000 7.630** 6.136** ----- 0.623 0.552 0.523 1.000 12.350** 10.268** 9.527** ----- 0.531 0.358 0.265 0.441 1.000 9.736** 5.943** 4.267** 7.618** ----- -0.013 0.076 0.038 -0.006 -0.070 1.000 -0.195 1.187 0.593 -0.086 -1.095 ----- 0.071 0.059 0.046 0.020 0.004 0.313 1.000 1.109 0.919 0.716 0.315 0.060 5.111** ----- -0.064 -0.049 -0.126 -0.155 -0.077 -0.127 -0.090 1.000 -0.989 -0.766 -1.966 -2.443* -1.196 -1.988* -1.396 ----- 0.012 -0.069 -0.095 0.031 0.006 -0.367 -0.160 0.043 0.192 -1.080 -1.483 0.487 0.097 -6.118** -2.517* 0.674 Significance at the 5% level ** 1.000 ----- Significance at the 1% level First, the positive relations between full board independence, committee independence and chairman independence are significant at the 1% level. Therefore a company with a higher percentage of independent directors on the board tends to have higher percentages of independent directors on the monitoring committees, with a higher chance that the chairman of the board is also an independent director. The performance variables fall into two clusters - accounting measures of ROA and ROE, and market-based measures of shareholder return and Tobin’s q; there is a strong positive correlation between ROA and ROE. Both shareholder return and Tobin’s q are negatively related to ROA; Tobin’s q is also inversely related to ROE at the 5% level of 116 significance. The findings are consistent with the Australian paper of Muth and Donaldson (1998), in which the correlation analysis shows that the performance variables fall into three distinct clusters – profit performance, shareholder return and sales growth; as noted by Muth and Donaldson (1998), similar results were obtained by Hamilton and Shergill (1992) in New Zealand when the authors subjected individual performance variables to factor analysis to generate a composite index of company performance. However, Table 6.5 indicates that, in general, there is no statistically significant association between past firm performance and board independence, although the average shareholder return for the past three years is negatively correlated with remuneration committee independence at the 5% level of significance. A comprehensive analysis of all research variables for the sample period 2000-2003 is provided in a list format, with p-values, in Appendix 4. There are quite a few significant coefficients in the correlations; the findings in relation to the variables which show some influences on board independence are summarised below. According to Appendix 4, larger firms as measured by market capitalisation have higher percentages of independent directors on the board and board committees. Companies with longer trading history on the stock exchange, or lower managerial shareholdings, have higher levels of full board and nomination committee independence. In addition, board size and diversification are positively related to full board and monitoring committee independence. It appears that higher blockholder ownership may lower the level of independence on the board, and audit and remuneration committees. Dividend payout has a positive effect on nomination and remuneration independence. Lastly, firm risk gives a negative impact on audit committee independence. 6.4 Correlations: The Sample Period 2003-2006 The correlations for board independence, subsequent performance and firm risk are exhibited in the following table. Similar to the findings in Table 6.5, for the sample period 2003-2006 there is a positive relation between ROA and ROE; Tobin’s q is inversely related to ROA and ROE at the 1% level of significance. 117 Table 6.6 Pearson Correlations: 2003-2006 Sample Period: 2003-2006 Included Observations: 243 Correlation t-Statistic FIND FIND 1.000 ACIND NCIND RCIND CMIND ROA2 ROE2 SHRET2 TOBQ2 RISK2 ----ACIND 0.752 1.000 17.731** ----NCIND 0.441 0.368 1.000 7.630** 6.136** ----RCIND 0.623 0.552 0.523 1.000 12.350** 10.268** 9.527** ----CMIND 0.531 0.358 0.265 0.441 1.000 9.736** 5.943** 4.267** 7.618** ----ROA2 ROE2 SHRET2 TOBQ2 RISK2 * -0.046 -0.003 -0.019 -0.036 -0.078 1.000 -0.719 -0.050 -0.288 -0.554 -1.220 ----- -0.055 -0.045 -0.028 -0.024 -0.066 0.328 -0.851 -0.699 -0.439 -0.370 -1.024 5.390** ----- -0.023 -0.041 -0.052 -0.032 0.005 0.005 0.029 1.000 -0.355 -0.632 -0.805 -0.504 0.071 0.071 0.452 ----- 0.073 -0.021 -0.001 0.004 0.092 -0.727 -0.197 -0.010 1.141 -0.326 -0.015 0.060 1.428 16.451** 3.126** -0.154 ----- -0.079 -0.081 -0.010 -0.067 -0.013 -0.044 -0.010 0.955 -0.008 -1.230 -1.259 -1.559 -1.037 -0.204 -0.677 -0.151 49.797** -0.126 Significance at the 5% level ** 1.000 1.000 1.000 ----- Significance at the 1% level From Table 6.6, it appears that full board independence, monitoring committee independence and chairman independence do not have any influence on subsequent firm performance and risk. In Appendix 5, a detailed analysis of all research variables for the period 2003-2006 is provided. In addition to the findings with respect of the relationships between board size, blockholder shareholdings, diversification, firm age, firm size or managerial ownership, and measures of board independence as reported in Section 6.3, Appendix 5 118 identifies a positive effect of board size on ROA, and a negative effect of board size on Tobin’s q. According to Appendix 4 and 5, the absolute values of correlations for the independent variables are well below 0.8 or 0.9, the rule of thumb for multicollinearity. Although there are high correlations among the various board independence measures, as introduced in Section 5.5 there is only one independence measure tested in each of the regressions. Therefore, there is no multicollinearity concern for the regression results, which are reported in the next chapter. 6.5 Summary The sample chosen in this study includes 243 firms drawn from the 2003 list of top 500 Australian companies; data is collected from databases and corporate annual reports. The sample is comprised of firms with widely differing attributes. It appears that in 2003 Australian publicly listed companies, in general, had not been in compliance with the recommendations that each company should establish a nomination committee and a majority of its members should be independent directors. However, most companies had an audit committee dominated by independent directors. To investigate the relationships between board independence, firm performance and risk, correlation analyses for the periods 2000-2003 and 2003-2006 are conducted. The results suggest that companies with a higher percentage of independent directors on the board tend to have higher percentages of independent directors on the monitoring committees, with higher chance that the chairman is also independent. The performance measures fall into two clusters - accounting and market-based measures. It is reported that larger firms have higher percentages of independent directors on the board and board committees. Companies with longer trading history or lower managerial ownership have higher levels of board and nomination committee independence. Board size and diversification are positively related to board and committee independence. Higher blockholder shareholdings may reduce the level of independence on the board, and audit and remuneration committees. Dividend payout has a positive effect on nomination and remuneration independence; firm risk shows a negative impact on audit independence. 119 Although a negative influence of shareholder return on remuneration committee independence is identified, there is no significant correlation between other measures for board independence, firm performance and risk. Therefore, the Pearson correlations indicate that the predictive power of agency theory, stewardship theory and organizational portfolio theory, with respect of the relationships between board independence and firm performance and risk as discussed in Chapter 4, may be limited. 120 Chapter 7. Multivariate Analysis 7.1 Introduction As illustrated in Chapter 4, there are three theoretical frameworks providing different expectations on the relationships between board independence and firm performance. According to agency theory, firms benefit from the oversight a board could provide by shielding the invested stakes of equity and debt holders from potential managerial selfinterest; where the board of directors is more independent of management, corporate performance should be higher. Stewardship theory offers opposing predictions about the structuring of effective boards; it is argued that managers have a wide range of motives beyond self-interest, and board of directors with a lower level of independence may lead to higher performance. As a new theory waiting for empirical testing, organizational portfolio model proposes that poor performance may trigger the installation of an independent chair and a higher proportion of independent directors on the board; the resulting risk-averse governance would foster a gradual decline in firm performance. From the above propositions several research hypotheses are developed: H1: There is a negative relationship between board independence and past firm performance (organizational portfolio theory); H2 : There is a negative relationship between board independence and subsequent firm performance (stewardship theory, organizational portfolio theory); H3: There is positive relationship between board independence and subsequent firm performance (agency theory); H4 : There is a negative relationship between board independence and subsequent firm risk (organizational portfolio theory); and H5: There is a positive relationship between board independence and subsequent firm risk (agency theory). To test these hypotheses, in this chapter a regression analysis of the research variables is undertaken. Section 2 reports the findings of OLS and logit models estimated for the effects of firm performance and other variables on board independence. The regressions 121 specified for the influences of board independence and other variables on performance and risk are introduced in Section 3 and 4, followed by a summary in the last section. 7.2 Regressions: Board Independence and Past Performance As discussed in Chapter 5, in the regressions to test H 1 board independence serves as the dependent variable. The independent variables include firm performance, board size, blockholder and managerial ownership, diversification, dividend payout, firm age and size, leverage and firm risk. The models are shown as follows: Yi 1 ( Performance) i 2 ( BoardSize) i 3 ( BlockholderOwnership) i 4 ( Diversification) i 5 ( DividendPayout ) i 6 ( FirmAge) i 7 ( FirmSize) i 8 ( Leverage) i 9 ( ManagerialOwnership) i 10 ( Risk ) i i Where, for the i th company Y = FIND, ACIND, NCIND, RCIND or CMIND = Constant of the equation = Coefficient of the variable Performance = ROA1, ROE1, SHRET1 or TOBQ1 Board Size = SIZE Blockholder Ownership Diversification Dividend Payout Firm Age = BLOCK = SEGMT = DIVR1 = AGE Firm Size = Log(MCAP1) Leverage = GEAR1 Managerial Ownership Firm Risk = EQED = RISK1 = Error term Table 7.1 provides regression results for the effects of ROA and other variables on board independence; the explanatory variables together account for 3.30%-21.20% of the cross sectional variation in independence measures. In the table, past ROA presents a negative influence on remuneration committee independence at 5% level of significance. 122 Table 7.1 OLS and Logit Regressions: Board Independence and Past Performance (ROA) Sample Period: 2000-2003 Included Observations: 243 Coefficient t/z-Statistic FIND ACIND NCIND RCIND CMIND Intercept 0.427 0.534 -0.187 0.147 0.403 5.720** 4.710** -1.794 1.277 0.558 -0.050 0.031 -0.109 -0.166 -0.870 -0.985 0.405 -1.532 -2.102* -1.596 -0.001 0.029 0.011 0.024 -0.056 -0.142 2.020* 0.827 1.684 -0.620 -0.258 -0.384 -0.155 -0.372 -1.088 -3.246** -3.186** -1.393 -3.033** -1.416 0.006 -0.003 0.005 -0.004 -0.044 0.768 -0.303 0.470 -0.360 -0.595 -0.001 0.025 0.026 0.083 0.343 -0.036 0.507 0.579 1.661 1.076 0.0007 0.0009 -2.78E-05 -0.0002 -0.002 0.868 0.743 -0.024 -0.169 -0.240 0.023 0.011 0.077 0.060 0.125 1.743 0.563 4.095** 2.888** 0.964 0.013 0.015 0.007 0.015 0.132 1.314 1.046 0.531 1.017 1.008 -0.033 0.0007 -0.014 0.083 -0.404 -0.404 0.005 -0.127 0.660 -0.505 -0.003 -0.0002 -0.007 -0.012 -0.096 -0.548 -0.026 -0.855 -1.421 -0.989 0.118 0.105 0.212 0.195 0.033 0.079 0.067 0.178 0.160 Std Error (Regression) 0.213 0.324 0.298 0.329 0.500 F/LR-Statistic 3.089 2.734 6.252 5.626 11.226 ROA1 SIZE BLOCK SEGMT DIVR1 AGE Log(MCAP1) GEAR1 EQED RISK1 R 2 /McFadden R 2 Adjusted R 2 123 Probability (F/LR-Statistic) 0.001 0.003 0.000 0.000 Durbin-Watson 1.685 1.906 2.152 1.761 * Significance at the 5% level ** 0.340 Significance at the 1% level Durbin-Watson is the classic test statistic for serial correlation; a number close to 2 is consistent with no serial correlation, while a number closer to 0 means there probably is a serial correlation. Therefore for the above models there is no indicator of serial correlation. Table 7.2 displays the effects of ROE and other variables on board independence; the independent variables account for 2.50%-20.40% of the variance in each of the independence measures. No significant relationship between past ROE and board independence is found. 124 Table 7.2 OLS and Logit Regressions: Board Independence and Past Performance (ROE) Sample Period: 2000-2003 Included Observations: 243 Coefficient t/z-Statistic FIND ACIND NCIND RCIND CMIND Intercept 0.462 0.539 -0.145 0.202 0.697 6.315** 4.851** -1.410 1.769 0.988 0.031 0.032 0.003 -0.012 0.018 1.309 0.887 0.083 -0.336 0.077 -0.001 0.029 0.011 0.025 -0.053 -0.106 2.025* 0.856 1.708 -0.593 -0.275 -0.390 -0.172 -0.392 -1.222 -3.472** -3.238** -1.537 -3.168** -1.599 0.005 -0.004 0.004 -0.005 -0.047 0.697 -0.314 0.411 -0.421 -0.638 -0.013 0.025 0.009 0.060 0.213 -0.414 0.531 0.205 1.207 0.694 0.0006 0.0009 -0.0001 -0.0003 -0.003 0.750 0.709 -0.093 -0.235 -0.320 0.021 0.011 0.073 0.055 0.099 1.552 0.562 3.903** 2.640** 0.778 0.011 0.015 0.006 0.014 0.117 1.188 1.012 0.449 0.928 0.952 -0.059 -0.020 -0.024 0.080 -0.453 -0.709 -0.160 -0.207 0.620 -0.565 -0.002 -0.0002 -0.005 -0.010 -0.068 -0.285 0.027 -0.663 -1.203 -0.800 0.120 0.108 0.204 0.180 0.025 0.082 0.069 0.170 0.145 Std Error (Regression) 0.213 0.323 0.300 0.332 0.503 F/LR-Statistic 3.173 2.804 5.958 5.101 8.331 ROE1 SIZE BLOCK SEGMT DIVR1 AGE Log(MCAP1) GEAR1 EQED RISK1 R 2 /McFadden R 2 Adjusted R 2 125 Probability (F/LR-Statistic) 0.0008 0.003 0.000 0.000001 Durbin-Watson 1.609 1.882 2.124 1.762 * Significance at the 5% level ** 0.597 Significance at the 1% level The regression results for shareholder return on board independence are presented in Table 7.3; the explanatory variables account for 2.60%-21.60% of the variance in the measures of board independence. A negative impact of shareholder return on remuneration committee independence, at 5% level of significance, is identified. 126 Table 7.3 OLS and Logit Regressions: Board Independence and Past Performance (Shareholder Return) Sample Period: 2000-2003 Included Observations: 243 Coefficient t/z-Statistic FIND ACIND NCIND RCIND CMIND Intercept 0.448 0.529 -0.135 0.227 0.720 6.183** 4.814** -1.332 2.042* 1.031 -0.023 -0.059 -0.120 -0.180 -0.275 -0.509 -0.855 -1.889 -2.584* -0.631 -0.001 0.028 0.011 0.025 -0.053 -0.124 2.009* 0.849 1.719 -0.598 -0.269 -0.389 -0.190 -0.425 -1.262 -3.381** -3.229** -1.711 -3.480** -1.650 0.005 -0.005 0.002 -0.009 -0.053 0.662 -0.395 0.174 -0.761 -0.710 -0.006 0.036 0.022 0.077 0.246 -0.200 0.748 0.505 1.578 0.795 0.0007 0.001 0.0001 2.41E-05 -0.002 0.876 0.849 0.113 0.019 -0.248 0.021 0.012 0.072 0.052 0.097 1.611 0.587 3.874** 2.565* 0.757 0.013 0.017 0.010 0.019 0.125 1.323 1.172 0.707 1.259 1.015 -0.037 0.0002 -0.028 0.063 -0.458 -0.460 0.002 -0.243 0.505 -0.581 0.011 0.033 0.064 0.094 0.089 0.409 0.821 1.709 2.284* 0.338 0.115 0.108 0.216 0.203 0.026 0.077 0.069 0.183 0.168 Std Error (Regression) 0.213 0.323 0.297 0.328 0.503 F/LR-Statistic 3.009 2.798 6.405 5.901 8.726 SHRET1 SIZE BLOCK SEGMT DIVR1 AGE Log(MCAP1) GEAR1 EQED RISK1 R 2 /McFadden R 2 Adjusted R 2 127 Probability (F/LR-Statistic) 0.001 0.003 0.000 0.000 Durbin-Watson 1.664 1.907 2.127 1.792 * Significance at the 5% level ** 0.558 Significance at the 1% level Table 7.4 provides regression estimates in relation to Tobin’s q on full board independence, committee independence and chairman independence. The models explain 2.50%-20.70% of the variation in dependent variables. According to the results, there is no statistically significant association between Tobin’s q and independence measures. 128 Table 7.4 OLS and Logit Regressions: Board Independence and Past Performance (Tobin’s Q) Sample Period: 2000-2003 Included Observations: 243 Coefficient t/z-Statistic FIND ACIND NCIND RCIND CMIND Intercept 0.430 0.532 -0.121 0.159 0.650 5.689** 4.639** -1.138 1.355 0.894 0.005 -0.003 -0.007 0.014 0.011 0.736 -0.278 -0.856 1.478 0.181 0.0004 0.028 0.009 0.030 -0.049 0.045 1.900 0.646 2.004* -0.538 -0.258 -0.384 -0.182 -0.375 -1.200 -3.240** -3.172** -1.630 -3.033** -1.566 0.006 -0.004 0.003 -0.003 -0.045 0.824 -0.323 0.296 -0.232 -0.607 -0.004 0.027 0.001 0.073 0.227 -0.120 0.551 0.030 1.468 0.728 0.0008 0.0009 -0.0003 -4.68E-05 -0.002 0.926 0.712 -0.214 -0.036 -0.286 0.019 0.014 0.078 0.046 0.093 1.381 0.669 4.009** 2.154* 0.708 0.013 0.015 0.005 0.015 0.119 1.327 1.038 0.379 1.040 0.960 -0.050 0.011 0.0009 0.027 -0.474 -0.605 0.087 0.008 0.210 -0.587 -0.002 -0.0006 -0.005 -0.010 -0.068 -0.428 -0.076 -0.686 -1.163 -0.810 0.116 0.105 0.207 0.187 0.025 0.078 0.067 0.173 0.152 Std Error (Regression) 0.213 0.324 0.299 0.331 0.503 F/LR-Statistic 3.041 2.725 6.049 5.353 8.358 TOBQ1 SIZE BLOCK SEGMT DIVR1 AGE Log(MCAP1) GEAR1 EQED RISK1 R 2 /McFadden R 2 Adjusted R 2 129 Probability (F/LR-Statistic) 0.001 0.003 0.000 0.000 Durbin-Watson 1.675 1.907 2.108 1.770 * Significance at the 5% level ** 0.594 Significance at the 1% level There are some consistent findings in Table 7.1, 7.2, 7.3 and 7.4 with respect to the relationships between blockholder ownership, firm size and board composition. The tables indicates that companies with higher blockholder shareholdings have lower percentages of independent directors on the board, and audit and remuneration committees; larger firms have relatively more independent directors on nomination and remuneration committees. The results for sensitivity tests without firm size control are reported in Appendix 6, 7, 8 and 9. The link between ROA and remuneration committee independence ceases to be significant; instead, a positive impact of Tobin’s q on remuneration committee independence emerges. In addition, there is a transfer of the positive effects on nomination and remuneration committee independence from firm size to board size; as introduced in Chapter 6, board size and firm size are strongly correlated. 7.3 Regressions: Board Independence and Subsequent Performance In the models to test H 2 and H 3 , performance is the dependent variable; the independent variables consist of board independence and other controls. The models are described as: Yi 1 ( Independence) i 2 ( BoardSize) i 3 ( BlockholderOwnership) i 4 ( Diversification) i 5 ( DividendPayout ) i 6 ( FirmAge) i 7 ( FirmSize) i 8 ( Leverage) i 9 ( ManagerialOwnership ) i 10 ( Risk ) i i Where, for the i th company Y = ROA2, ROE2, SHERT2 or TOBQ2 = Constant of the equation = Coefficient of the variable Independence = FIND, ACIND, NCIND, RCIND or CMIND Board Size = SIZE Blockholder Ownership = BLOCK Diversification = SEGMT 130 Dividend Payout Firm Age = DIVR2 = AGE Firm Size = Log(MCAP2) Leverage = GEAR2 Managerial Ownership Firm Risk = EQED = RISK2 = Error term The influences of full board independence and other variables on firm performance are reported in Table 7.5; the explanatory variables account for 10.60%-94.50% of the variance in performance measures. In the table, full board independence gives a positive contribution to Tobin’s q, and a negative contribution to ROE, at 5% level of significance. 131 Table 7.5 OLS Regressions: Full Board Independence and Subsequent Performance Sample Period: 2003-2006 Included Observations: 243 Coefficient t-Statistic ROA2 ROE2 SHERT2 TOBQ2 Intercept -0.226 0.407 -0.625 2.153 -1.112 0.740 -7.125** 1.816 -0.319 -1.088 0.097 1.987 -1.853 -2.339* 1.307 1.982* -0.003 -0.070 -0.017 -0.352 -0.120 -1.093 -1.666 -2.533* -0.088 -1.203 0.135 0.918 -0.403 -2.051* 1.440 0.726 -0.0008 0.105 -0.014 -0.103 -0.043 1.984* -1.639 -0.909 0.200 0.532 0.070 -1.040 2.445* 2.404* 1.980* -2.181* 0.001 -0.002 0.0001 -0.014 0.471 -0.410 0.135 -1.093 0.061 0.157 0.107 0.307 2.007* 1.906 8.150** 1.728 -0.019 -0.703 -0.016 -0.016 -1.083 -14.604** -2.130* -0.150 -0.366 0.062 -0.125 3.582 -1.656 0.104 -1.316 2.780** -0.002 0.048 0.579 -0.054 -0.081 0.797 60.540** -0.420 0.106 0.502 0.945 0.111 0.068 0.481 0.942 0.072 Std Error (Regression) 0.560 1.514 0.242 3.264 F-Statistic 2.754 23.416 397.120 2.882 FIND SIZE BLOCK SEGMT DIVR2 AGE Log(MCAP2) GEAR2 EQED RISK2 R2 Adjusted R 2 132 Probability (F-Statistic) 0.003 0.000 0.000 0.002 Durbin-Watson 2.052 2.078 2.014 1.926 * Significance at the 5% level ** Significance at the 1% level Table 7.6 displays regression results for audit committee independence and other variables on performance; the independent variables account for 9.70%-94.40% of the variance in ROA, ROE, shareholder return and Tobin’s q. As shown below, there is a negative influence of audit committee independence on ROE at the 5% level. 133 Table 7.6 OLS Regressions: Audit Committee Independence and Subsequent Performance Sample Period: 2003-2006 Included Observations: 243 Coefficient t-Statistic ROA2 ROE2 SHERT2 TOBQ2 Intercept -0.298 0.294 -0.599 2.769 -1.496 0.547 -6.972** 2.377* -0.126 -0.681 0.029 0.468 -1.104 -2.216* 0.600 0.703 0.001 -0.048 -0.018 -0.369 0.060 -0.741 -1.745 -2.612** -0.051 -1.176 0.120 0.569 -0.235 -2.006* 1.284 0.447 -0.003 0.097 -0.013 -0.092 -0.144 1.835 -1.568 -0.805 0.205 0.561 0.069 -1.056 2.490* 2.529* 1.938 -2.194* 0.001 -0.002 0.0002 -0.013 0.445 -0.401 0.161 -1.036 0.055 0.139 0.109 0.350 1.809 1.691 8.346** 1.969 -0.020 -0.705 -0.016 -0.015 -1.093 -14.625** -2.117* -0.147 -0.359 0.086 -0.128 3.535 -1.615 0.144 -1.336 2.724** -0.001 0.049 0.578 -0.059 -0.056 0.814 60.353** -0.451 0.098 0.501 0.944 0.097 0.059 0.480 0.942 0.058 Std Error (Regression) 0.563 1.516 0.242 3.288 F-Statistic 2.510 23.307 394.693 2.503 ACIND SIZE BLOCK SEGMT DIVR2 AGE Log(MCAP2) GEAR2 EQED RISK2 R2 Adjusted R 2 134 Probability (F-Statistic) 0.007 0.000 0.000 0.007 Durbin-Watson 2.036 2.046 2.020 1.934 * Significance at the 5% level ** Significance at the 1% level The regression results in relation to the nomination committee independence are provided in Table 7.7; the explanatory variables account for 10.00%-94.50% of the variance in performance measures. No relationship between nomination committee independence and subsequent performance is identified. 135 Table 7.7 OLS Regressions: Nomination Committee Independence and Subsequent Performance Sample Period: 2003-2006 Included Observations: 243 Coefficient t-Statistic ROA2 ROE2 SHERT2 TOBQ2 Intercept -0.392 -0.109 -0.588 3.112 -2.067* -0.211 -7.176** 2.798** -0.229 -0.361 -0.046 0.795 -1.897 -1.094 -0.872 1.122 0.003 -0.061 -0.016 -0.372 0.112 -0.928 -1.573 -2.651** -0.038 -0.969 0.102 0.510 -0.178 -1.668 1.108 0.409 3.80E-05 0.104 -0.013 -0.103 0.002 1.937 -1.524 -0.892 0.195 0.523 0.069 -1.021 2.389* 2.340* 1.963 -2.129* 0.0009 -0.003 0.0002 -0.013 0.394 -0.504 0.190 -1.006 0.065 0.149 0.112 0.317 2.099* 1.769 8.399** 1.756 -0.017 -0.700 -0.016 -0.023 -0.974 -14.382** -2.094* -0.221 -0.363 0.081 -0.129 3.551 -1.644 0.135 -1.349 2.740** -0.002 0.048 0.578 -0.055 -0.111 0.801 60.326** -0.420 0.107 0.493 0.945 0.100 0.068 0.471 0.942 0.062 Std Error (Regression) 0.560 1.528 0.242 3.282 F-Statistic 2.773 22.577 395.414 2.587 NCIND SIZE BLOCK SEGMT DIVR2 AGE Log(MCAP2) GEAR2 EQED RISK2 R2 Adjusted R 2 136 Probability (F-Statistic) 0.003 0.000 0.000 0.005 Durbin-Watson 2.044 2.038 2.015 1.947 * Significance at the 5% level ** Significance at the 1% level Table 7.8 gives regression estimates for the effect of remuneration committee independence on performance measures; the regressions explain 10.10%-94.50% of the variation in dependent variables. It is found that remuneration committee independence is inversely associated with the accounting performance measures of ROA and EOE, at 5% level of significance. 137 Table 7.8 OLS Regressions: Remuneration Committee Independence and Subsequent Performance Sample Period: 2003-2006 Included Observations: 243 Coefficient t-Statistic ROA2 ROE2 SHERT2 TOBQ2 Intercept -0.317 0.066 -0.573 2.858 -1.668 0.128 -6.955** 2.559* -0.221 -0.617 -0.048 0.740 -2.032* -2.090* -1.030 1.159 0.005 -0.048 -0.016 -0.380 0.210 -0.734 -1.505 -2.689** -0.094 -1.168 0.089 0.694 0.665 -1.988* 0.850 0.546 -0.003 0.099 -0.013 -0.093 -0.138 1.865 -1.593 -0.812 0.207 0.551 0.072 -1.061 2.534* 2.483* 2.034* -2.211* 0.0008 -0.003 0.0002 -0.013 0.361 -0.541 0.173 -0.986 0.063 0.157 0.112 0.324 2.055* 1.891 8.450** 1.811 -0.021 -0.707 -0.017 -0.012 -1.156 -14.626** -2.183* -0.117 -0.340 0.141 -0.124 3.471 -1.538 0.235 -1.296 2.677** 3.76E-05 0.053 0.578 -0.063 0.002 0.890 60.451** -0.486 0.109 0.500 0.945 0.101 0.070 0.478 0.942 0.062 Std Error (Regression) 0.559 1.518 0.242 3.282 F-Statistic 2.831 23.200 395.952 2.596 RCIND SIZE BLOCK SEGMT DIVR2 AGE Log(MCAP2) GEAR2 EQED RISK2 R2 Adjusted R 2 138 Probability (F-Statistic) 0.002 0.000 0.000 0.005 Durbin-Watson 2.033 2.073 1.993 1.938 * Significance at the 5% level ** Significance at the 1% level An analysis of chairman independence and firm performance is presented in Table 7.9. The independent variables account for 10.50%-94.40% of the variation in performance measures. The table suggests that chairman independence does not have significant effect on ROA, ROE, shareholder return and Tobin’s q. 139 Table 7.9 OLS Regressions: Chairman Independence and Subsequent Performance Sample Period: 2003-2006 Included Observations: 243 Coefficient t-Statistic ROA2 ROE2 SHERT2 TOBQ2 Intercept -0.287 0.081 -0.585 2.584 -1.477 0.154 -6.956** 2.276* -0.127 -0.242 0.004 0.710 -1.743 -1.216 0.114 1.664 -0.004 -0.072 -0.017 -0.344 -0.177 -1.110 -1.669 -2.471* -0.043 -0.991 0.110 0.615 -0.203 -1.703 1.192 0.493 -0.004 0.097 -0.013 -0.085 -0.201 1.813 -1.574 -0.746 0.207 0.544 0.070 -1.079 2.527* 2.434* 1.971* -2.254* 0.0008 -0.003 0.0002 -0.012 0.348 -0.535 0.193 -0.965 0.058 0.140 0.109 0.329 1.920 1.694 8.328** 1.854 -0.018 -0.701 -0.016 -0.022 -1.018 -14.421** -2.137* -0.213 -0.370 0.068 -0.128 3.597 -1.670 0.112 -1.332 2.784** -0.0004 0.052 0.578 -0.063 -0.018 0.859 60.301** -0.485 0.105 0.494 0.944 0.106 0.066 0.472 0.942 0.068 Std Error (Regression) 0.560 1.527 0.242 3.272 F-Statistic 2.711 22.633 394.069 2.754 CMIND SIZE BLOCK SEGMT DIVR2 AGE Log(MCAP2) GEAR2 EQED RISK2 R2 Adjusted R 2 140 Probability (F-Statistic) 0.004 0.000 0.000 0.003 Durbin-Watson 2.047 2.033 2.013 1.933 * Significance at the 5% level ** Significance at the 1% level Regarding the control variables used in the analysis, some consistent patterns emerge from the above tables. It appears that larger board or lower managerial shareholdings could lead to poor performance as measured by Tobin’s q. During the test period, dividend payments of sample firms reflect the accounting performance measures of ROA and ROE, however, diverge from the market measure of Tobin’s q. In general, larger firms or firms with lower leverage have better shareholder return. The regression results are assessed by additional tests without firm size control, which are reported in Appendix 10, 11, 12, 13 and 14. The tests discern a positive effect of full board independence on shareholder return; the negative correlation between remuneration committee independence and accounting performance measures, however, disappears. 7.4 Regressions: Board Independence and Firm Risk In the models to test H 4 and H 5 , the dependent variable is firm risk; the independent factors include board characteristics, firm performance and other control variables. Yi 1 ( Independence) i 2 ( BoardSize) i 3 ( BlockholderOwnership) i 4 ( Diversification) i 5 ( DividendPayout ) i 6 ( FirmAge) i 7 ( FirmSize) i 8 ( Leverage) i 9 ( ManagerialOwnership ) i 10 ( Performance) i i Where, for the i th company Y = RISK2 = Constant of the equation = Coefficient of the variable Independence = FIND, ACIND, NCIND, RCIND or CMIND Board Size = SIZE Blockholder Ownership = BLOCK Diversification = SEGMT Dividend Payout = DIVR2 Firm Age = AGE 141 Firm Size = Log(MCAP2) Leverage = GEAR2 Managerial Ownership = EQED Performance = SHERT2 = Error term As shown in Table 7.10, the models explain 94.50% of the variance in firm risk; no significant relationship is identified between firm risk and measures of board independence. According to the regressions, firm risk may be jointly determined by dividend payout, firm size, leverage and shareholder return, i.e., smaller companies, companies with higher gearing or shareholder return, or companies with lower dividend payout tend to be riskier. 142 Table 7.10 OLS Regressions: Board Independence and Firm Risk Sample Period: 2003-2006 Included Observations: 243 Coefficient t-Statistic Intercept FIND RISK2 1.072 1.027 1.007 0.981 1.001 7.327** 7.172** 7.354** 7.140** 7.131** -0.168 1.354 ACIND -0.053 -0.641 NCIND 0.062 0.703 RCIND 0.085 1.073 CMIND -0.003 -0.066 SIZE BLOCK SEGMT DIVR2 AGE Log(MCAP2) GEAR2 EQED 0.026 0.027 0.025 0.023 0.026 1.486 1.572 1.413 1.322 1.491 -0.123 -0.098 -0.068 -0.043 -0.079 -0.783 -0.623 -0.442 -0.274 -0.507 0.021 0.020 0.019 0.020 0.020 1.476 1.401 1.369 1.428 1.410 -0.146 -0.144 -0.146 -0.150 -0.147 -2.482* -2.438* -2.471* -2.540* -2.478* -3.23E-05 -7.21E-05 -0.0001 -9.35E-05 -0.0001 -0.020 -0.045 -0.077 -0.059 -0.078 -0.180 -0.184 -0.187 -0.188 -0.184 -8.182** -8.386** -8.398** -8.499** -8.375** 0.029 0.028 0.028 0.029 0.029 2.216* 2.203* 2.189* 2.271* 2.223* 0.117 0.120 0.122 0.114 0.120 0.728 0.746 0.758 0.707 0.744 143 SHERT2 1.626 1.626 1.627 1.626 1.626 60.540** 60.353** 60.326** 60.451** 60.301** 0.945 0.945 0.945 0.945 0.945 0.943 0.943 0.943 0.943 0.943 Std Error (Regression) 0.405 0.406 0.406 0.405 0.406 F-Statistic 402.324 399.738 399.891 401.084 398.998 Probability (F-Statistic) 0.000 0.000 0.000 0.000 0.000 Durbin-Watson 1.994 1.998 1.995 1.973 R2 Adjusted R * 2 Significance at the 5% level ** Significance at the 1% level Robustness tests without firm size control are conducted; the results, which are shown in Appendix 15, discern a negative contribution of full board independence on firm risk; it seems that the role of firm size in reducing risk is replaced by board size. Moreover, without firm size control managerial shareholdings display a positive effect on risk. 7.5 Summary To investigate the specific influence of firm performance on board independence, OLS and logit regressions in which board independence serves as the dependent variable, and firm performance and some other factors serve as the explanatory variables, were carried out. It is found that companies with higher blockholder shareholdings tend to reduce the percentages of independent directors on the board, and audit and remuneration committees; larger firms have relatively more independent members sitting on nomination and remuneration committees. Although poor performance in terms of ROA and shareholder return may lead to a higher proportion of independent directors on remuneration committee, the regressions indicate that, past performance, as measured by ROA, ROE, shareholder return and Tobin’s q, have no significant effect on the level of independence on the board, audit committee and nomination committees, and the decision to appoint an independent chairman. To search for the potential impact of board independence on firm performance, OLS models, in which performance measures are used as the dependent variables, and the independent variables consist of board independence and other controls, were tested. 144 The resulting evidence looks ambiguous; full board independence gives a positive contribution on Tobin’s q, and a negative contribution on ROE. Higher level of independence on audit and remuneration committees may lower accounting performance; chairman independence and nomination committee independence, however, do not affect subsequent performance. Additional findings include that larger board or lower managerial shareholdings could lead to poor performance as measured by Tobin’s q. Dividend payments of sample firms reflect the accounting performance measures of ROA and ROE, however, diverge from the market measure of Tobin’s q. Moreover, during the test period larger firms and firms with lower leverage have better shareholder return. According to the regressions specified for board independence and other variables on risk, there is no significant relationship between board independence and subsequent firm risk; smaller companies, companies with higher gearing or shareholder return, or companies with lower dividend payout tend to be riskier. The results of sensitivity tests without firm size control are disclosed in the appendices, which, along with the findings as summarised above, are discussed further in Chapter 8, to address each of the five hypotheses, and the general research questions raised earlier in this thesis. 145 Chapter 8. Discussion and Conclusions 8.1 Introduction This thesis attempts to test the applicability of several theories which make different predictions about the effect of board independence on firm performance and vice versa; unlike the prior studies which do not ask whether board characteristics are endogenously related to performance, two research questions are raised. Does board independence have any influence on firm performance among Australian listed companies? Does firm performance have any influence on board independence among Australian listed companies? In this final chapter the results emanating from the data analysis are discussed; the discussion responds specifically to the above questions. There are five sections in the chapter; the next section provides an examination of the key findings presented in Chapter 6 and 7, which leads to the conclusions and recommendations in Section 3. Limitations in the current study and possible areas for future research are introduced in Section 4, followed by a summary in the last section. 8.2 Discussion of Findings In Chapter 4, regarding the potential links between board independence, corporate performance and risk, five testable hypotheses were developed from agency theory, stewardship theory and organizational portfolio model. This section summarises and discusses the results for hypothesis tests as reported in the preceding chapters. 8.2.1 Board Independence and Past Performance The correlation analysis for the period 2000-2003 indicates that the average shareholder return in the past three years is inversely related to remuneration committee independence; there is no association between other measures of performance and board independence. Similarly, the regression models locate a negative impact of ROA and shareholder return on remuneration committee independence; past performance, as measured by ROA, ROE, shareholder return and Tobin’s q, does not influence full board independence, audit and nomination committee independence, and chairman independence. 146 Since the explanatory power of the regression models, as displayed in Tables 7.1-7.4, is quite low, and the link between ROA and remuneration committee independence becomes insignificant in the sensitivity tests, the results may be too weak to endorse H 1 (There is a negative relationship between board independence and past firm performance). There are several papers which test the effect of prior performance on board independence. As shown in Table 8.1, the conclusion reached here is consistent with the evidence in the Australian paper of Lawrence and Stapledon (1999), in which the authors ran regressions for performance measures during the 1985-1995 periods on board composition in mid-1995. Table 8.1 Relationship between Board Independence and Past Performance Papers Country Results Hermalin and Weisbach (1988) U.S. Negative Pearce II and Zahra (1992) U.S. Negative Lawrence and Stapledon (1999) Australia Insignificant Denis and Sarin (1999) U.S. Positive Bhagat and Black (2000) U.S. Negative Panasian et al (2003) Canada Negative Current Study Australia Insignificant 8.2.2 Board Independence and Subsequent Performance The Pearson correlations for the period 2003-2006 show that full board independence, monitoring committee independence and chairman independence are not associated with subsequent performance. The OLS regressions yield ambiguous findings - full board independence gives a positive contribution on Tobin’s q, and a negative contribution on ROE, probably due to the fact that the performance variables in this research fall into two distinct groups – accounting measures and market-based measures. The findings indicate that the Australian market may tend to reward companies with relatively more independent board members even there is no corresponding improvement in operating result. As defined in Table 5.1, Tobin’s q is 147 calculated by dividing market value by book value of total assets. Since this measure is related to the value placed on the firm by the market, it seems that, in the context of the global movement to promote board independence, shareholders may expect that independent directors would enhance the value of their investments. Although higher levels of independence on audit and remuneration committees may reduce accounting performance, chairman independence and nomination committee independence do not have significant effect on performance. Additional tests without firm size control discern a positive effect of full board independence on shareholder return; the negative influence of remuneration committee independence on accounting performance measures, however, disappears. Thus it could be concluded that the data analysis does not give a clear support to H 2 (There is a negative relationship between board independence and subsequent firm performance) or H 3 (There is positive relationship between board independence and subsequent firm performance). The results, and the evidence provided by prior Australian research as summarised in Table 8.2, suggest that independent directors may not add value to Australian public corporations as expected by the ASX Corporate Governance Council. The possible explanations for the findings are explored in Section 8.3. Table 8.2 Relationship between Board Independence and Subsequent Performance Papers Country Results Muth and Donaldson (1998) Australia Negative Calleja (1999) Australia Insignificant Lawrence and Stapledon (1999) Australia Negative Cotter and Silverster (2003) Australia Insignificant Kiel and Nicholson (2003) Australia Negative Bonn et al (2004) Australia Positive Balatbat et al (2004) Australia Insignificant Hutchinson and Gul (2004) Australia Insignificant Current Study Australia Insignificant 148 8.2.3 Board Independence and Firm Risk According to the correlation analysis for 2003-2006 and regressions for the effect of board independence and other variables on firm risk, full board independence, monitoring committee independence and chairman independence are not related to subsequent risk. H 4 (There is a negative relationship between board independence and subsequent firm risk) and H 5 (There is a positive relationship between board independence and subsequent firm risk) are therefore rejected. As introduced in Chapter 3, Schellenger et al (1989) believed that the conflicting evidence with respect to the existence or non-existence of a board composition effect on financial performance might be due to failure to control risk; after testing a sample of 526 U.S. firms with complete data for the year 1986, they found that the percentage of outsiders and standard deviation of returns were negatively correlated. They asserted that their findings provided support for advocates of non-executive representation on the boards of directors. There are two possible reasons why the results in this study are inconsistent with those in Schellenger et al (1989). First, in this research board composition is measured by the percentages of independent directors, rather than NEDs, on the board and monitoring committees. Secondly, as clear in Appendix 2, there may not be enough control variables in Schellenger et al (1989); similar to the findings in Schellenger et al (1989), it is reported in Chapter 7 that the tests without firm size control identify a negative contribution of full board independence on firm risk. 8.2.4 Summary of Other Findings The data analysis suggests that companies with higher blockholder shareholdings tend to reduce the percentages of independent directors on the board, and audit and remuneration committees. The findings are consistent with the conclusion in the Australian study of Cotter and Silverster (2003) that an absence of substantial shareholders is compensated for with greater board independence, and support the view that firms may choose among a number of different governance mechanisms in order to create the appropriate structure for themselves, given the environment in which they operate (Coles et al, 2001). 149 It is presumed that blockholders have the capacity to monitor their investments and, by virtue of the magnitude of their investments, can affect managerial behaviour; the threat that blockholders will sell large blocks of shares if the firm fails to provide an acceptable return, or is not responsive to governance concerns that investors view as critical, is a significant issue for managers (Coles et al, 2001). There is evidence that institutional investors and other blockholders do impact managerial behaviour (e.g., Barclay and Holderness, 1991; Van Nuys, 1993; Brickley, Lease and Smith, 1994; Shome and Singh, 1995; Bethel, Liebeskind and Opler, 1998; Allen and Phillips, 2000). The data analysis indicates that larger firms have relatively more independent directors sitting on nomination and remuneration committees. One possible interpretation is that larger firms may draw more attention from media, shareholders and other stakeholders; consequently they restructure their committees to be more independent, in fear of the criticisms of interest conflicts and excesses, in accordance with the recent demand for greater accountability (Vafeas and Theodorou, 1998; Cotter and Silverster, 2003). With respect to the determinants of firm performance, it is reported that smaller board or higher managerial shareholdings could lead to better performance as measured by Tobin’s q. For the effect of board size on performance, the literature survey in Finegold, Benson and Hecht (2007) shows that the empirical evidence is inconclusive, although the general consensus is that smaller boards are more effective at monitoring performance (Coles, Naveen and Lalitha, 2008). The argument is that smaller groups are more cohesive and more productive, while larger groups suffer from the problems such as social loafing and higher co-ordination costs (Lipton and Lorsch, 1992; Jensen, 1993; Yermack, 1996). Similarly, although scholars have frequently tested the impact of executive ownership on firm performance, the evidence is mixed (Sundaramurthy, Rhoades and Rechner, 2005). Jensen and Meckling (1976) proposed that increasing managerial ownership could mitigate agency conflicts - the higher the proportion of equity owned by managers, the greater the alignment between managers and shareholder interests; the studies supporting their view include Morck et al (1988), Kim et al (1988) and Hudson et al (1992). Some authors, for example Tsetsekos and DeFusco (1990) and Sundaramurthy et al (2005), could not locate any significant relationship between managerial shareholdings and performance. There are a number of papers, for example, 150 McConnell and Servaes (1990) and Brailsford, Oliver and Pua (2002), which identify a non-linear relationship. It is not surprising that dividend payments of sample firms reflect the accounting measures of ROA and ROE, however, diverge from the market measure of Tobin’s q, taking into account that in the correlation analysis the performance variables fall into two distinct clusters. As discussed in Chapter 5, accounting measures are historical and therefore experience a backward and inward looking focus; market-based measures are forward looking indicators that reflect current plans and strategies, and in theory represent the discounted present value of future cash flows (Devinney et al, 2005; Fisher and McGowan, 1983). It is therefore concluded that in Australia dividend payout is based on the historical performance, rather than the market expectation. Moreover, smaller companies, companies with higher gearing or shareholder return, or companies with lower dividend payout appear to be riskier. The findings, in general, are in line with the finance literature (e.g., Brigham, 1985; Reilly, 1985; Ross and Westerfield, 1988; Ross, Westerfield and Jaffe, 2005; Reilly and Brown, 2006). For the test period of 2003-2006, larger firms or firms with lower leverage have better shareholder return; therefore in 2003-2006 the Australian shareholders tended to value larger companies and companies with lower leverage. It is noted that the negative consequence of leverage coincides with the results of Alaganar (2004) who examined the top ASX 100 companies from 1994 to 2003; according to the author, one possible explanation is that newly acquired debt may be deployed on projects that have a negative impact on shareholder wealth. This may have been fuelled by the prevailing low interest rate environment where firms were inclined to undertake such projects (Alaganar, 2004). 8.3 Conclusions and Recommendations As discussed in the last section, regarding the relationships between board independence, corporate performance and risk, the data analysis, in general, does not support the five research hypotheses as proposed by agency theory, stewardship theory and organizational portfolio theory. The relevant findings are summarised in Table 8.3. 151 Table 8.3 Results: Board Independence, Firm Performance and Firm Risk Sample Period: 2000-2006 Included Observations: 243 Relation FIND ACIND NCIND RCIND CMIND ROA Insignificant Insignificant Insignificant Negative* Insignificant ROE Insignificant Insignificant Insignificant Insignificant Insignificant Shareholder Return Insignificant Insignificant Insignificant Negative* Insignificant Tobin’s q Insignificant Insignificant Insignificant Insignificant Insignificant ROA Insignificant Insignificant Insignificant Negative* Insignificant ROE Negative* Negative* Insignificant Negative* Insignificant Shareholder Return Insignificant Insignificant Insignificant Insignificant Insignificant Tobin’s q Positive* Insignificant Insignificant Insignificant Insignificant Insignificant Insignificant Insignificant Insignificant Past Performance Subsequent Performance Subsequent Risk Standard Deviation of Return Insignificant * Significance at the 5% level As described in Table 6.1, FIND, ACIND, NCIND and RCIND represent the percentage of independent directors on the full board, audit committee, nomination committee or remuneration committee respectively; CMIND is a binary variable to assess whether or not the chairperson is an independent director. With respect to the general research questions raised earlier, the above results, together with those of the Australian studies as shown in Appendix 1, suggest that, for Australian listed companies, there does not appear to be a strong relationship between board independence, and past or subsequent performance. The literature gives some possible reasons why the predictive power of agency theory, stewardship theory and organizational portfolio model, for the board independencefinancial performance link, is so limited. Based on the institutional theory which has been used to deal with the rationale behind the emergence of practices without obvious economic value (Myer and Rowan, 1977), Peng (2004) argued that appointing outside 152 directors to the board might merely represent firms’ attempts to comply with institutional pressures, and therefore might not be necessarily linked to firm performance. A core assumption of institutional theory is that organizations would act to protect or enhance their legitimacy; copying other reputable organizations, even without knowing the direct benefits of doing so, may be a low cost strategy to gain legitimacy (Peng, 2004). Emerging practices, e.g., total quality management or board independence, are generally regarded as state-of-the-art techniques (Westphal, Gulati and Shortell, 1997); jumping on such a “bandwagon” may be perceived “as a form of innovation when it is contrasted with the more passive act of ignoring industry trends or the more active stance of rejecting them altogether” (Staw and Epstein, 2000, p.528). Although firms may comply with the institutional demands for more outside directors so that they would not be noticed as different and consequently singled out for criticism, they could still employ a number of tactics to neutralize the power of outsiders (Oliver, 1991; Zajac and Westphal, 1996; Westphal, 1999). This can be done, for example, by appointing individuals who are demographically similar and therefore more sympathetic to executives (Westphal and Zajac, 1995), or individuals with experience on other passive boards instead of more active boards (Zajac and Westphal, 1996), or individuals who are from strategically irrelevant backgrounds without the knowledge base to effectively participate in strategic decision making and challenge executives’ power (Carpenter and Westphal, 2001). The rising number of external members on the board may therefore “occur as the result of processes that make organizations more similar without necessarily making them more efficient” (DiMaggio and Powell, 1983, p.147). According to DiMaggio and Powell (1983), the concept that best captures the process of homogenization is isomorphism, which is a constraining process that forces one unit in a population to resemble other units that face the same set of environmental conditions; they identified three mechanisms through which institutional isomorphic change could occur, each with its own antecedents: Coercive isomorphism that stems from political influence and legitimacy; Mimetic isomorphism resulting from standard responses to uncertainty; and Normative isomorphism associated with professionalization. 153 It appears that coercive isomorphism may play an important role behind the rising number of independent directors on the board; coercive isomorphism results from both formal and informal pressures exerted on organizations by other organizations upon which they are dependent, in this case the stock exchanges, and by cultural expectations in the society in which organizations function. Although the ASX follows a voluntary approach to promote board independence, in which “[i]f a company considers that a recommendation is inappropriate to its particular circumstances, it has the flexibility not to adopt it – a flexibility tempered by the requirement to explain why” (Guidelines, 2003, p.5), it is noted that the pressures for isomorphism could be felt as persuasion, or as invitations to join in collusion, as well as force (DiMaggio and Powell, 1983). Mimetic isomorphism may also help us understand some of the dynamics in appointing outsiders on corporate boards. Uncertainty is a powerful force that encourages imitation; when an organization faces a problem with ambiguous causes or unclear solutions, it may model itself after similar organizations that it perceives to be more legitimate or successful (DiMaggio and Powell, 1983). As discussed in Chapter 2, although corporate governance has become a prominent topic in recent years, there are significant disagreements in the field of corporate governance research (Murphy and Topyan, 2005; Gillan, 2006). In Pettigrew (1992), for example, it is noted that corporate governance lacks any form of coherence, either empirically, methodologically or theoretically, with only piecemeal attempts to understand and explain how the modern corporation is run. Tricker (2000) contended that corporate governance did not have an accepted theoretical base or commonly accepted paradigm, and the term “corporate governance” was scarcely used until 1980. Murphy and Topyan (2005) maintained that researchers investigated corporate governance less as a planned, systematic inquiry, and more as a response to observed problems in corporations. In the situation some firms may decide to appoint independent outsiders to their boards, as a response to the uncertainty in corporate governance issues. Regarding the relationship between board characteristics and firm risk, as shown earlier in Chapter 4, some agency theorists asserted that conflicts relating to managerial risk aversion may arise because of portfolio diversification constraints of managerial income; the conflicts may be heightened when executive compensation is composed largely of a fixed salary, or where their specific skills are difficult to transfer from one 154 company to another (Fama, 1980; Knoeber, 1986; Baysinger and Hoskisson, 1990; Prentice, 1993; Vafeas and Theodorou, 1998; Coles et al, 2001; Godfrey et al, 2003). Risk increasing investment decisions may increase the likelihood of bankruptcy; such a corporate event could severely damage a manager’s reputation, making it difficult to find alternative employment (Jensen, 1986; McColgan, 2001). Managerial risk aversion may also affect the financial policy of the firm. Higher debt is believed to reduce agency conflicts and carries potentially valuable tax shields (Jensen, 1986; Haugen and Sendbet, 1986); managers, however, may prefer equity financing because debt increases the risk of bankruptcy and default (Brennan, 1995). It is corporate governance mechanisms, including board of directors due to its presumed independence, that would harmonize these agency conflicts and safeguard invested capital. The board could ensure that managers are not the sole evaluators of their own performance, and the board’s legal responsibilities to hire, fire, and reward executives are seen as key elements in controlling conflicts of interest (Fama and Jensen, 1983; Williamson, 1984; Baysinger and Hoskisson, 1990; McColgan, 2001). However, it is found in this research that the level of board independence does not affect subsequent firm risk. The results indicate that there may not be any significant difference between the risk preferences of independent directors and executives, or, the concern over agency conflicts relating to managerial risk aversion may not be justified. According to Heslin and Donaldson (1999), since independent directors are often appointed in order to curb the ostensibly radical excesses of management, they may tend to be risk-averse; the pressure of strong public criticism and threat of legal action for failure in their fiduciary responsibility may reinforce the risk-aversion of outside directors (Davis and Thompson, 1994). Baysinger et al (1991) demonstrated that executives would be more likely to approve risky proposals such as increasing expenditures on R&D; this probably reflects their intimate knowledge of the business and resulting confidence that anticipated benefits would flow from their proposed investments (Lorsch and MacIver, 1989). In contrast, external board members may have less first-hand familiarity with the business; they tend to emphasise those factors that are relatively certain, e.g., costs, while being more 155 sceptical about less certain factors, e.g., the chance of anticipated benefits being realised (Baysinger et al, 1991). In addition, Hill and Snell (1988), Baysinger et al (1991) and Westphal (1999) reported that outside directors would lead to increased diversification which could lower the level of firm risk. Chandler (1994) documented that non-executives might be more reluctant to make the R&D investments in capital intensive, technologically complex industries. Daily and Dalton (1994) concluded that outside directors would tend to prevent performance from dropping to the level that may cause bankruptcy. Appointing independent directors to corporate boards could become a widespread practice in Australia, following the release of the ASX Corporate Governance Guidelines in 2003, in which the monitoring role of independent directors hypothesized by agency theory is endorsed. The evidence offered in this dissertation, however, casts doubts on the hope that promoting board independence would improve corporate performance, and calls for more attention on the actual roles played by independent directors in public companies. As introduced in Chapter 2, claims are often made that the Australian market is an outsider system of corporate governance (Scott, 1997; Weimar and Paper, 1999; Bradley et al, 1999; Campbell, 2002), in which the main concern is the agency conflicts between strong managers and weak dispersed shareholders. However, it is found in Dignam and Galanis (2004) that the Australian listed market is characterized by: significant blockholders engaged in private rent extraction; institutional investor powerlessness; a strong relationship between management and blockholders, which results in a weak market for corporate control; and a historic weakness in public and private securities regulation, which allows the creation and perpetuation of crucial blocks to information flow. These characteristics suggest that Australia may have been misclassified as an outsider system; rather, it may tend towards an insider system. Dignam and Galanis (2004, p.651) commented that “… a central assumption of Australian’s recent reform process – that the reform initiatives from the UK and the US should be adopted in Australia – may be incorrect”. Therefore, the regulatory bodies in Australia and other countries should 156 be mindful of the differences between the markets when they look for the solutions to their corporate governance issues. Moreover, some academics, for example, Kole (1995), Himmelberg, Hubbard and Palia (1999) and McColgan (2001), noted that agency conflicts might be heterogeneous across different firms in different industries and cultures. Himmelberg et al (1999) referred to differing firms with different contracting environments, which refreshes an important point from Jensen and Meckling (1976) original theory that no two firms would have the same “nexus of contacts”. The scope of agency conflicts may differ from one company to another, as would the effectiveness of governance mechanisms in reducing them; thus what is required is a more detailed understanding of what makes these mechanisms important for some companies and ineffective for others (McColgan, 2001). Tipgos (2007) confirmed that agency theory was the most influential basis for the recent development in corporate governance; he then questioned the foundation of the theory, the principal-agent assumption of shareholders and managers. It is illustrated that shareholders are not owners/principals of the modern public corporations; they are investors whose interests are defined by their risk-return models. Consequently agency theory may be inadequate to describe the complex relationships inherent in the public corporations today. It could be argued that some types of independent directors may be valuable, while others may not (Chan and Li, 2008); the argument, however, would lead to the conclusion that to push for greater board independence may be fruitless, unless the independent directors have some particular attributes, which are currently unclear, other than their independence from management. Therefore, for policy-makers, practitioners and scholars in Australia and elsewhere, the findings reported here suggest that, despite agency theory’s theoretical logic and policy influence, whether certain corporate governance practices recommended by the perspective would improve corporate performance should be empirically tested. 8.4 Limitations and Future Research Although this thesis tries to address some of the limitations identified in prior research, such as small sample size, short-term observation of firm performance, limited 157 performance measures and control variables, and simplistic dichotomy of inside and outside directors as an empirical proxy for board independence, it is still subject to a number of limitations. For example, as reliance is placed on the disclosures within the corporate annual reports, certain weaknesses are evidenced in the collection of data on board members. The ASX Corporate Governance Council (2003, p.19) acknowledged that “[a]n independent director is independent of management and free of any business or other relationship that could materially interfere with – or could reasonably be perceived to materially interfere with – the exercise of their unfettered and independent judgement.” This statement notes that there may be other relationships, e.g., friendship and demographically similarity as explored in Westphal and Zajac (1995), which might compromise the independence of the directors. Whilst the possible existence of these impediments to independence may be present and may limit the assessment of independence in this study, there is a lack of available data to identify such a relationship. Moreover, the use of three-year averages may hide the richness of raw data on performance. The weak results achieved on the tests of hypotheses may indicate that corporate governance mechanisms may have a complementary or substitution effect on each other. As argued by Larcker et al (2007), the inconsistent results for the association between typical measures of corporate governance and performance outcomes may be partially attributed to the difficulty in generating reliable and valid measures for the complex construct that is termed “corporate governance”. So the use of a composite corporate governance index or board independence index may provide stronger statistical relationships with performance than modelling these mechanisms separately. Future research could therefore consider running other methods of analysis, such as confirmatory factor analysis, which have been used by some academics in the U.S. (e.g., Molz, 1988; Larcker et al, 2007), and hierarchical and step-wise regressions. Cho (1998), Hermalin and Weisbach (1998), Himmelberg et al (1999) and Dafinone (2001) noted that corporate governance involved complex interrelated mechanisms, such as board composition, dividends, blockholder and managerial shareholdings, and leverage; this study provides some insights on this topic, e.g., the correlation between board composition and blockholder ownership. Further research may investigate the 158 substitute and complementary effects of corporate governance variables on performance, perhaps with the use of structural equation modelling. Although Azim and Shailer (2006) found some inconsistent evidence for the effects between board monitoring, auditor monitoring and shareholder monitoring, in Australia the relevant literature is still limited. The ASX Guidelines highlights the need to apply recommendations regarding board composition and structure flexibly. It is unclear whether merely complying with the recommendations would improve accountability; to address this issue future research may consider a survey of shareholders, regulation bodies and other stakeholders. It is also recommended to conduct a survey of directors’ perceptions of the extent to which they believe the recommendations could improve board effectiveness and accountability. Another interesting question to be investigated is who would be risk averse, independent directors or executives. The evidence offered in this project suggests that there may not be any significant difference between their risk preferences, or, managers may not be as risk averse as believed by some agency theorists. Currently there appears to be very little empirical work carried out to deal with this issue, which could involve an attitudinal survey of independent directors and managers. 8.5 Summary To address the general research questions raised earlier in the thesis, the univariate and multivariate analyses in relation to board independence, firm performance and risk are reviewed. It is concluded that the results do not support the hypotheses developed from agency theory, stewardship theory and organizational portfolio model. For Australian listed companies, there does not appear to be a strong relationship between board independence, and past or subsequent firm performance. The literature provides some possible reasons why the explanatory power of these theories with respect to the board independence-financial performance link is so limited. Some authors argued that appointing outside directors to the board might merely represent firms’ attempts to comply with institutional pressures, and therefore might not be necessarily linked to performance; although firms may comply with the demands for 159 more outside directors, they could employ a number of tactics to neutralize the power of outsiders. It is discovered that independent directors, in general, do not have any significant influence on the level of performance risk. The potential explanations for the findings include that independent directors may not be significantly different from executives in terms of risk preference, or, the concern over managerial risk aversion may not be justified. After the ASX Corporate Governance Council introduced its Guidelines in 2003, appointing independent directors to corporate boards may have become a popular practice. This study casts doubts on the hope that promoting board independence would improve performance; it is recommended that whether certain corporate governance practices developed from the agency perspective would lead to better performance may need to be tested. The limitations of this study and future research opportunities are also discussed. 160 References Agrawal, A. and Knoeber, C. R, 1996. Firm Performance and Mechanisms to Control Agency Problems between Managers and Shareholders. Journal of Financial and Quantitative Analysis 31(3): 377-397. Agrawal, A. and Knoeber, C. R, 1999. Outside Directors, Politics, and Firm Performance. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=85310. Aguilera, R. V, 2005. Corporate Governance and Director Accountability: An Institutional Comparative Perspective. British Journal of Management 16: 39-53. AICD, AEOA and ASA, 2007. Executive Equity Plan Guidelines. http://www.companydirectors.com.au/Policy/Policies+And+Papers/2007/Executive+Eq uity+Plan+Guidelines.htm. Alaganar, V. T, 2004. Australian Corporate Debt: Neither Risky nor Rewarding. http://www.ssga.com/library/resh/Australiancorporatedebt20041025/page.html. Allen, J. W. and Phillips, G. M, 2000. Corporate Equity Ownership, Strategic Alliance, and Product Market Relationships. Journal of Finance 55: 2791-2816. Alshimmiri, T, 2004. Board Composition, Executive Remuneration, and Corporate Performance: The Case of REITs. Corporate Ownership & Control 2(1): 104-118. Anderson, R. C. and Reeb, D. M, 2004. Board Composition: Balancing Family Influence in S & P 500 Firms. Administrative Science Quarterly 49(2): 209-237. Andrews, K. R, 1986. The Concept of Corporate Strategy. Homewood, Irwin. Aoki, M, 1990. Toward an Economic Model of the Japanese Firm. Journal of Economic Literature 28(1): 1-27. Aoki, M, 2001. Toward a Comparative Institutional Analysis. MIT Press, Cambridge, Massachusetts. Argyris, C, 1964. Integrating the Individual and the Corporation. Wiley and Sons, New York. 161 Arthur, N, 2001. Board Composition as the Outcome of an Internal Bargaining Process: Empirical Evidence. Journal of Corporate Finance 7(3): 307-340. Aspect Huntley, 2003. Huntleys’ Shareholder. Wrightbooks, Queensland. ASX, 2003. Principles of Good Corporate Governance and Best Practice Recommendations (Guidelines). ASX, 2005. Analysis of Corporate Governance Practices Reported in 2004 Annual Reports. ASX, 2006. ASX Corporate Governance Council Principles of Good Corporate and Best Practice Recommendations Frequently Asked Questions. http://www.asx.com.au/about/pdf/FAQsCorporateGovernance.pdf. ASX, 2007. Corporate Governance Principles and Recommendations (Guidelines). ASX, 2008. Analysis of Corporate Governance Disclosures in 2007 Annual Reports. http://www.asx.com.au/about/pdf/analysis_of_cg_in_2007_annual_reports.pdf. Azim, M. and Shailer, G, 2006. Substitute and Complementary Effects of Alternate Monitoring Mechanisms on Firm Performance. 2006 AFAANZ Conference, Wellington, 2-4 July 2006. Bacon, J, 1973. Corporate Directorship Practices: Membership and Committees of the Board. Conference Board, New York. Bacon, J. and Brown, K, 1975. Corporate Directorship Practices: Role, Selection and Legal Status of the Board. Conference Board, New York. Baird, I. S. and Thomas, H, 1990. What is Risk anyway? in Bettis, R. A. and Thomas, H. (Ed.), Risk, Strategy and management, JAI Press, Greenwich, CT: 21-52. Balatbat, M. C. A., Taylor, S. L. and Walter, T. S, 2004. Corporate Governance, Insider Ownership and Operating Performance of Australian Initial Public Offerings. Accounting and Finance 44(3): 299-328. Barclay, M. and Holderness, C, 1991. Control of Corporations by Active Block Investors. Journal of Applied Corporate Finance 4: 68-77. 162 Barnhart, S. W. and Rosenstein, S, 1998. Board Composition, Managerial Ownership and Firm Performance: An Empirical Analysis. Financial Review 33(4): 1-16. Bathala, C. T. and Rao, R. P, 1995. The Determinants of Board Composition: An Agency Theory Perspective. Managerial and Decision Economics 16(1): 59-69. Baysinger, B. D. and Butler, H. N, 1985. Corporate Governance and the Board of Directors: Performance Effects of Changes in Board Composition. Journal of Law, Economics and Organization 1(1): 101-124. Baysinger, B. D. and Hoskisson, R. E, 1990. The Composition of Board of Directors and Strategic Control: Effects on Corporate Strategy. Academy of Management Review 15: 72-87. Baysinger, B. D., Kosnik, R. T. and Turk, T. A, 1991. Effects of Board and Ownership Structure on Corporate R&D Strategy. Academy of Management Journal 34(1): 205214. Beasley, M. S, 1996. An Empirical Analysis of the Relation between Board of Directors Compensation and Financial Statement Fraud. Accounting Review 71(4): 443-465. Beasley, M. S. and Petroni, K, 1998. Board Independence and Audit Firm Type. Department of Accounting, North Carolina State University. Beatty, R. P. and Zajac, E. J, 1994. Managerial Incentives, Monitoring, and Risk Bearing: A Study of Executive Compensation, Ownership, and Board Structure in Initial Public Offerings. Administrative Science Quarterly 39: 313-335. Bebchuk, L. A, 1999. A Rent-Protection Theory of Corporate Ownership and Control. NBER, Cambridge, Massachusetts. Bebchuk, L. A, 2000. Using Options to Divide Value in Corporate Bankruptcy. European Economics Review 44: 829-843. Bebchuk, L. A., Coates, J. and Subramanian, G, 2002. The Powerful Anti-Takeover Force of Staggered Boards: Theory, Evidence and Policy. Stanford Law Review 54(5): 887-951. 163 Becht, M., Bolton, P. and Roell, A, 2005. Corporate Governance and Control. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=343461. Berger, P. and Ofek, E, 1996. Bustup Takeovers of Value-Destroying Firms. Journal of Finance 51: 1175-1200. Berle, A. A. and Means, G. C, 1932. The Modern Corporation and Private Property. MacMillan, New York. Bethel, J., Liebeskind, J. and Opler, T, 1998. Block Share Repurchases and Corporate Performance. Journal of Finance 53: 605-634. Bhagat, S. and Black, B, 1999. The Uncertain Relationship between Board Composition and Firm Performance. Business Lawyer 54(3): 921-963. Bhagat, S. and Black, B, 2000. Boar Independence and Long-Term Firm Performance. Graduate School of Business, University of Colorado. Bhagat, S., Dennis C. C. and Elson, C, M, 1999. Director Ownership, Corporate Performance, and Management Turnover. Business Lawyer 54(3): 885-920. Black, B. S, 1990. Shareholder Passivity Reexamined. Michigan Law Review 89(3): 520-608. Bonn, I., Yoshikawa, T. and Phan, P, H, 2004. Effects of Board Structure on Firm Performance: A Comparison between Japan and Australia. Asian Business & Management 2004(3): 105-125. Borch, O. J. and Huse, M, 1993. Informal Strategic Networks and Boards of Directors. Entrepreneurship: Theory and Practice 18 (1): 23-36. Borokhovich, K. A., Parrino, R. and Trapani, T, 1996. Outside Directors and CEO Selection. Journal of Quantitative Analysis 31(3): 337-355. Bouy, C, 2005. On Corporate Governance, Management and Entrepreneurship in OECD Countries. Corporate Governance & Control 2(3): 36-50. Bradley, M., Schipani, C. A., Sundaram, A. K. and Walsh, J. P, 1999. The Purposes and Accountability of the Corporation in Contemporary Society: Corporate Governance at a Crossroads. Law and Contemporary Problems 62(3): 9-51. 164 Brailsford, T. J., Oliver, B. R. and Pua, S. L. H, 2002. On the Relation between Ownership Structure and Capital Structure. Accounting and Finance 42: 1-26. Brazilian Institute of Corporate Governance, 1999. Code of Best Practice (IBGC Code). Brealey, R. A. and Myers, S. C, 1996. Principles of Corporate Finance. McGraw-Hill, New York. Brennan, M. J, 1995. Corporate Finance over the Past 25 Years. Financial Management 24: 9-22. Brick, I. E., Palmon, O. and Wald, J. K, 2006. CEO Compensation, Director Compensation, and Firm Performance: Evidence of Cronyism. Journal of Corporate Finance 12: 403-423. Brickley, J. A., Coles, J. L. and Terry, R. L, 1994. Outside Directors and the Adoption of Poison Pills. Journal of Financial Economics 35: 371-390. Brickley, J. A., Lease, R. C. and Smith, C. W, 1994. Corporate Voting: Evidence from Charter Amendment Proposals. Journal of Corporate Finance 1: 5-31. Brigham, E. F, 1985. Financial Management: Theory and Practice. Dryden Press, Chicago. Brussels Stock Exchange, 1998. Report of the Belgium Commission on Corporate Governance (Cardon Report). BRT, 1997. Statement on Corporate Governance (BRT Report). Byrd, J. W. and Hickman, K. A, 1992. Do Outside Directors Monitor Managers? Evidence from Tender Offer Bids. Journal of Financial Economics 32: 195-222. Calleja, N, 1999. “To Delegate or Not to Delegate”: Board Committees and Corporate Performance in Australia’s Top 100 Companies. Sydney Law Review 21: 5-35. Campbell, I, 2002. Foreword to Contemporary Issues in Corporate Governance. University of New South Wales Law Journal 25(2): 267-269. Carpenter, M. and Westphal, J. D, 2001. The Strategic Context of External Network Ties: Examining the Impact of Board Appointments on Board Involvement in Strategic 165 Decision Making. Academy of Management Journal 44: 639-660. Castaldi, R. and Wortmann Jr, M. S, 1984. Boards of Directors in Small Corporations: An Untapped Resource. American Journal of Small Business 9(2): 1-10. Chaganti. R. S., Mahajan, V. and Sharma, S, 1985. Corporate Board Size, Composition and Corporate Failures in Retailing Industry. Journal of Management Studies 22: 400417. Chakravarthy, B. S, 1986. Measuring Strategic Performance. Strategic Management Journal 7: 437-458. Chambers, R. J, 1993. Positive Accounting Theory and the PA Cult. ABACUS 29(1): 126. Chan, K. C. and Li, J, 2008. Audit Committee and Firm Value: Evidence on Outside Top Executives as Expert-Independent Directors. Corporate Governance: An International Review 16(1): 16-31. Chander, A. D. Jr, 1962. Strategy and Structure: Chapters in the History of the Industrial Enterprise. MIT Press, Cambridge, Massachusetts. Chander, A. D. Jr, 1994. The Functions of the HQ Unit in the Multibusiness Firm. Fundamental Issues in Strategy: A Research Agenda. Harvard Business Press, Boston. Chang, A. and Leng, A, 2004. The Impact of Corporate Governance Practices on Firms’ Financial Performance. ASEAN Economic Bulletin 21(3): 308-318. Chen, Z., Cheung, Y. L., Stouraitis, A. and Wong, A. W. S, 2005. Ownership Concentration, Firm Performance, and Dividend Policy in Hong Kong. Pacific-Basin Finance Journal 13: 431-449. Child, J, 1972. Organizational Structure, Environment and Performance: The Role of Strategic Choice. Sociology 6: 1-22. Chin, T., Vos, E. and Casey, Q, 2004. Level of Ownership Structure, Board Composition and Board Size Seem Unimportant in New Zealand. Corporate Ownership & Control 2(1): 119-128. 166 Cho, M. H, 1998. Ownership Structure, Investment, and the Corporate Value: An Empirical Analysis. Journal of Financial Economics 47: 103-121. Choi, J. J., Park, S. W. and Yoo, S. S, 2007. The Value of Outside Directors: Evidence from Corporate Governance Reform in Korea. Journal of Financial and Quantitative Analysis 42(4): 941-962. Christenson, C, 1983. The Methodology of Positive Accounting. The Accounting Review January: 1-22. Chun, L., Rosenstein, S., Rangan, N. and Davidson, W, 1992. Board Composition and Shareholders Wealth: The Case of Management Buyout. Financial Management Spring: 58-72. Chung, K. H. and Pruitt, S. W, 1994. A Simple Approximation of Tobin’s q. Financial Management 23(3): 70-74. CIMA, 2006. Management Accounting – Financial Strategy. Claessens, S., Djankov, S., Fan, J. and Lang, L, 1998. The Pattern and Valuation Effects of Corporate Diversification: A Comparison of the U.S., Japan, and Other East Asian Economies. World Bank. Claessens, S., Djankov, S., Fan, J. and Lang, L, 2002. Expropriation of Minority Shareholders in East Asia. Journal of Finance 57: 2741-2771. Clifford, P. and Evans, R, 1997. Non-executive Directors: A Question of Independence. Corporate Governance: An International Review 5(4): 224-231. Coffee, J. C, 1999. The Future as History: The Prospects for Global Convergence in Corporate Governance and its Implication. Northwestern University Law Review 93: 641-708. Coffee, J. C, 2002. Convergence and its Critics: What are the Preconditions to the Separation of Ownership and Control? Corporate Governance Regimes: Convergence and Diversity. Oxford University Press, Oxford. Coles, J. L., Naveen, D. D. and Lalitha, N, 2008. Boards: Does One Size Fit All? Journal of Financial Economics 87: 329-356. 167 Coles, J. W., McWilliams, V. B. and Sen, N, 2001. An Examination of Relationship of Governance Mechanisms to Performance. Journal of Management 27: 23-50. Comment, R. and Schwert, G. W, 1995. Poison or Placebo? Evidence on the Deterrence and Wealth Effects of Modern Antitakeover Measures. Journal of Financial Economics 39: 3-43. Committee on Corporate Governance, 1997. Corporate Governance in the Netherlands – Forty Recommendations (Peters Code). Committee on Corporate Governance, 1998. Committee on Corporate Governance: Final Report (Hampel Report). Conseil National du Patronat Francais and Association Francaise des Enterprises Privees, 1995. Boards of Directors of Listed Companies in France (Vienot I Report). Conseil National du Patronat Francais and Association Francaise des Enterprises Privees, 1999. Report of the Committee on Corporate Governance (Vienot II Report). Corporate Governance Forum of Japan, 1998. Corporate Governance Principles – A Japanese View. Corrado, C. J., 1989. A Nonparametric Test for Abnormal Security-Price Performance in Event Studies. Journal of Financial Economics 23: 385-396. Cotter, J., Shivdasani, A. and Zenner, M, 1997. Do Independent Directors Enhance Target Shareholder Wealth during Tender Offers? Journal of Financial Economics 43: 195-218. Cotter, J. and Silverster, M, 2003. Board and Monitoring Committee Independence. ABACUS 39 (2): 211-232. Dafinone, D, 2001. The Effectiveness of Audit Committees: An Analysis of Corporate Governance Mechanisms as Surrogates for Effectiveness. PhD thesis, Business School, London City University. Daily, C. M. and Dalton, D. R, 1992. The Relationship between Governance Structure and Corporate Performance in Entrepreneurial Firms. Journal of Business Venturing 7(5): 375-387. 168 Daily, C. M. and Dalton, D. R, 1993. Board of Directors Leadership and Structure: Control and Performance Implications. Entrepreneurship Theory and Practice 17(3): 65-81. Daily, C. M. and Dalton, D. R, 1994. Bankruptcy and Corporate Governance: The Impact of Board Composition and Structure. Academy of Management Journal 37: 1603-1617. Daily, C. M., Johnson, J. L. and Dalton, D. R, 1999. On the Measurement of Board Composition: Poor Consistency and a Serious Mismatch of Theory and Operationalizations. Decision Sciences 30(1): 83-106. Daily, C. M., McDougall, P. P., Covin, J. G. and Dalton, D. R, 2002. Governance and Strategic Leadership in Entrepreneurial Firms. Journal of Management 28(3): 387-412. Dalton, D. R., Daily, C. M., Ellstrand, A. E. and Johnson, J. L, 1998. Meta-Analytic Reviews of Board Composition, Leadership Structure, and Financial Performance. Strategic Management Journal 19: 269-290. Danielson, M. G. and Karpoff, J. M, 1998. On the Use of Corporate Governance Provisions. Journal of Corporate Finance, Contracting, Governance and Organization 4: 347-371. Davis, G. F. and Thompson, T. A, 1994. A Social Movement Perspective on Corporate Control. Administrative Science Quarterly 39: 141-173. Davis, J. H., Schoorman, F. D. and Donaldson, L, 1997. Towards a Stewardship Theory of Management. Academy of Management Review 22: 20-47. Dechow, P. W., Sloan, R. G. and Sweeney, A. P, 1996. Causes and Consequences of Earnings Manipulation: An Analysis of Firms Subject to Enforcement Actions by the SEC. Contemporary Accounting Research 13: 1-36. Dehaene, A., Vuyst, V. D. and Ooghe, H, 2001. Corporate Performance and Board Structure in Belgian Companies. Long Range Planning 34: 383-398. Denis, D., Denis, D. and Sarin, A, 1997. Agency Problems, Equity Ownership, and Corporate Diversification. Journal of Finance 52: 135-160. 169 Denis, D. J. and Sarin, A, 1999. Ownership and Board Structures in Publicly Traded Corporations. Journal of Financial Economics 52: 187-223. Denis, D. K. and McConnell, J. J, 2003. International Corporate Governance. Journal of Financial and Quantitative Analysis 38(1): 1-36. Department for Trade and Industry, 1992. Report on the Committee on the Financial Aspects of Corporate Governance (Cadbury Report). Department for Trade and Industry, 2003. Higgs Review on the Role and Effectiveness of Non-Executive Directors (Higgs Report). Devinney, T. M., Richard, P. J., Yip, G. S. and Johnson, G, 2005. Measuring Organizational Performance in Management Research: A Synthesis of Measurement Challenges and Approaches. Australian Graduate School of Management, University of New South Wales. Dignam, A. and Galanis, M, 2004. Australia Inside-Out: The Corporate Governance System of the Australian Listed Market. Melbourne University Law Review 28: 623653. DiMaggio, P. J. and Powell, W. W, 1983. The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields. American Sociological Review 48: 147-160. Donaldson, L, 1987. Strategy and Structural Adjustment to Regain Fit and Performance: In Defense of Contingency Theory. Journal of Management Studies 24: 1-24. Donaldson, L, 1990. The Ethereal Hand: Organizational Economics and Management Theory. Academy of management Review 15: 369-381. Donaldson, L, 2000. Organizational Portfolio Theory: Performance-Driven Organizational Change. Contemporary Economic Policy 18(4): 386-396. Donaldson, L. and Davis, J. H, 1991. Stewardship Theory and Agency Theory: CEO Governance and Shareholder Returns. Australian Journal of Management 16: 49-64. 170 Donaldson, L. and Davis, J. H, 1994. Boards and Company Performance: Research Challenges the Conventional Wisdom. Corporate Governance: An International Review 2(3): 151-160. Doucouliagos, C, 1994. A Note on the Evolution of Homo Economics. Journal of Economics Issues 3: 877-883. Dulewicz, V. and Herbert, P, 2004. Does the Composition and Practice of Boards of Directors Bear any Relationship to the Performance of their Companies? Corporate Governance: An International Review 12(3): 263-280. Easterbrook, F. H, 1984. Two Agency Cost Explanations of Dividends. American Economic Review 74(4): 650-659. Easterbrook, F. H, 1997. International Corporate Differences: Market or Law? Journal of Applied Corporate Finance 9: 23-29. Easterbrook, F. H. and Fischel, D. R, 1991. The Economic Structure of Corporate Law. Harvard University Press, Cambridge, Massachusetts. Eisenberg, M. A, 1976. The Structure of the Corporation: A Legal Analysis. Little Brown and Company, Boston. Epstein, E. J, 1986. Who Owns the Corporation? Management vs. Shareholders. Priority Press Publications, New York. Etzioni, A, 1975. A Comparative Analysis of Complex Organizations: On Power, Investment and their Correlates. Free Press, New York. Ezzamel, M. and Watson, R, 1993. Organizational Form, Ownership Structure and Corporate Performance: A Contextual Empirical Analysis of U.K. Companies. British Journal of Management 4: 161-176. Fama, E. F, 1980. Agency Problem and the Theory of Firm. Journal of Political Economy 88(2): 288-307. Fama, E. F. and Jensen, M. C, 1983. Separation of Ownership and Control. Journal of Law and Economics 26: 301-325. Ferris, S. P., Jagannathan, M. and Pritchard, A. C, 2003. Too Busy to Mind the 171 Business? Monitoring by Directors with Multiple Board Appointments. Journal of Finance 58: 1087-1111. Ferris, S. P., Lawless, R. M. and Makhija, A. K, 2001. Derivative Lawsuits as a Corporate Governance Mechanism: Empirical Evidence on Board Changes Surrounding Filings. Contracting and Organizations Research Institute, University of Missouri-Columbia. Financial Reporting Council, 2003. The Smith Guidance on Audit Committee (Smith Report). Finegold, D., Benson, G. S. and Hecht, D, 2007. Corporate Boards and Company Performance: Review of Research in Light of Recent Reforms. Corporate Governance: An International Review 15(5): 865-878. Fisher, F. M. and McGowan, J. J, 1983. On the Misuse of Accounting Rates of Return to Infer Monopoly Profits. American Economic Review 73: 82-97. Fleischer, A. Jr., Hazard, G. C. and Klipper, M. Z, 1988. Board Games: The Changing Shape of Corporate Power. Little Brown, Boston. Forbes, D. and Milliken, F, 1999. Cognition and Corporate Governance: Understanding Boards of Directors as Strategic Decision Making Groups. Academy of Management Review 24: 489-505. Fosberg, R. H, 1989. Outside Directors and Managerial Monitoring. Akron Business and Economic Review 20(2): 24-32. Fox, M. A. and Hamilton, R. T, 1994. Ownership and Diversification: Agency Theory or Stewardship Theory. Journal of Management Studies 31: 69-81. Frank, R. H, 1994. Microeconomics and Behaviour. McGraw-Hill, New York. Frankfort-Nachmias, C. and Leon-Guerrero, A, 2006. Social Statistics for a Diverse Society. Pine Forge Press, Thousand Oaks, California. Franks, J. R. and Mayer, C, 2001. Ownership and Control of German Corporations. Review of Financial Studies 14(4): 943-977. 172 Fukao, M, 1995. Financial Integration, Corporate Governance, and the Performance of Multinational Companies. Brookings Institution, Washington DC. Gay, K, 2002. Board Theories and Governance Practices: Agents, Stewards and their Evolving Relationship with Stakeholders. Journal of General Management 27: 36-61. Gill, J. and Johnson, P, 2002. Research Methods for Managers. Sage Publications, London. Gillan, S. L, 2006. Recent Developments in Corporate Governance: An Overview. Journal of Corporate Finance 12: 381-402. Gillan, S. L. and Starks, L. T, 2003. Corporate Governance, Corporate Ownership, and the Role of Institutional Investors: A Global Perspective. Journal of Applied Finance 13(2): 4-22. Godfrey, J., Hodgson, A. and Holmes, S, 2003. Accounting Theory. Wiley, Milton, Queensland. Gray, R., Owen, D. and Adams, C, 1996. Accounting and Accountability. Prentice Hall, London. Green, H, 1984. Why Directors Can’t Protect the Shareholders. Fortune 110(6): 28-32. Gregory, H. J, 1999. Overview of Corporate Governance Guidelines and Codes of Best Practice in Developing and Emerging Markets. Weil, Gotshal & Manges LLP, New York. Gregory, H. J, 2001a. International Comparison of Board “Best Practices” in Developed Markets. Weil, Gotshal & Manges LLP, New York. Gregory, H. J, 2001b. International Comparison of Board “Best Practices” in Developing and Emerging Markets. Weil, Gotshal & Manges LLP, New York. Gregory, H. J, 2001c. International Comparison of Board “Best Practices” – Investor Viewpoints. Weil, Gotshal & Manges LLP, New York. Grossman, S. J. and Hart, O. D, 1980. Takeover Bids, the Free-rider Problem and the Theory of the Corporation. Bell Journal of Economics 11: 42-64. 173 Grossman, S. J. and Hart, O. D, 1982. Corporate Financial Structure and Managerial Incentives. The Economics of Information and Uncertainty. University of Chicago Press, Chicago. Grundfest, J. A, 1990. Subordination of American Capital. Journal of Financial Economics 27: 89-114. Gujarati, D. N, 2003. Basic Econometrics. McGraw-Hill, New York. Hall, P. A. and Soskice, D, 2001. Varieties of Capitalism: The Institutional Foundations of Comparative Advantage. Oxford University Press, Oxford. Hambrick, D. C. and Manson, P, 1984. Upper Echelons: The Organization as a Reflection of its Top Managers. Academy of Management Review 9: 193-206. Hamilton, R. T. and Shergill, G. S, 1992. The Relationship between Strategy-Structure Fit and Financial Performance in New Zealand: Evidence of Generality and Validity with Enhanced Controls. Journal of Management Studies 29: 95-113. Hansmann, H. and Kraakman, R, 2001. The End of History for Corporate Law. Georgetown Law Journal 89: 439-468. Harris, M. and Raviv, A, 1991. The Theory of Capital Structure. Journal of Finance 46(1): 297-355. Haugen, R. and Sendbet, L, 1986. Corporate Finance and Taxes: A Review. Financial Management Autumn: 5-21. Hausman, J. A, 1978. Specification Tests in Econometrics. Econometrica 46: 12511271. Helland, E. and Sykuta, M, 2005. Who’s Monitoring the Monitor? Do Outside Directors Protect Shareholders’ Interests? Financial Review 40: 155-172. Henn, H. G, 1974. Cases and Materials on the Laws of Corporations. West Publishing, St Paul, Minnesota. Hermalin, B. E. and Weisbach, M. S, 1988. The Determinants of Board Composition. Rand Journal of Economics 19: 589-606. 174 Hermalin, B. E. and Weisbach, M. S, 1991. The Effect of Board Composition and Direct Incentives on Firm Performance. Financial Management 21(4): 101-112. Hermalin, B. E. and Weisbach, M. S, 1998. Endogenously Chosen Boards and their Monitoring of the CEO. American Economic Review 88(1): 96-118. Hermalin, B. E. and Weisbach, M. S, 2003. Boards of Directors as an Endogenously Determined Institution: A Survey of the Economic Literature. Economic Policy Review 9: 7-26. Herzberg, F, 1966. Work and the Nature of Man. World Publishing, New York. Heslin, P. A. and Donaldson, L, 1999. An Organizational Portfolio Theory of Board Composition. Corporate Governance: An International Review 7(1): 81-88. Hill, C. W. L. and Snell, S. A, 1988. External Control, Corporate Strategy, and Firm Performance in Research-Intensive Industries. Strategic Management Journal 9(6): 577-590. Himmelberg, C. P., Hubbard, R. G. and Palia, D, 1999. Understanding the Determinants of Ownership and the Link between Ownership and Performance. Journal of Financial Economics 53: 353-384. HKSE, 1995. Guide for Directors of Listed Companies (HKSE Guide). Hofer, C. W, 1983. ROVA: A New Measure for Assessing Organizational Performance. Advances in Strategic Management 2: 43-55. Holmstrom, B. and Kaplan, S. N, 2001. Corporate Governance and Merger Activity in the United States: Making Sense of the 1980s and 1990s. Journal of Economic Perspectives 15(2): 121-144. Holthausen, R. W. and Larcker, D. F, 1993. Boards of Directors, Ownership Structure, and CEO Compensation. Wharton School, University of Pennsylvania. Hoshi, T., Kashyap, A. and Scharfstein, D, 1990. The Role of Banks in Reducing the Costs of Financial Distress in Japan. Journal of Financial Economics 27: 67-88. Hossain, M., Prevost, A. K. and Rao, R. P, 2001. Corporate Governance in New Zealand: The Effect of the 1993 Companies Act on the Relation between Board 175 Composition and Firm Performance. Pacific-Basin Finance Journal 9: 119-145. Howieson, B, 1996. Whither Financial Accounting Research: A Modern-Day Bo-Peep? Australian Accounting Review 6(1): 29-36. Hudson, C. D., Jahera, J. S. and Lloyd, W. P, 1992. Further Evidence on the Relationship between Ownership and Performance. Financial Review 27: 227-240. Hung, H, 1998. A Typology of the Theories of the Roles of Governing Boards. Corporate Governance: An international Review 6(2): 101-111. Huse, M, 1998. Researching the Dynamics of Board-Stakeholder Relations. Long Range Planning 31: 218-226. Hussey, J. and Hussey, R, 1997. Business Research: A Practical Guide for Undergraduate and Postgraduate Students. Macmillan, London. Hutchinson, M. and Gul, F. A, 2004. Investment Opportunity Set, Corporate Governance Practices and Firm Performance. Journal of Corporate Finance 10: 595614. ICAEW, 1999. Internal Control Guidance for Directors on the Combined Code (Turnbull Report). Instituto Universitario Euroforum Escorial, 1998. Governance of Spanish Companies. Jackson, G, 2001. The Origins of Nonliberal Corporate Governance in Germany and Japan. The Origins of Nonliberal Capitalism: Germany and Japan in Comparison. Cornell University Press, Ithaca. Jackson, G. and Moerke, A, 2005. Continuity and Change in Corporate Governance: Comparing Germany and Japan. Corporate Governance: An International Review 13(3): 351-361. Jansson, E, 2005. The Stakeholder Model: The Influence of the Ownership and Governance Structures. Journal of Business Ethics 56: 1-13. Jensen, M. C, 1986. Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers. American Economic Review 76(2): 323-329. 176 Jensen, M. C, 1993. The Modern Industrial Revolution, Exit and the Failure of Internal Control Systems. Journal of Finance 48: 831-880. Jensen, M. C. and Meckling, W. H, 1976. Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure. Journal of Financial Economics 2: 305-360. Jensen, M. C. and Meckling, W. H, 1994. The Nature of Man. Journal of Applied Corporate Finance 7(2): 4-19. Johnson, S, 2000. Corporate Governance in the Asian Financial Analysis. Journal of Financial Economics 58: 141-186. Johnson, J. L., Daily, C. M. and Ellstrand, A. E, 1996. Boards of Directors: A Review and Research Agenda. Journal of Management 22: 409-438. Judge, G. C., Hill, R. C., Griffiths, W. E., Lutkepohl, H. and Lee, T. C, 1982. Introduction to the Theory and Practices of Econometrics. Wiley, New York. Juran, J. M. and Louden, J. K, 1966. The Corporate Director. American Management Associate, New York. Kang, J. K. and Stulz, R. M, 2000. Do Banking Shocks Affect Borrowing Firm Performance? An Analysis of the Japanese Experience. Journal of Business 73: 1-23. Kesner, I. F, 1987. Directors’ Stock Ownership and Organizational Performance: An Investigation of Fortune 500 Companies. Journal of Management 13: 499-507. Kiel, G. C. and Nicholson, G. J, 2003. Board Composition and Corporate Performance: How the Australian Experience Informs Contrasting Theories of Corporate Governance. Corporate Governance: An International Review 11(3): 189-205. Kiel, G. C. and Nicholson, G. J, 2007. Can Directors Impact Performance? A CaseBased Test of Three Theories of Corporate Governance. Corporate Governance: An International Review 15(4): 585-608. Kim, W. S., Lee, J. W. and Francis, J. C, 1988. Investment Performance of Common Stocks in Relation to Insider Ownership. Financial Review 23: 53-64. Klein, A, 1998. Firm Performance and Board Committee Structure. Journal of Law and Economics 41: 275-303. 177 Klein, A, 1999. An Empirical Analysis of the Relation between Board of Directors' Composition, Firm Performance, and the Degree of CEO Domination over the Board of Directors. Stern School of Business, New York University. Knoeber, C. R, 1986. Golden Parachutes, Shark Repellents and Hostile Tender Offers. American Economic Review 76: 155-167. Kole, S, 1995. Measuring Managerial Equity Ownership: A Comparison of Sources of Ownership Data. Journal of Corporate Finance 1: 413-435. Kole, S, 1996. Managerial Ownership and Firm Performance: Incentives or Rewards? Advances in Financial Economics 2: 119-149. Koontz, H, 1967. The Board of Directors and Effective Management. McGraw-Hill, New York. Kosnik, R. D, 1987. Greenmail: A Study of Board Performance in Corporate Governance. Administrative Science Quarterly 32: 163-185. Kreiken, J, 1985. The Board’s Role in Strategic Planning and Resource Allocation. Handbook for Corporate Directors. McGraw-Hill, New York. Krivogorsky, V, 2006. Ownership, Board Structure, and Performance in Continental Europe. International Journal of Accounting 41: 176-197. Lamba, A. and Stapledon, G, 2001. The Determinants of Corporate Ownership Structure: Australian Evidence. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=279015. La Porta, R., Lopez-de-Silanes, F. and Shleifer, A, 1998. Law and Finance. Journal of Political Economy 106: 1113-1155. La Porta, R., Lopez-de-Silanes, F. and Shleifer, A, 1999. Corporate Ownership around the World. Journal of Finance 54: 471-517. La Porta, R., Lopez-de-Silanes, F., Shleifer, A. and Vishny, R. W, 2000. Investor Protection and Corporate Governance. Journal of Financial Economics 58: 3-27. Larcker, D. F., Richardson, S. A. and Tuna, I, 2007. Corporate Governance, Accounting Outcomes, and Organizational Performance. Accounting Review 82(4): 963-1008. 178 Lawrence, J. J. and Stapledon, G. P, 1999. Do Independent Directors Add Value? Centre for Corporate Law and Securities Regulation, University of Melbourne. Leamer, E. E, 1978. Specification Searches: Adhoc Inference with Non-Experimental Data. John Wiley, New York. Lindenberg, E. B. and Ross, S. A, 1981. Tobin’s q Ratio and Industrial Organization. Journal of Business 54(1): 1-32. Lipton, M. and Lorsch, J, 1992. A Modest Proposal for Improved Corporate Governance. Business Lawyer 1: 59-77. Litch, A, 2002. Accountability and Corporate Governance. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=328401. Loevinger, L, 1986. Courts and Boards: A Decade of Turmoil. Directors & Boards 11 (1): 36-38. LSE, 1998, 2003. Combined Code: Principles of Good Governance and Code of Best Practice (Combined Code). Lorsch, J. W. and Maclver, E, 1989. Pawns or Potentates: The Reality of America’s Corporate Boards. Harvard Business School Press, Boston. Louden, J. K, 1982. The Director: A Professional Guide to Effective Board Work. Amacom, New York. Luan, C. J. and Tang, M. J, 2007. Where is Independent Director Efficacy? Corporate Governance: An International Review 15(4): 636-643. Lynall, M. D., Golden, B. R. and Hillman, A. J, 2003. Board Composition from Adolescence to Maturity: A Multi-Theoretical View. Academy of Management Review 28(3): 416-431. MacAvoy, P. W., Cantor, S., Dana, J. and Peck, S, 1983. ALI Proposals for Increased Control of the Corporation by the Board of Directors: An Economic Analysis. Statement of the Business Roundtable on the American Law Institute's Proposed "Principles of Corporate Governance and Structure: Restatement and Recommendations, exhibit C-1. 179 Mace, M. L, 1971. Directors: Myth and Reality. Graduate School of Business Administration, Harvard University. Mason, J, 1996. Qualitative Researching. Sage Publications, Thousand Oaks, California. McCauley, R. N. and Zimmer, S. A, 1994. Exchange Rates and International Differences in the Cost of Capital. Exchange Rates and Corporate Performance. New York University. McClelland, D, 1961. The Achieving Society. Free Press, New York. McColgan, P, 2001. Agency Theory and Corporate Governance: A Review of the Literature from a UK Perspective. Department of Accounting and Finance, University of Strathclyde. McConell, J. J. and Servaes, H, 1990. Additional Evidence on Equity Ownership and Corporate Value. Journal of Financial Economics 27: 595-612. Mikkelson, W. H. and Partch, M. M, 1997. The Decline of Takeovers and Disciplinary Managerial Turnover. Journal of Financial Economics 44: 205-228. Molz, R, 1988. Managerial Domination of Boards of Directors and Financial Performance. Journal of Business Research 16: 235-249. Morck, R., Shleifer, A. and Vishny, R. W, 1988. Management Ownership and Market Valuation: An Empirical Analysis. Journal of Financial Economics 20(1): 293-315. Murphy, A. and Topyan, K, 2005. Corporate Governance: A Critical Survey of Key Concepts, Issues, and Recent Reforms in the U.S. Employee Responsibilities and Rights Journal 17(2): 75-89. Muth, M. M. and Donaldson, L, 1998. Stewardship Theory and Board Structure: A Contingency Approach. Corporate Governance: An International Review 6(1): 5-28. Myer, J. and Rowan, B, 1977. Institutional Organizations: Formal Structure as Myth and Ceremony. American Journal of Society 83: 340-363. NACD, 1996. Report of the NACD Commission on Director Professionalism (NASD Report). 180 Narayanan, M. P, 1985. Managerial Incentives for Short-term Results. Journal of Finance 40(5): 1469-1484. Newman, H. A. and Wright, D. W, 1995. Compensation Committee Composition and its Influence on CEO Compensation Practices. School of Business, University of Michigan. Novak, J. and Gowin, R, 1984. Learning how to Learn. Oxford University Press, London. OECD, 1999. Principles of Corporate Governance. OECD Business Sector Advisory Group on Corporate Governance, 1999. Corporate Governance: Improving Competitiveness and Access to Capital in Global Markets (Millstein Report). Oliver, C, 1991. Strategic Responses to Institutional Processes. Academy of Management Review 16: 145-179. Ong, C. H. and Lee, S. H, 2000. Board Functions and Firm Performance: A Review and Directions for Future Research. Journal of Comparative International Management 3(1): Panasian, C., Prevost, A. K. and Bhabra, H. S, 2003. Board Composition and Firm Performance: The Case of the Dey Report and Publicly Listed Canadian Firms. Department of Finance, Montreal University. Parkinson, J. E, 1993. Corporate Power and Responsibility. Oxford University Press, Oxford. Patton, A. and Baker, J. C, 1987. Why do not Directors Rock the Boat? Harvard Business Review 65: 10-13. Pearce II, J. A. and Zahra, S. A, 1991. The Relative Power of CEOs and Boards of Directors: Associations with Corporate Performance. Strategic Management Journal 12: 135-153. Pearce II, J. A. and Zahra, S. A, 1992. Board Composition from a Strategic Contingency Perspective. Journal of Management Studies 29(4): 411-438. 181 Peasnell, K. V., Pope, P. E. and Young, S, 1998. Outside Directors, Board Effectiveness, and Earnings Management. Lancaster University. Peng, M. W, 2004. Outside Directors and Firm Performance during Institutional Transactions. Strategic Management Journal 25: 453-471. Pennington, 1986. Duties and Liabilities of Company Directors in Public Listed Companies. Professional Resources Agency Pte Ltd, Singapore. Perotti, E. C. and Von Thadden, E. L, 2003. Strategic Transparency and Informed Trading: Will Capital Market Integration Force Convergence of Corporate Governance? Journal of Financial and Quantitative Analysis 38: 61-85. Perrow, C, 1986. Complex Organizations: A Critical Essay. McGraw-Hill, New York. Pettigrew, A, 1992. On Studying Managerial Elites. Strategic Management Journal 13(8): 163-182. Pfeffer, J, 1972. Size and Composition of Corporate Boards of Directors: The Organization and its Environment. Administrative Science Quarterly 17: 218-228. Pfeffer, J. and Salancik, G. R, 1978. The External Control of Organizations: A Resource Dependence Perspective. Harper and Row, New York. Phillips, P. C. B, 1980. The Exact Distribution of Instrumental Variable Estimators in an Equation Containing n+1 Endogenous Variables. Economics 48: 861-878. Pindyck, R. S. and Rubinfeld, D. L, 1991. Econometric Models and Economic Forecasts. McGraw-Hill, New York. Porter, M. E, 1992a. Capital Disadvantage: America’s Failing Capital Investment System. Harvard Business Review: 68-82. Porter, M. E, 1992b. Capital Choices: Changing the Way America Invests in Industry. Harvard Business School, Harvard University. Prahalad, C. K. and Hamel, G, 1990. The Core Competence of the Corporation. Harvard Business Review 68(3): 79-91. 182 Prentice, D. D, 1993. Some Aspects of Corporate Governance Debate. Contemporary Issues in Corporate Governance. Clarendon Press, Oxford. Prevost, A. K., Rao, R. P. and Hossain, M, 2002. Determinants of Board Composition in New Zealand: A Simultaneous Equations Approach. Journal of Empirical Finance 9: 373-397. Prowse, S. D, 1990. Institutional Investment Patterns and Corporate Financial Behavior in the U.S. and Japan. Journal of Financial Economics 27: 43-66. Quinn, D. P. and Jones, T. M, 1995. An Agent Morality View of Business Politics. Academy of Management Review 20(1): 22-42. Randoy, T. and Jenssen, J. I, 2004. Board Independence and Product Market Competition in Swedish Firms. Corporate Governance: An International Review 12(3): 281-289. Rechner, P. L. and Dalton, D. R, 1991. CEO Duality and Organizational Performance: A Longitudinal Analysis. Strategic Management Journal 12: 155-160. Reilly, F. K, 1985. Investment Analysis and Portfolio Management. Dryden Press, Chicago. Reilly, F. K. and Brown, K. C, 2006. Investment Analysis and Portfolio Management. Thomson South-Western, Manson, Ohio. Rhodes, G. F. Jr. and Westbrook, M. D, 1981. A Study of Estimator Densities and Performance under Misspecification. Journal of Econometrics 16: 311-337. Roe, M. J, 1990. Political and Legal Restraints on Ownership and Control of Public Companies. Journal of Financial Economics 27: 7-41. Roe, M. J, 1991. A Political Theory of American Corporate Finance. Columbia Law Review 91: 10-67. Roe, M. J, 1994. Strong Managers, Weak Owners: The Political Roots of American Corporate Finance. Princeton University Press, Princeton. Romano, R, 1991. Corporate Governance in the Aftermath of the Insurance Crisis. Emory Law Journal 39: 1155-1189. 183 Romano, R, 1993. The Genius of American Corporate Law. AEI Press, Washington DC. Rosenstein, J, 1987. Why don’t U.S. Boards Get More Involved in Strategy? Long Range Planning 20(3): 20-34. Rosenstein, S. and Wyatt, J. G, 1990. Outside Directors, Board Independence, and Shareholder Wealth. Journal of Financial Economics 26: 175-191. Rosenstein, S. and Wyatt, J. G, 1997. Inside Directors, Board Effectiveness, and Shareholder Wealth. Journal of Financial Economics 44: 229-250. Ross, S. A, 1987. The Interrelations of Finance and Economics. American Economic Review 77: 29-34. Ross, S. A. and Westerfield, R. W, 1988. Corporate Finance. Times Mirror/Mosby College Publishing, St. Louis. Ross, S. A., Westerfield, R. W. and Jaffe, J, 2005. Corporate Finance. McGraw-Hill, Irwin. Rozeff, M. S, 1982. Growth, Beta and Agency Costs as Determinants of Dividend Payout Ratio. Journal of Financial Research 5(3): 249-260. Ryan, B., Scapens, R. and Theobald, M, 2002. Research Method and Methodology in Finance and Accounting. Academic Press, London. Schellenger, M. H., Wood, D. D. and Tashakori, A, 1989. Board of Directors Composition, Shareholder Wealth, and Dividend Policy. Journal of Management 15: 457-467. Schmidt, R, 1975. Does Board Composition Really Make a Difference? Conference Board Record 12(10): 38-41. Schmidt, S. J, 2005. Econometrics. McGraw-Hill, Boston. Scott, J, 1997. Corporate Business and Capitalist Classes. Oxford University Press, New York. Scott, K. E, 1983. Corporation Law and the American Law Corporate Governance 184 Project. Stanford Law Review 35: 927-947. Servaes, H, 1996. The Value of Diversification during the Conglomerate Merger Wave. Journal of Finance 51: 1201-1225. SET, 1997. The SET Code of Best Practice for Directors of Listed Companies (SET Code). Shankmann, N. E, 1999. Reframing the Debate between Agency and Stakeholder Theories of the Firm. Journal of Business Ethics 19: 319-334. Shinn, J. and Gourevitch, P, 2002. How Shareholder Reforms can Pay Foreign Policy Dividends. Council on Foreign Relations, New York. Shleifer, A. and Summers, L. H, 1988. Breach of Trust in Hostile Takeover. Corporate Takeovers: Causes and Consequences. University of Chicago Press, Chicago. Shleifer, A. and Vishny, R. W, 1989. Equilibrium Short Horizons of Investors and Firms. American Economic Review 80(2): 148-153. Shleifer, A. and Vishny, R. W, 1997a. A Survey of Corporate Governance. Journal of Finance 52: 737-783. Shleifer, A. and Vishny, R. W, 1997b. The Takeover Wave of the 1980s. Studies in International Corporate Finance and Governance Systems. Oxford University Press, New York. Shome, D. and Singh, S, 1995. Firm Value and External Blockholdings. Financial Management 24: 3-14. Short, H. and Keasey, K, 1999. Managerial Ownership and the Performance of Firms: Evidence from the U.K. Journal of Corporate Finance, Contracting, Governance and Organization 5: 79-101. Shrader, C. B., Taylor, L. and Dalton, D. R, 1984. Strategic Planning and Organizational Performance: A Critical Appraisal. Journal of Management 10: 149-171. Singh, M. and Davidson III, W. N, 2003. Agency Costs, Ownership Structure and Corporate Governance Mechanisms. Journal of Banking & Finance 27: 793-816. 185 Smith, J. A, 2003. Qualitative Psychology: A Practical Guide to Research Methods. Sage Publications, London. Sridharan, U. V, 1996. CEO Influence and Executive Compensation. Financial Review 31: 51-66. S&P, 2004. Understanding Indices. http://www2.standardandpoors.com/spf/pdf/index/A5_CMYK_UIndices.pdf. Stapledon, G. P. and Lawerence, J. J, 1996. Corporate Governance in Top 100: An Empirical Study of the Top 100 Companies’ Boards of Directors. Centre of Corporate Law and Securities Regulation, University of Melbourne. Startz, R, 2007. EViews Illustrated for Version 6. Quantitative Micro Software, Irvine, California. Staw, B. and Epstein, L, 2000. What Bandwagons Bring: Effects of Popular Management Techniques on Corporate Performance, Reputation, and CEO Pay. Administrative Science Quarterly 45: 523-556. Stein, J. C, 1988. Takeovers Treats and Managerial Myopia. Journal of Political Economy 96: 61-80. Stein, J. C, 1989. Efficient Capital Markets, inefficient Firms: A Model of Myopic Corporate Behavior. Quarterly Journal of Economics 104: 655-669. Sterling, R. R, 1990. Positive Accounting: An Assessment. ABACUS 26(2): 97-135. Stiles, P. and Taylor, B, 2001. Boards at Work – How Directors View their Roles and Responsibilities. Oxford University Press, Oxford. Storey, D, 1994. Understanding the Small Business Sector. Routledge, London. Stroh, L. K., Brett, J. M., Baumann, J. P. and Reilly, A. H, 1996. Agency Theory and Variable Pay Composition Strategies. Academy of Management Journal 39(3): 751-767. Study Group on Director’s Remuneration, 1995. Final Report (Greenury Report). Suchard, J. A., Singh, M. and Barr, R, 2001. The Market Effects of CEO Turnover in Australian Firms. Pacific-Basin Finance Journal 9: 1-27. 186 Sundaramurthy, C., Rhoades, D. L. and Rechner, P. L, 2005. A Meta-Analysis of the Effects of Executive and Institutional Ownership on Firm Performance. Journal of Managerial Issues 17(4): 494-510. Swedish Academy of Directors, Western Region, 1994. Introduction to a Swedish Code of “Good Boardroom Practice”. Teece, D., Pisano, G. and Sheun, A, 1997. Dynamic Capabilities and Strategic Management. Strategic Management Journal 18: 509-533. Thomas, R. H, 1996. Research Methods and Data Analysis in the Social Science. HarperCollins College Publishers, New York. Tinker, T., Merino. B. and Niemark, N, 1982. The Normative Origins of Positive Theories: Ideology and Accounting Thought. Accounting Organizations and Society 7: 167-200. Tipgos, M. A, 2007. How Wrong is the Agency Theory as Basis for Corporate Governance? School of Business, Indiana University Southeast. TSE Committee on Corporate Governance in Canada, 1994. “Where were the Directors?” Guidelines for Improved Corporate Governance in Canada (Dey Report). Tsetsekos, G. P. and DeFusco, R, 1990. Portfolio Performance, Managerial Ownership, and the Size Effect. Journal of Portfolio Management 16(3): 33-39. Tricker, R. I, 1994. Editorial: On Theories of Corporate Governance. Corporate Governance: An International Review 2: 55-57. Tricker, R. I, 2000. Editorial: Corporate Governance – The Subject Whose Time has Come. Corporate Governance: An International Review 8(4): 289-296. Vafeas, N. and Theodorou, E, 1998. The Relationship between Board Structure and Firm Performance in the U.K. British Accounting Review 30: 383-407. Vance, S. C, 1964. Board of Directors: Structure and Performance. Harvard Business School Press, Boston. Vance, S. C, 1978. Corporate Governance: Assessing Corporate Performance by Boardroom Attributes. Journal of Business Research 6: 203-220. 187 Vance, S. C, 1983. Corporate Leadership: Boards, Directors, and Strategy. McGrawHill, New York. Van Nuys, K, 1993. Corporate Governance through the Proxy Process: Evidence from the 1989 Honeywell Proxy Solicitation. Journal of Financial Economics 34: 101-132. Veal, A, 2005. Business Research Methods: A Managerial Approach. Pearson Education, Australia. Walsh, J. P. and Steward, J. K, 1990. On the Efficiency of Internal and External Corporate Control Mechanisms. Journal of Management Review 15: 421-458. Weimar, J. and Paper, J, 1999. A Taxonomy of Systems of Corporate Governance. Corporate Governance: An International Review 7(2): 152-167. Weisbach, M. S, 1988. Outside Directors and CEO Turnover. Journal of Financial Economics 20: 431-460. Weisberg, H. F., Krosnick, J. A. and Bowen, B. D, 1996. An Introduction to Survey Research, Polling, and Data Analysis. Sage Publications, Thousand Oaks, California. Westphal, J. D, 1997. CEO’s Board Games. Proceedings of the Academy of Management Annual Conference. Academy of Management, Washington DC. Westphal, J. D, 1999. Collaboration in the Boardroom: Behavioral and Performance Consequences of CEO-Board Social Ties. Academy of Management Journal 42: 7-24. Westphal, J. D., Gulati, R. and Shortell, S, 1997. Customization or Conformity? An Institutional and Network Perspective on the Content and Consequences of TQM Adoption. Administrative Science Quarterly 39: 367-390. Westphal, J. D. and Zajac, E. J, 1995. Who Shall Govern? CEO/Board Power, Demographic Similarity, and New Director Selection. Administrative Science Quarterly 40: 60-83. Williams, P. E, 1989. The Logic of Positive Accounting Research. Accounting Organizations and Society 14: 455-468. 188 Williamson, O. E, 1971. Managerial Discretion, Organization Form and the Multidivision Hypothesis. The Corporate Economy: Growth, Competition and Innovation Potential. Harvard University Press, Cambridge, Massachusetts. Williamson, O. E, 1975. Markets and Hierarchies. Free Press, New York. Williamson, O. E, 1984. Corporate Governance. Yale Law Journal 93: 1197-1230. Wojcik, D, 2001. Change in the German Model of Corporate Governance: Evidence from Blockholdings 1997-2001. University of Oxford. Womack, J. P., Jones, D. T. and Roos, D, 1991. The Machine that Changed the World: How Japan’s Secret Weapon in the Global Auto Wars will Revolutionize Western Industry. Harper Perennial, New York. Working Group Representing Australian Institute of Company Directors, Australian Society of Certified Practicing Accountants, Business Council of Australia, Law Council of Australia, Institute of Chartered Accountants in Australia, and Securities Institute of Australia, 1995. Corporate Practices and Conduct (Bosch Report). Wright, D. W, 1996. Evidence on the Relation between Corporate Governance Characteristics and the Quality of Financial Reporting. School of Business, University of Michigan. Yermack, D, 1996. Higher Market Valuation of Companies with a Small Board of Directors. Journal of Financial Economics 40: 185-212. Zahra, S. A, 1990. Increasing the Board’s Involvement in Strategy. Long Range Planning 23(6): 109-117. Zahra, S. A. and Pearce II, J. A, 1989. Boards of Directors and Corporate Financial Performance: A Review and Integrative Model. Journal of Management 15: 291-334. Zajac, E. J. and Westphal, J. D, 1996. Director Reputation, CEO-Board Power, and the Dynamics of Board Interlocks. Administrative Science Quarterly 41: 507-529. 189 Appendix 1 Summary of Australian Research Author Sample Variable* Board Characteristics Firm Performance Control Test Relevant Findings Muth and 145 Australian Board independence Shareholder wealth, sales Firm size, firm Correlations, A higher board independence Donaldson (1998) firms factor composed of board growth, and profit risk, leverage and rotated factor factor leads to lower subsequent size, CEO duality, performance factor industry rate of analysis, ANOVA shareholder wealth and sales percentage of NEDs and composed of profit return and multivariate growth interest alignment with margin, ROA and ROE regressions owners Calleja (1999) 83 Australian Board size, percentage of Shareholder return None firms NEDs and number of statistics and OLS board committees regressions Lawrence and 100 Australian The percentage of Share price performance, Board size and firm Stapledon (1999) firms independent directors or total assets, revenue, net size executive directors profit, EBIT, number of employees, cash flow, revenue to assets, net profit to revenue, revenue to employees, cash flow to revenue, and 190 Descriptive OLS regressions No significant relation is found The proportion of independent directors is negatively related to the revenue to assets ratio percentage growth in assets, revenue, net profit, EBIT and cash flow Cotter and 109 Australian The percentage of Silverster (2003) firms Market value of equity Firm size Descriptive Neither board nor committee independent directors on statistics, independence is significantly the board, on the audit or correlations and associated with firm value compensation committee, OLS regressions and absence of the CEO from the committee Tobin’s q and ROA Descriptive Tobin’s q is negatively related to outside directors and CEO statistics, the proportion of outside duality correlations, directors, and positively related ANOVA and to board size Kiel and Nicholson 348 Australian Board size, percentage of (2003) firms None multivariate regressions Bonn et al (2004) 104 Australian Board size, and firms and 160 Japanese firms ROA and MBT Firm age and Correlations and For Australian firms, the percentage of outside average age of multivariate proportion of outside directors is directors or female directors regressions positively related to ROA, and directors the female ratio is positively associated with MBT; for Japanese firms, board size and the average age of directors are negatively associated with MBT 191 Balatbat et al 1,316 firm-year The percentage of (2004) observations of Australian 313 Operating return Firm age, leverage, Descriptive There is no evidence that board independent directors and retained ownership, statistics, and OLS composition is associated with CEO duality and the extent to and 2SLS variation in performance; firms which firm is regressions with dual leadership are found to IPOs comprised of asset- perform better in-place rather than options Hutchinson and 310 Australian The ratio of non- Gul (2004) firms ROE Firm size, leverage, Descriptive executive to executive prior performance, statistics, directors executive director correlations and shareholdings, OLS regressions No significant relation is found managers’ remuneration and interaction terms * This table summarises the variables for board composition and structure as defined in Chapter 2 (p.19), and firm performance and control variables used in relevant studies 192 Appendix 2 Summary of Overseas Research Author Pfeffer (1972) Sample 80 U.S. firms Variable* Board Characteristics Firm Performance Board size and percentage Net income to sales of inside directors Control None Test Relevant Findings Correlations and Firms that deviate more from an and net income to multivariate optimal inside-outside director stockholders’ regressions orientation are less successful when investment compared to industry standards than those that deviate less from an optimal board composition Baysinger and 266 U.S. firms Butler (1985) Kenser (1987) 250 U.S. firms The percentage of Industry-adjusted independent directors ROE None Descriptive Firms that had invited relatively statistics, more independent directors onto their correlations and boards in 1970 enjoyed relatively cross-lagged better records of financial regressions performance in 1980 Descriptive The proportion of inside directors is The percentage of inside ROA, ROE, EPS, directors profit margin, stock statistics and positively related to current market performance correlations performance in terms of ROA and None and total return to profit margin, and future investors performance measured by total return to investors Hermalin and 1,521 firm-year The departures or Stock return and Firm size, number and 193 Descriptive Poor stock return leads to the Weisbach (1988) observations of additions of inside or 142 U.S. firms outside directors earnings change median tenure of inside statistics, resignations of inside directors; or outside directors, correlations and outside directors are added after poor CEO retirement and Poisson models performance measured by both stock tenure, current board return and earnings change composition and possible vacancies, and change in the number of industries Molz (1988) 50 U.S. firms Managerial control factor ROA, ROE and total consisting of joint return to shareholders Correlations, There was no significant relationship confirmatory factor between the degree of managerial chairman/CEO, analysis, control on the board and financial chairman/CEO tenure, discriminant performance outside-dominated social analysis, ANOVA responsibility committee, and MANOVA inside versus outside directors, frequency of board meetings, salary ratio of the highest paid to the second highest-paid executive, inside and outside director stockholdings, and woman and minority group representation 194 None Fosberg (1989) 127 pairs of U.S. The percentage of outside ROE, sales, expenses, firms directors number of None Paired sample There is no relationship between the analysis proportion of outside directors and employees, sales to managerial performance total assets, expenses to total assets and number of employees to total assets Schellenger et al 526 U.S. firms (1989) The percentage of outside ROA, ROE, directors None Descriptive ROA and risk-adjusted shareholder shareholder return, statistics and return correlate positively with the and risk-adjusted correlations percentage of outside directors Descriptive The appointments of an outside statistics, director are accompanied by positive frequency excess returns, even though most distribution, and boards are dominated by outsiders parametric and before the appointments shareholder return Rosenstein and 1,251 outside The appointments of only Stock abnormal Wyatt (1990) director one outside director and returns announcements no inside directors None nonparametric tests Hermalin and 1,521 firm-year The percentage of inside Weisbach (1991) observations of or outside directors Tobin’s q and EBIT 142 U.S. firms 195 Firm size, R&D and OLS regressions There does not appear to be a relation advertising expenses, between board composition and firm CEO tenure, median performance, although firms with tenures of insiders and longer median tenures of outsiders outsiders, and family tend to have higher performance company control measured as q Pearce II and 119 U.S. firms Zahra (1992) The percentage of ROA, ROE, EPS and affiliated or independent net profit margin Firm size directors Canonical analysis, Effective past performance is ANOVA, and associated with larger boards and MANOVA lower outsiders’ representation; large boards and high representation of outsiders are associated positively with future performance Daily and Dalton 100 U.S. firms (1992) The number and ROA, ROE and percentage of outside price/earnings ratio None Descriptive There is a positive relation between statistics, total numbers and proportion of directors, and CEO correlations, outside directors and firm duality contingency performance measured by analysis and price/earnings ratio MANOVA Daily and Dalton 186 U.S. firms (1993) The number and ROA, ROE and Firm age and industry Descriptive The results demonstrate a positive percentage of outside price/earnings ratio control statistics, relationship between total numbers directors, and CEO correlations, and proportion of outside directors duality contingency and financial performance analysis and MANOVA Yermack (1996) 3,438 firm-year Board size and percentage Tobin’s q, ROA, Firm size, profitability, Descriptive In the OLS model there is a negative observations of of outside directors sales to assets and investment and growth statistics, OLS association between the percentage return on sales opportunities, regressions, and of outside directors and q; in the diversification, insider fixed-effect, Probit fixed-effect model a positive relation ownership, stockholder- and Poisson is found. Firms are valued more 452 U.S. firms 196 directors, CEO duality, models and industry and highly when the CEO and chairman positions are separated individual year controls Agrawal and 383 U.S. firms Knoeber (1996) The percentage of outside Tobin’s q directors Rosenstein and 170 inside The appointments of one Stock abnormal Wyatt (1997) director or more inside directors returns announcements and not outside directors Klein (1998) 485 U.S. firms Firm size, OLS and 2SLS Fewer outside directors lead to diversification, regressions improved firm performance or, better regulated firms and performance may lead to fewer founding CEOs outsiders on the board None Descriptive The market reaction to the statistics, announcements is negative when frequency inside directors own less than 5% of distribution, and the firm’s stock, and positive when parametric and their ownership level is between 5% nonparametric tests and 25% The percentage of inside ROA, stock abnormal Firm risk, director Descriptive There are positive linkages between directors returns, Jensen shareholdings, director statistics and OLS the percentages of inside directors on Productivity and quality, R&D, regressions finance and investment committees market returns relationship investing, and performance measures; firms capital expenditures, increasing inside director and CEO influence representation on these committees experience higher abnormal returns than firms decreasing the percentage of inside directors on these two committees 197 Barnhart and 321 U.S. firms Rosenstein Board size and percentage Tobin’s q of independent directors (1998) Firm size, leverage, Descriptive There may be a curvilinear relation R&D and advertising statistics, between the proportion of expenses, institutional correlations, and independent directors and firm ownership and industry OLS and 3SLS performance control regressions None Meta-analyses Dalton et al 228 samples of The percentage of inside, Accounting and (1998) U.S. firms outside, affiliate, market performance effect on accounting and market independent or indicators performance indicators; there is no Board composition has virtually no interdependent directors, relationship between board and chairman leadership leadership structure and firm performance Vafeas and 250 U.K. firms The percentage of ROA, MBT, market ROA, sales, leverage, Descriptive There is no significant link between Theodorou independent or “grey” to book value of R&D expenses, statistics and OLS board characteristics and firm (1998) directors on the board, equity, and stock dividend yield and regressions performance percentage of NEDs on return industry control the audit, nomination or remuneration committee, and chairman independence Denis and Sarin 4,563 firm-year Board size and percentage Market-adjusted Firm size, firm age, Descriptive Large changes in outside (1999) observations of of independent directors stock return firm risk, leverage, statistics, representation and board size are growth opportunities, correlations, associated with CEO replacements new CEOs, control frequency and corporate control threats, and 583 U.S. firms 198 threats, and firms in distribution, and stock return appears to be higher which founders left OLS regressions among firms that subsequently increase the proportion of independent directors Bhagat and 934 U.S. firms Black (2000) Board size, and board Tobin’s q, ROA, Firm size, CEO Descriptive There is a correlation between poor independence level market-adjusted stock ownership, outside statistics, and OLS performance and subsequent increase proxied by the percentage price return and sales director ownership, and 3SLS in board independence of independent directors to assets blockholder ownership, regressions minus percentage of firm and industry sales inside directors growth, and industry control Coles et al 144 U.S. firms (2001) The percentage of EVA and MVA independent directors and Firm size, blockholder Descriptive There is a positive contribution of ownership, and industry statistics, CEO duality to EVA, and a negative correlations and influence of independent directors multivariate and CEO salary sensitivity on MVA CEO duality regressions Dehaene et al 122 Belgian Board size, percentage of Industry-adjusted (2001) firms outside directors and CEO ROA and ROE None duality Descriptive There is a positive relation between statistics and the number of outside directors and multivariate ROE; however, where the functions regressions of CEO and chairman are combined, ROA appears higher Hossain et al 633 firm-year Board size, percentage of (2001) observations of independent directors and Tobin’s q 199 Firm size, leverage, Descriptive The proportion of independent diversification and statistics, directors has a positive influence on New Zealand CEO duality growth opportunities firms correlations and firm performance OLS regressions Singh and 236 firm-year Board size, and Asset turnover and Firm size, leverage, Descriptive Boar composition does not seem to Davidson III observations of percentage of inside, SG&A expenses to managerial ownership, statistics and significantly influence agency costs; (2003) U.S. firms affiliated or independent sales blockholder ownership multivariate boar size is negatively related to asset and industry regressions turnover, but unrelated to directors discretionary expenses Tobin’s q Firm size, ROA, capital Descriptive For firms with prior average q less unrelated directors and structure, blockholder statistics and fixed- than one, there is a positive CEO duality ownership, intangibles effect models correlation between the proportion of Panasian et al 274 Canadian Board size, percentage of (2003) firms to total assets, and unrelated directors and performance investment opportunities Anderson and 2,686 firm-year The percentage of Reeb (2004) observations of 403 U.S. firms Peng (2004) Tobin’s q Firm size, firm age, Descriptive Among firms with founding-family independent, affiliated or firm risk, ROA and statistics, fixed- ownership, there is a positive relation family directors investment effect models and between the proportion of opportunities 2SLS regressions independent directors and q 1,211 firm-year The percentage of ROE and sales Firm size, firm age, Descriptive Outside directors do make a observations of affiliated or non-affiliated growth prior performance, statistics, difference in firm performance, if 405 Chinese outside directors and CEO inside or outside correlations, and such performance is measured by firms duality director ownership, and weighted sales growth; they have little impact state ownership and generalized least- on financial performance such as 200 state directors squares and ROE proportional hazards regressions Dulewicz and 86 U.K. firms Herbert (2004) Board size, NED presence CFROTA and sales factor consisting of the turnover None Descriptive The number of executive directors is statistics, positively associated with CFROTA, number and percentage of correlations and and the number and proportion of NEDs in relation to board rotated factor NEDs are negatively related to sales; size or executive analysis companies with a chairman who is directors, and chairman not the CEO or is a NED tend to independence factor perform better in terms of CFROTA consisting of nonexecutive chairman versus executive chairman Randoy and 294 firm-year Board size and percentage Jenssen (2004) observations of of outside directors Tobin’s q and ROE 98 Swedish firms Firm size, firm age, Correlations and Among firms facing high levels of leverage, blockholder OLS regressions product market competition, there is ownership and foreign a negative relation between the exchange listings percentage of outside directors and firm performance measured as q and ROE; there is a positive relation between the percentage of outsiders and q among firms facing low levels of product market competition Chin et al (2004) 426 firm-year Board size and percentage Tobin’s q None 201 Descriptive No significant relation is found observations of of outside directors statistics, New Zealand correlations and firms OLS regressions Chang and Leng 231 firm-year The percentage of NEDs, ROE and dividend (2004) observations of NED chairman on the payout 77 Malaysia audit committee and CEO firms duality 1,648 firm-year Board size, percentage of ROA, ROE, and observations of independent NEDs, MBT 412 Hong Kong firms Chen et al (2005) None Fixed-effect Board characteristics do not have any models influence on firm performance Firm size, leverage and Descriptive Board composition has little impact sales growth statistics, on firm performance; there is a presence of audit frequency negative association between CEO committee, outsider- distribution and duality and MBT dominated board and fixed-effect models CEO duality Krivogorsky 87 continental The percentage of inside (2006) European firms directors, independent Firm size, firm age, Descriptive The percentage of independent leverage and growth statistics, directors has a positive correlation directors, or scholars and correlations and with ROA and ROE CEO duality OLS regressions Luan and Tang 259 Taiwanese Independent director (2007) firms assignment Choi et al (2007) 1,834 firm-year Board size, percentage of ROA, ROE and MBT ROE Tobin’s q Firm size, prior Descriptive After controlling for a firm’s past performance and statistics, performance, independent director absorptive capacity, correlations and appointments have a positive impact OLS regressions on a firm’s performance Descriptive The results indicate that outside Firm size, firm age, 202 observations of outside, independent or leverage, chaebol statistics, directors have significant and South Korea affiliated directors, and affiliation, depository correlations and positive effects on firm performance; firms foreign outside directors receipt, exports to sales, random-effect the effects are stronger for R&D expenditure to models independent directors than for “grey” sales, ROA, beta, and directors who may have professional industry and individual ties with the firm year controls Chan and Li (2008) * 200 U.S. firms Board size, committee Tobin’s q Firm size, number of Descriptive The independence of audit committee size, percentage of employees, director’s statistics and results in higher firm value when a independent or expert- age and tenure, number simultaneous majority of expert-independent independent directors on of other directorships, equation models directors serve on the board; the the board or audit Fortune 500 rank, ROA presence of CEO duality is related to committee, and CEO and holding period negative q duality return This table summarises the variables for board composition and structure as defined in Chapter 2 (p.19), and firm performance and control variables used in relevant studies 203 Appendix 3 Theories and Hypotheses in Prior Research Author Pfeffer (1972) Theory Organizational theories Hypothesis* Organizations that deviate relatively more from a Conclusion The hypothesis is supported preferred inside-outside director orientation should be relatively less successful when compared to industry standards than those that deviate less from a preferred board composition Baysinger and Butler Legalistic view, agency theory and (1985) transaction costs, and strategy board composition would be inappropriate in an economy formulation and implementation where firms are different and are continually changing Kenser (1987) Hermalin and Financial dependence perspective None Weisbach (1988) Molz (1988) None A public policy prescription in favour of any particular The proportion of inside directors serving on a firm’s Higher inside representation is associated with greater board is positively related to organizational performance profitability and higher asset utilization The CEO succession process and firm performance will The findings are consistent with both hypotheses affect board composition None Firms with managerial dominated boards will have The hypothesis is not supported superior financial performance Fosberg (1989) Schellenger et al Agency theory Agency theory Increasing the percentage of outside directors on the Agency theory could not be confirmed by the analysis in board enhances management performance this study None The findings provides support for advocates of outsider (1989) representation on the boards of corporations 204 Rosenstein and Wyatt None None The addition of an outside director increases firm value; (1990) outside directors are selected, on average, in the interests of shareholders Hermalin and Agency theory None There appears to be no relation between board Weisbach (1991) composition and firm performance Pearce II and Zahra Strategic contingency model (i.e., (1992) resource dependence theory) Poor past performance is associated positively with The strategic contingency approach is a viable means of small board size and low outsider’ representation; studying board of directors’ composition and Large board size and higher representation of outside directors on the board are associated positively with future corporate financial performance Daily and Dalton None (1992) CEO duality will be associated with lower firm There is modest support for the proposition that firms performance; with greater numbers of outside directors outperform Fewer total outside directors will be associated with those with fewer outside directors lower firm performance; and The proportion of outside directors will be associated with higher firm performance Daily and Dalton None (1993) CEO duality will be associated with lower firm It seems that firms adhering to suggested board reforms performance; realize performance advantages Fewer total outside directors will be associated with lower firm performance; and Proportion of outside directors will be associated 205 with higher firm performance Yermack (1996) None None The effect of the board composition variable is ambiguous and appears sensitive to the inclusion of firm effects in the model Agrawal and Agency theory None The negative effect of outsiders on the board on firm Knoeber (1996) performance suggests that firms tend to have too many outside directors Rosenstein and Wyatt Agency theory None The expected benefits of an inside director’s expert (1997) knowledge outweigh the expected costs of managerial entrenchment only when managerial and outside shareholder interests are closely aligned Muth and Donaldson Agency theory, stewardship theory (1998) and resource dependence theory Where board members are more independent of Agency theory predictions relating to board independence management, company performance will be higher and firm performance are not upheld while those of (agency theory); and stewardship theory are supported Boards with a lower level of board independence lead to higher company performance (stewardship theory) Klein (1998) Agency theory None Inside directors provide valuable information to boards about the firms’ long-term investment decisions Barnhart and None None There may be a curvilinear relation between the Rosenstein (1998) proportion of independent directors and firm performance; board composition, managerial ownership 206 and performance may be jointly determined Dalton et al (1998) Agency theory, stewardship theory None The findings provide support for neither agency theory and resource dependence theory Vafeas and Agency theory nor stewardship theory Theodorou (1998) There is a positive association between the fraction The results are consistent with governance needs varying of non-executive directors on the board and firm across firms, and contrast the notion that uniform board value; structures should be mandated There is a positive association between the split in the roles of the CEO and the chairman of the board (dual leadership structure) and firm value; and There is a positive association between the fraction of non-executive directors serving on monitoring committees and firm value Calleja (1999) None None The small sample size made it difficult to reach any firm conclusions Lawrence and Agency theory None Independent directors did not appear to have added value Stapledon (1999) Denis and Sarin to Australian largest listed companies None None The predominant factors associated with ownership and (1999) control changes appear to be top executive changes, prior stock price performance and corporate control threats Bhagat and Black None None There is a reasonably strong correlation between poor (2000) performance and subsequent increase in board independence; there is no evidence that greater board 207 independence leads to improved firm performance Coles et al (2001) Agency theory and stewardship theory Firms that separate the positions of CEO and chair of While there is evidence to support that firms may use the board will have better performance than will governance packages to deal with agency issues, further firms that join the two positions; and research could provide important evidence on these Firms that select higher proportions of independent issues by focusing on examining a more refined, outsiders to serve on their board of directors will industry-level context have better performance than will firms with a higher proportion of insides on the board Dehaene et al (2001) Agency theory None Distinct corporate governance models for companies exist because they operate in different business context; comparing these models in isolation can lead to futile conclusions Hossain et al (2001) Agency theory None The legislation, directly designed to increase the fiduciary responsibilities of directors, did not seem to enhance or weaken the positive relationship between outside board representation and firm performance Cotter and Silverster Agency theory (2003) Kiel and Nicholson Agency theory, stewardship theory (2003) and resource dependence theory There is a positive relationship between the independence There is no evidence that firm value is enhanced through of boards of directors (and their monitoring committees) stronger monitoring committee or full board and firm value independence Board size is positively correlated with firm The arguments put forward by stewardship theory are performance; supported The proportion of outside directors is uncorrelated 208 with firm performance; and Separation of chairman and CEO is uncorrelated with firm performance Singh and Davidson Agency theory None Independent outsiders on a board do not appear to protect III (2003) Panasian et al (2003) the firm from agency costs Agency theory None The increase of outside directors is most beneficial for firms that are most likely to have agency problems Bonn et al (2004) Agency theory and resource dependence theory Board size is positively associated with firm There may be different agency relationships in different performance in Australian firms; countries Board size is negatively associated with firm performance in Japanese firms; The proportion of outside directors on the board is positively associated with firm performance for Australian firms; and The proportion of outside directors on the board is negatively associated with firm performance for Japanese firms Balatbat et al (2004) Agency theory None The circumstances faced by IPO firms may frequently render more traditional corporate governance practices irrelevant, especially as they relate to monitoring managers who provide highly firm-specific human capital as a key component of the firm’s value 209 Hutchinson and Gul Agency theory None The results demonstrate the importance of corporate (2004) governance for firms with more growth opportunities Anderson and Reeb Agency theory and stewardship (2004) theory The greater the fraction of independent directors in The findings highlight the importance of independent public firms with founding-family ownership, the directors in mitigating conflicts between shareholder better the performance of the firm; groups and imply that the interests of minority investors At low (high) levels of family board representation are best protected when, through independent directors, relative to independent directors, the higher the ratio they have power relative to family shareholders of family directors to independent directors, the better (poorer) the performance of the firm; and The greater the fraction of affiliated directors on the board in public firms with founding-family ownership, the poorer (agency theory) better (stewardship theory) the performance of the firm Peng (2004) Agency theory, resource dependence theory and institutional theory Greater representation on the board by outside While agency theory hypotheses typically fail to be directors has a positive effect on firm performance; supported in emerging economies, resource dependence Greater representation on the board by outside and institutional claims tend to be substantiated directors has a negligible effect on firm performance; There is a positive relationship between poor prior performance of the firm in a previous year and the appointment of outside directors to the board in a given year; There is a positive relationship between a young firm age in a previous year and the appointment of outside 210 directors to the board in a given year; and The strength of the relationship between (a) poor prior performance, (b) large firm size, and (c) young firm age on one hand and the appointment of outside directors to the board on the other hand will decrease over time Dulewicz and Agency theory, stewardship theory Herbert (2004) and stakeholder theory Where the roles of chairman and CEO are separated The results generally provide scant support for the main (agency theory) combined (stewardship theory), theories of board of directors company performance will be higher; A higher proportion of outside directors (agency theory) executive directors (stewardship theory) on the board leads to higher company performance; Larger (agency theory) smaller (stewardship theory) boards will have higher company performance; and Where board members are more (agency theory) less (stewardship theory) independent of management, company performance will be higher Randoy and Jenssen Agency theory and stewardship (2004) theory Board independence has a negative influence on firm Board independence is less relevant or even redundant in performance in firms that face a high level of product highly competitive industries, where the firm is already market competition, and monitored by a competitive product market; board Board independence has a positive influence on firm independence enhances firm performance among performance in firms that face a less competitive companies facing less competitive product market product market 211 Chin et al (2004) Agency theory There is a positive correlation between the There does not seem to be a rise-fall relationship in percentage of outside directors and firm performance relating to the percentage of outside performance; and directors, nor to board size There is an inverse relationship between board size and firm performance Chang and Leng Agency theory None Several corporate governance variables do not have any (2004) impact on corporate performance in Malaysia, including the proportion of NEDs, CEO duality and NED chairman of the audit committee Chen et al (2005) Agency theory None More effort is needed in order to ensure the true independence of non-executive directors and that they are able to perform an adequate monitoring function Krivogorsky (2006) Agency theory The proportion of independent directors and scholars The theoretical predictions of agency theory on a positive on the board has a strong effect on the profitability of relationship between outside (independent) director and the company; firm performance are also applicable in the European The proportion of inside directors does not have a environment strong effect on the profitability of the company; and power concentration (CEO=Chairman) negatively affects the supervisory ability of the board and, thus, the profitability of the company Luan and Tang Agency theory, stewardship theory (2007) and resource dependence theory Ceteris paribus, appointing independent outside The findings suggest that firms that chose to appoint directors on the board has a positive effect on firm independent outside directors have a higher corporate 212 performance; and profit Absorptive capacity of the firms has a positive effect on firm performance when appointing independent outside directors on the board Choi et al (2007) Agency theory Firm Performance increases with board independence The presence of independent outsiders is critical in an emerging market that is subject to external shocks and that may lack sufficient liquidity as well as indigenous institutional infrastructure Chan and Li (2008) * Agency theory Independence of audit committee does not affect firm The presence of expert-independent directors on the value board and audit committee enhances firm value This table exhibits the hypotheses explicitly stated in the studies, regarding the relationships between board composition and structure as defined in Chapter 2 (p.19), and firm performance 213 Appendix 4 Pearson Correlations: 2000-2003 Sample Period: 2000-2003 Included Observations: 243 Variable Correlation t-Statistic Probability FIND FIND 1.000000 ----- ----- ACIND FIND 0.752380 17.73122 0.0000 ACIND ACIND 1.000000 ----- ----- NCIND FIND 0.441090 7.629908 0.0000 NCIND ACIND 0.367563 6.135615 0.0000 NCIND NCIND 1.000000 ----- ----- RCIND FIND 0.622563 12.35004 0.0000 RCIND ACIND 0.551664 10.26792 0.0000 RCIND NCIND 0.523030 9.526538 0.0000 RCIND RCIND 1.000000 ----- ----- CMIND FIND 0.531313 9.736099 0.0000 CMIND ACIND 0.357522 5.943042 0.0000 CMIND NCIND 0.265020 4.266787 0.0000 CMIND RCIND 0.440519 7.617651 0.0000 CMIND CMIND 1.000000 ----- ----- ROA1 FIND -0.012554 -0.194902 0.8456 ROA1 ACIND 0.076213 1.186596 0.2366 ROA1 NCIND 0.038183 0.593186 0.5536 ROA1 RCIND -0.005512 -0.085569 0.9319 ROA1 CMIND -0.070382 -1.095334 0.2745 ROA1 ROA1 1.000000 ----- ----- ROE1 FIND 0.071257 1.109022 0.2685 ROE1 ACIND 0.059075 0.918697 0.3592 ROE1 NCIND 0.046069 0.715949 0.4747 ROE1 RCIND 0.020257 0.314545 0.7534 214 ROE1 CMIND 0.003849 0.059747 0.9524 ROE1 ROA1 0.312705 5.110788 0.0000 ROE1 ROE1 1.000000 ----- ----- SHRET1 FIND -0.063596 -0.989275 0.3235 SHRET1 ACIND -0.049272 -0.765830 0.4445 SHRET1 NCIND -0.125631 -1.965900 0.0505 SHRET1 RCIND -0.155481 -2.443429 0.0153 SHRET1 CMIND -0.076792 -1.195667 0.2330 SHRET1 ROA1 -0.127047 -1.988416 0.0479 SHRET1 ROE1 -0.089536 -1.395585 0.1641 SHRET1 SHRET1 1.000000 ----- ----- TOBQ1 FIND 0.012342 0.191612 0.8482 TOBQ1 ACIND -0.069391 -1.079844 0.2813 TOBQ1 NCIND -0.095079 -1.482743 0.1394 TOBQ1 RCIND 0.031365 0.487150 0.6266 TOBQ1 CMIND 0.006217 0.096509 0.9232 TOBQ1 ROA1 -0.366648 -6.117966 0.0000 TOBQ1 ROE1 -0.160041 -2.516939 0.0125 TOBQ1 SHRET1 0.043346 0.673538 0.5013 TOBQ1 TOBQ1 1.000000 ----- ----- SIZE FIND 0.134638 2.109356 0.0359 SIZE ACIND 0.221020 3.518154 0.0005 SIZE NCIND 0.332164 5.466970 0.0000 SIZE RCIND 0.302538 4.927567 0.0000 SIZE CMIND 0.001162 0.018043 0.9856 SIZE ROA1 0.191112 3.022564 0.0028 SIZE ROE1 0.082249 1.281182 0.2014 SIZE SHRET1 -0.109309 -1.707158 0.0891 SIZE TOBQ1 -0.226900 -3.616774 0.0004 215 SIZE SIZE 1.000000 ----- ----- BLOCK FIND -0.235016 -3.753562 0.0002 BLOCK ACIND -0.194588 -3.079688 0.0023 BLOCK NCIND -0.099272 -1.548766 0.1228 BLOCK RCIND -0.179168 -2.827188 0.0051 BLOCK CMIND -0.112494 -1.757532 0.0801 BLOCK ROA1 0.098244 1.532577 0.1267 BLOCK ROE1 0.142809 2.239950 0.0260 BLOCK SHRET1 0.013232 0.205433 0.8374 BLOCK TOBQ1 -0.056240 -0.874467 0.3827 BLOCK SIZE 0.050692 0.787958 0.4315 BLOCK BLOCK 1.000000 ----- ----- SEGMT FIND 0.178790 2.821021 0.0052 SEGMT ACIND 0.132247 2.071221 0.0394 SEGMT NCIND 0.274461 4.430932 0.0000 SEGMT RCIND 0.193908 3.068509 0.0024 SEGMT CMIND -0.005431 -0.084309 0.9329 SEGMT ROA1 0.166541 2.622023 0.0093 SEGMT ROE1 0.066234 1.030499 0.3038 SEGMT SHRET1 -0.100931 -1.574920 0.1166 SEGMT TOBQ1 -0.172673 -2.721477 0.0070 SEGMT SIZE 0.463862 8.128477 0.0000 SEGMT BLOCK -0.110906 -1.732414 0.0845 SEGMT SEGMT 1.000000 ----- ----- DIVR1 FIND 0.048966 0.761068 0.4474 DIVR1 ACIND 0.104408 1.629755 0.1045 DIVR1 NCIND 0.145403 2.281512 0.0234 DIVR1 RCIND 0.179599 2.834212 0.0050 DIVR1 CMIND 0.062330 0.969501 0.3333 DIVR1 ROA1 0.327495 5.380822 0.0000 216 DIVR1 ROE1 0.163276 2.569198 0.0108 DIVR1 SHRET1 -0.135804 -2.127953 0.0344 DIVR1 TOBQ1 -0.229225 -3.655879 0.0003 DIVR1 SIZE 0.236187 3.773373 0.0002 DIVR1 BLOCK 0.037580 0.583805 0.5599 DIVR1 SEGMT 0.165504 2.605242 0.0098 DIVR1 DIVR1 1.000000 ----- ----- AGE FIND 0.136348 2.136643 0.0336 AGE ACIND 0.121245 1.896216 0.0591 AGE NCIND 0.134395 2.105475 0.0363 AGE RCIND 0.098822 1.541673 0.1245 AGE CMIND 0.006702 0.104043 0.9172 AGE ROA1 0.135286 2.119697 0.0351 AGE ROE1 0.071293 1.109583 0.2683 AGE SHRET1 0.000659 0.010226 0.9918 AGE TOBQ1 -0.206636 -3.278609 0.0012 AGE SIZE 0.194783 3.082895 0.0023 AGE BLOCK -0.068960 -1.073100 0.2843 AGE SEGMT 0.278588 4.503129 0.0000 AGE DIVR1 0.186591 2.948445 0.0035 AGE AGE 1.000000 ----- ----- Log(MCAP1) FIND 0.215770 3.430458 0.0007 Log(MCAP1) ACIND 0.204916 3.250120 0.0013 Log(MCAP1) NCIND 0.434249 7.483795 0.0000 Log(MCAP1) RCIND 0.347302 5.749455 0.0000 Log(MCAP1) CMIND 0.054389 0.845592 0.3986 Log(MCAP1) ROA1 0.274635 4.433977 0.0000 Log(MCAP1) ROE1 0.114678 1.792106 0.0744 Log(MCAP1) SHRET1 -0.136846 -2.144589 0.0330 Log(MCAP1) TOBQ1 -0.072846 -1.133889 0.2580 217 Log(MCAP1) SIZE 0.682392 14.49218 0.0000 Log(MCAP1) BLOCK -0.007428 -0.115323 0.9083 Log(MCAP1) SEGMT 0.554557 10.34559 0.0000 Log(MCAP1) DIVR1 0.298028 4.846890 0.0000 Log(MCAP1) AGE 0.294663 4.786941 0.0000 Log(MCAP1) Log(MCAP1) 1.000000 ----- ----- GEAR1 FIND 0.099450 1.551574 0.1221 GEAR1 ACIND 0.098929 1.543360 0.1241 GEAR1 NCIND 0.089434 1.393982 0.1646 GEAR1 RCIND 0.112031 1.750202 0.0814 GEAR1 CMIND 0.081559 1.270372 0.2052 GEAR1 ROA1 0.095074 1.482669 0.1395 GEAR1 ROE1 0.081022 1.261956 0.2082 GEAR1 SHRET1 -0.014963 -0.232319 0.8165 GEAR1 TOBQ1 -0.085129 -1.326366 0.1860 GEAR1 SIZE 0.113373 1.771441 0.0778 GEAR1 BLOCK 0.024304 0.377409 0.7062 GEAR1 SEGMT -0.038995 -0.605822 0.5452 GEAR1 DIVR1 0.081045 1.262312 0.2081 GEAR1 AGE 0.054459 0.846689 0.3980 GEAR1 Log(MCAP1) 0.153891 2.417837 0.0164 GEAR1 GEAR1 1.000000 ----- ----- EQED FIND -0.146766 -2.303371 0.0221 EQED ACIND -0.104789 -1.635771 0.1032 EQED NCIND -0.149318 -2.344318 0.0199 EQED RCIND -0.103623 -1.617373 0.1071 EQED CMIND -0.072204 -1.123845 0.2622 EQED ROA1 -0.012381 -0.192223 0.8477 EQED ROE1 0.183276 2.894227 0.0041 EQED SHRET1 0.100503 1.568169 0.1182 218 EQED TOBQ1 0.218672 3.478904 0.0006 EQED SIZE -0.179533 -2.833130 0.0050 EQED BLOCK 0.243215 3.892606 0.0001 EQED SEGMT -0.183199 -2.892973 0.0042 EQED DIVR1 -0.034604 -0.537522 0.5914 EQED AGE -0.185882 -2.936853 0.0036 EQED Log(MCAP1) -0.257292 -4.133399 0.0000 EQED GEAR1 -0.039258 -0.609918 0.5425 EQED EQED 1.000000 ----- ----- RISK1 FIND -0.062750 -0.976068 0.3300 RISK1 ACIND -0.045748 -0.710939 0.4778 RISK1 NCIND -0.104178 -1.626130 0.1052 RISK1 RCIND -0.131980 -2.066969 0.0398 RISK1 CMIND -0.076923 -1.197711 0.2322 RISK1 ROA1 -0.170313 -2.683162 0.0078 RISK1 ROE1 -0.118134 -1.846860 0.0660 RISK1 SHRET1 0.977674 72.22970 0.0000 RISK1 TOBQ1 0.045601 0.708659 0.4792 RISK1 SIZE -0.102204 -1.594979 0.1120 RISK1 BLOCK 0.030118 0.467769 0.6404 RISK1 SEGMT -0.076942 -1.198007 0.2321 RISK1 DIVR1 -0.169281 -2.666428 0.0082 RISK1 AGE -0.020402 -0.316786 0.7517 RISK1 Log(MCAP1) -0.131366 -2.057178 0.0407 RISK1 GEAR1 -0.046672 -0.725338 0.4689 RISK1 EQED 0.109994 1.717998 0.0871 RISK1 RISK1 1.000000 ----- ----- 219 Appendix 5 Pearson Correlations: 2003-2006 Sample Period: 2003-2006 Included Observations: 243 Variable Correlation t-Statistic Probability FIND FIND 1.000000 ----- ----- ACIND FIND 0.752380 17.73122 0.0000 ACIND ACIND 1.000000 ----- ----- NCIND FIND 0.441090 7.629908 0.0000 NCIND ACIND 0.367563 6.135615 0.0000 NCIND NCIND 1.000000 ----- ----- RCIND FIND 0.622563 12.35004 0.0000 RCIND ACIND 0.551664 10.26792 0.0000 RCIND NCIND 0.523030 9.526538 0.0000 RCIND RCIND 1.000000 ----- ----- CMIND FIND 0.531313 9.736099 0.0000 CMIND ACIND 0.357522 5.943042 0.0000 CMIND NCIND 0.265020 4.266787 0.0000 CMIND RCIND 0.440519 7.617651 0.0000 CMIND CMIND 1.000000 ----- ----- ROA2 FIND -0.046246 -0.718706 0.4730 ROA2 ACIND -0.003196 -0.049621 0.9605 ROA2 NCIND -0.018568 -0.288305 0.7734 ROA2 RCIND -0.035684 -0.554320 0.5799 ROA2 CMIND -0.078351 -1.220083 0.2236 ROA2 ROA2 1.000000 ----- ----- ROE2 FIND -0.054760 -0.851385 0.3954 ROE2 ACIND -0.045002 -0.699333 0.4850 ROE2 NCIND -0.028291 -0.439366 0.6608 ROE2 RCIND -0.023827 -0.370005 0.7117 220 ROE2 CMIND -0.065815 -1.023942 0.3069 ROE2 ROA2 0.328017 5.390445 0.0000 ROE2 ROE2 1.000000 ----- ----- SHRET2 FIND -0.022836 -0.354610 0.7232 SHRET2 ACIND -0.040670 -0.631896 0.5281 SHRET2 NCIND -0.051806 -0.805329 0.4214 SHRET2 RCIND -0.032458 -0.504146 0.6146 SHRET2 CMIND 0.004554 0.070691 0.9437 SHRET2 ROA2 0.004599 0.071397 0.9431 SHRET2 ROE2 0.029088 0.451761 0.6518 SHRET2 SHRET2 1.000000 ----- ----- TOBQ2 FIND 0.073300 1.140987 0.2550 TOBQ2 ACIND -0.021016 -0.326333 0.7445 TOBQ2 NCIND -0.000980 -0.015212 0.9879 TOBQ2 RCIND 0.003885 0.060315 0.9520 TOBQ2 CMIND 0.091602 1.428048 0.1546 TOBQ2 ROA2 -0.727304 -16.45134 0.0000 TOBQ2 ROE2 -0.197391 -3.125840 0.0020 TOBQ2 SHRET2 -0.009940 -0.154325 0.8775 TOBQ2 TOBQ2 1.000000 ----- ----- SIZE FIND 0.134638 2.109356 0.0359 SIZE ACIND 0.221020 3.518154 0.0005 SIZE NCIND 0.332164 5.466970 0.0000 SIZE RCIND 0.302538 4.927567 0.0000 SIZE CMIND 0.001162 0.018043 0.9856 SIZE ROA2 0.152030 2.387902 0.0177 SIZE ROE2 0.067511 1.050446 0.2946 SIZE SHRET2 -0.027501 -0.427090 0.6697 SIZE TOBQ2 -0.193652 -3.064286 0.0024 221 SIZE SIZE 1.000000 ----- ----- BLOCK FIND -0.235016 -3.753562 0.0002 BLOCK ACIND -0.194588 -3.079688 0.0023 BLOCK NCIND -0.099272 -1.548766 0.1228 BLOCK RCIND -0.179168 -2.827188 0.0051 BLOCK CMIND -0.112494 -1.757532 0.0801 BLOCK ROA2 -0.011262 -0.174837 0.8614 BLOCK ROE2 -0.054571 -0.848427 0.3970 BLOCK SHRET2 0.136493 2.138957 0.0334 BLOCK TOBQ2 0.046301 0.719556 0.4725 BLOCK SIZE 0.050692 0.787958 0.4315 BLOCK BLOCK 1.000000 ----- ----- SEGMT FIND 0.178790 2.821021 0.0052 SEGMT ACIND 0.132247 2.071221 0.0394 SEGMT NCIND 0.274461 4.430932 0.0000 SEGMT RCIND 0.193908 3.068509 0.0024 SEGMT CMIND -0.005431 -0.084309 0.9329 SEGMT ROA2 0.114218 1.784816 0.0755 SEGMT ROE2 0.161249 2.536444 0.0118 SEGMT SHRET2 -0.047912 -0.744654 0.4572 SEGMT TOBQ2 -0.132743 -2.079120 0.0387 SEGMT SIZE 0.463862 8.128477 0.0000 SEGMT BLOCK -0.110906 -1.732414 0.0845 SEGMT SEGMT 1.000000 ----- ----- DIVR2 FIND 0.040177 0.624224 0.5331 DIVR2 ACIND 0.098798 1.541292 0.1246 DIVR2 NCIND 0.077777 1.211099 0.2270 DIVR2 RCIND 0.108367 1.692271 0.0919 DIVR2 CMIND 0.059763 0.929440 0.3536 DIVR2 ROA2 0.192989 3.053394 0.0025 222 DIVR2 ROE2 0.058814 0.914628 0.3613 DIVR2 SHRET2 -0.098499 -1.536583 0.1257 DIVR2 TOBQ2 -0.132364 -2.073081 0.0392 DIVR2 SIZE 0.196521 3.111510 0.0021 DIVR2 BLOCK 0.112978 1.765184 0.0788 DIVR2 SEGMT 0.084908 1.322906 0.1871 DIVR2 DIVR2 1.000000 ----- ----- AGE FIND 0.136348 2.136643 0.0336 AGE ACIND 0.121245 1.896216 0.0591 AGE NCIND 0.134395 2.105475 0.0363 AGE RCIND 0.098822 1.541673 0.1245 AGE CMIND 0.006702 0.104043 0.9172 AGE ROA2 0.121209 1.895650 0.0592 AGE ROE2 0.088235 1.375133 0.1704 AGE SHRET2 0.032754 0.508747 0.6114 AGE TOBQ2 -0.118192 -1.847786 0.0659 AGE SIZE 0.194783 3.082895 0.0023 AGE BLOCK -0.068960 -1.073100 0.2843 AGE SEGMT 0.278588 4.503129 0.0000 AGE DIVR2 0.120315 1.881459 0.0611 AGE AGE 1.000000 ----- ----- Log(MCAP2) FIND 0.231315 3.691076 0.0003 Log(MCAP2) ACIND 0.197470 3.127135 0.0020 Log(MCAP2) NCIND 0.387699 6.529388 0.0000 Log(MCAP2) RCIND 0.310960 5.079215 0.0000 Log(MCAP2) CMIND 0.080945 1.260743 0.2086 Log(MCAP2) ROA2 0.237853 3.801569 0.0002 Log(MCAP2) ROE2 0.086819 1.352895 0.1774 Log(MCAP2) SHRET2 0.038968 0.605413 0.5455 Log(MCAP2) TOBQ2 -0.100862 -1.573827 0.1168 223 Log(MCAP2) SIZE 0.645477 13.11962 0.0000 Log(MCAP2) BLOCK -0.009569 -0.148554 0.8820 Log(MCAP2) SEGMT 0.500235 8.968501 0.0000 Log(MCAP2) DIVR2 0.262399 4.221455 0.0000 Log(MCAP2) AGE 0.327850 5.387357 0.0000 Log(MCAP2) Log(MCAP2) 1.000000 ----- ----- GEAR2 FIND 0.005548 0.086127 0.9314 GEAR2 ACIND -0.016345 -0.253784 0.7999 GEAR2 NCIND 0.058644 0.911974 0.3627 GEAR2 RCIND -0.013825 -0.214649 0.8302 GEAR2 CMIND 0.047822 0.743254 0.4581 GEAR2 ROA2 -0.047507 -0.738336 0.4610 GEAR2 ROE2 -0.668375 -13.94949 0.0000 GEAR2 SHRET2 -0.014255 -0.221319 0.8250 GEAR2 TOBQ2 0.007060 0.109601 0.9128 GEAR2 SIZE -0.007503 -0.116475 0.9074 GEAR2 BLOCK -0.026620 -0.413405 0.6797 GEAR2 SEGMT -0.035968 -0.558728 0.5769 GEAR2 DIVR2 0.096079 1.498486 0.1353 GEAR2 AGE -0.063307 -0.984768 0.3257 GEAR2 Log(MCAP2) 0.072461 1.127860 0.2605 GEAR2 GEAR2 1.000000 ----- ----- EQED FIND -0.146766 -2.303371 0.0221 EQED ACIND -0.104789 -1.635771 0.1032 EQED NCIND -0.149318 -2.344318 0.0199 EQED RCIND -0.103623 -1.617373 0.1071 EQED CMIND -0.072204 -1.123845 0.2622 EQED ROA2 -0.161353 -2.538129 0.0118 EQED ROE2 -0.084041 -1.309302 0.1917 EQED SHRET2 -0.144474 -2.266621 0.0243 224 EQED TOBQ2 0.190904 3.019147 0.0028 EQED SIZE -0.179533 -2.833130 0.0050 EQED BLOCK 0.243215 3.892606 0.0001 EQED SEGMT -0.183199 -2.892973 0.0042 EQED DIVR2 0.060777 0.945266 0.3455 EQED AGE -0.185882 -2.936853 0.0036 EQED Log(MCAP2) -0.321018 -5.262038 0.0000 EQED GEAR2 0.055542 0.863572 0.3887 EQED EQED 1.000000 ----- ----- RISK2 FIND -0.078967 -1.229744 0.2200 RISK2 ACIND -0.080858 -1.259384 0.2091 RISK2 NCIND -0.099949 -1.559433 0.1202 RISK2 RCIND -0.066631 -1.036691 0.3009 RISK2 CMIND -0.013159 -0.204306 0.8383 RISK2 ROA2 -0.043583 -0.677241 0.4989 RISK2 ROE2 -0.009704 -0.150647 0.8804 RISK2 SHRET2 0.954684 49.79724 0.0000 RISK2 TOBQ2 -0.008087 -0.125556 0.9002 RISK2 SIZE -0.119215 -1.864010 0.0635 RISK2 BLOCK 0.121476 1.899885 0.0586 RISK2 SEGMT -0.108621 -1.696283 0.0911 RISK2 DIVR2 -0.175115 -2.761174 0.0062 RISK2 AGE -0.028643 -0.444847 0.6568 RISK2 Log(MCAP2) -0.135237 -2.118916 0.0351 RISK2 GEAR2 0.002996 0.046518 0.9629 RISK2 EQED -0.076994 -1.198821 0.2318 RISK2 RISK2 1.000000 ----- ----- 225 Appendix 6 OLS and Logit Regressions: Board Independence and Past Performance (ROA) Sample Period: 2000-2003 Included Observations: 243 Coefficient t/z-Statistic FIND ACIND NCIND RCIND CMIND Intercept 0.474 0.557 -0.032 0.269 0.659 6.793** 5.285** -0.316 2.461* 0.327 -0.039 0.037 -0.073 -0.138 -0.814 -0.772 0.479 -1.000 -1.731 -1.517 0.007 0.033 0.039 0.046 -0.009 0.922 2.742** 3.432** 3.741** -0.124 -0.255 -0.383 -0.144 -0.364 -1.066 -3.193** -3.178** -1.258 -2.922** -1.392 0.010 -0.001 0.020 0.007 -0.021 1.426 -0.122 1.887 0.632 -0.300 0.006 0.028 0.048 0.101 0.381 0.172 0.579 1.041 1.989* 1.192 0.0009 0.001 0.0005 0.0002 -0.001 1.063 0.812 0.418 0.149 -0.137 0.015 0.016 0.014 0.021 0.145 1.548 1.131 1.039 1.384 1.101 -0.054 -0.010 -0.083 0.029 -0.533 -0.667 -0.079 -0.718 0.231 -0.677 -0.003 -0.0003 -0.007 -0.013 -0.102 -0.589 -0.040 -0.926 -1.471 -1.034 0.106 0.104 0.155 0.166 0.031 0.071 0.070 0.123 0.134 Std Error (Regression) 0.214 0.323 0.308 0.334 0.500 F/LR-Statistic 3.068 3.012 4.761 5.161 10.291 Probability (F/LR-Statistic) 0.002 0.002 0.000008 0.000002 0.327 Durbin-Watson 1.702 1.912 2.157 1.7870.658 ROA1 SIZE BLOCK SEGMT DIVR1 AGE GEAR1 EQED RISK1 R 2 /McFadden R 2 Adjusted R * 2 Significance at the 5% level ** 226 Significance at the 1% level Appendix 7 OLS and Logit Regressions: Board Independence and Past Performance (ROE) Sample Period: 2000-2003 Included Observations: 243 Coefficient t/z-Statistic FIND ACIND NCIND RCIND CMIND Intercept 0.502 0.561 -0.004 0.308 0.890 7.305** 5.400** -0.040 2.843** 1.346 0.033 0.033 0.011 -0.006 0.027 1.403 0.925 0.316 -0.171 0.120 0.007 0.033 0.038 0.045 -0.016 0.849 2.751** 3.368** 3.633** -0.210 -0.272 -0.388 -0.159 -0.383 -1.200 -3.418** -3.228** -1.385 -3.055** -1.574 0.009 -0.001 0.019 0.006 -0.028 1.296 -0.131 1.788 0.496 -0.404 -0.006 0.029 0.035 0.079 0.248 -0.187 0.622 0.767 1.593 0.811 0.0008 0.001 0.0004 8.05E-05 -0.002 0.928 0.778 0.339 0.062 -0.235 0.013 0.016 0.013 0.019 0.128 1.403 1.098 0.949 1.275 1.035 -0.078 -0.031 -0.094 0.028 -0.557 -0.955 -0.250 -0.791 0.215 -0.703 -0.002 7.16E-05 -0.006 -0.011 -0.072 -0.335 0.009 -0.768 -1.273 -0.831 0.111 0.107 0.152 0.156 0.023 0.077 0.072 0.119 0.123 Std Error (Regression) 0.213 0.323 0.309 0.336 0.502 F/LR-Statistic 3.238 3.090 4.643 4.771 7.723 Probability (F/LR-Statistic) 0.001 0.002 0.00001 0.000008 0.562 Durbin-Watson 1.627 1.888 2.138 1.787 ROE1 SIZE BLOCK SEGMT DIVR1 AGE GEAR1 EQED RISK1 R 2 //McFadden R 2 Adjusted R * 2 Significance at the 5% level ** 227 Significance at the 1% level Appendix 8 OLS and Logit Regressions: Board Independence and Past Performance (Shareholder Return) Sample Period: 2000-2003 Included Observations: 243 Coefficient t/z-Statistic FIND ACIND NCIND RCIND CMIND Intercept 0.489 0.551 0.002 0.327 0.905 7.168** 5.360** 0.016 3.093** 1.381 -0.026 -0.060 -0.129 -0.187 -0.288 -0.567 -0.878 -1.973 -2.648** -0.662 0.007 0.033 0.038 0.044 -0.017 0.864 2.747** 3.341** 3.601** -0.230 -0.265 -0.387 -0.176 -0.415 -1.239 -3.321** -3.217** -1.545 -3.361** -1.624 0.009 -0.002 0.016 0.001 -0.035 1.268 -0.211 1.511 0.099 -0.492 0.002 0.041 0.050 0.097 0.283 0.056 0.853 1.101 1.987* 0.920 0.0009 0.001 0.0007 0.0004 -0.001 1.073 0.928 0.566 0.325 -0.159 0.015 0.018 0.017 0.024 0.136 1.560 1.270 1.230 1.617 1.101 -0.056 -0.010 -0.091 0.017 -0.553 -0.699 -0.084 -0.789 0.133 -0.710 0.012 0.034 0.068 0.097 0.091 0.452 0.839 1.764 2.328* 0.346 0.105 0.106 0.166 0.180 0.024 0.070 0.072 0.133 0.149 Std Error (Regression) 0.214 0.323 0.306 0.332 0.502 F/LR-Statistic 3.034 3.079 5.140 5.690 8.150 Probability (F/LR-Statistic) 0.002 0.002 0.000002 0.000 0.519 Durbin-Watson 1.683 1.913 2.131 1.808 SHRET1 SIZE BLOCK SEGMT DIVR1 AGE GEAR1 EQED RISK1 R 2 /McFadden R 2 Adjusted R * 2 Significance at the 5% level ** 228 Significance at the 1% level Appendix 9 OLS and Logit Regressions: Board Independence and Past Performance (Tobin’s Q) Sample Period: 2000-2003 Included Observations: 243 Coefficient t/z-Statistic FIND ACIND NCIND RCIND CMIND Intercept 0.455 0.550 -0.017 0.220 0.776 6.201** 4.957** -0.161 1.923 1.100 0.007 -0.001 0.002 0.020 0.022 1.138 -0.106 0.199 2.103* 0.384 0.008 0.033 0.039 0.048 -0.013 0.985 2.749** 3.380** 3.844** -0.171 -0.251 -0.378 -0.153 -0.358 -1.161 -3.152** -3.139** -1.331 -2.878** -1.521 0.010 -0.001 0.019 0.007 -0.027 1.417 -0.087 1.814 0.582 -0.383 0.005 0.033 0.038 0.095 0.271 0.161 0.701 0.829 1.927 0.881 0.0009 0.001 0.0005 0.0004 -0.001 1.144 0.823 0.381 0.289 -0.183 0.015 0.017 0.013 0.020 0.132 1.557 1.157 0.982 1.381 1.049 -0.073 -0.006 -0.091 -0.027 -0.593 -0.891 -0.045 -0.767 -0.214 -0.750 -0.003 -0.0008 -0.006 -0.011 -0.073 -0.477 -0.101 -0.805 -1.232 -0.840 0.109 0.103 0.152 0.171 0.023 0.074 0.069 0.119 0.139 Std Error (Regression) 0.214 0.323 0.309 0.333 0.503 F/LR-Statistic 3.155 2.985 4.636 5.349 7.856 Probability (F/LR-Statistic) 0.001 0.002 0.00001 0.00001 0.549 Durbin-Watson 1.696 1.915 2.147 1.797 TOBQ1 SIZE BLOCK SEGMT DIVR1 AGE GEAR1 EQED RISK1 R 2 /McFadden R 2 Adjusted R * 2 Significance at the 5% level ** 229 Significance at the 1% level Appendix 10 OLS Regressions: Full Board Independence and Subsequent Performance Sample Period: 2003-2006 Included Observations: 243 Coefficient t-Statistic ROA2 ROE2 SHERT2 TOBQ2 Intercept -0.120 0.680 -0.439 2.688 -0.606 1.274 -4.575** 2.339* -0.272 -0.969 0.178 2.220 -1.587 -2.090* 2.141* 2.225* 0.021 -0.009 0.025 -0.232 1.009 -0.163 2.442* -1.920 -0.049 -1.105 0.202 1.110 -0.227 -1.880 1.912 0.877 0.009 0.130 0.003 -0.055 0.463 2.517* 0.336 -0.496 0.229 0.606 0.121 -0.894 2.827** 2.769** 3.072** -1.897 0.002 -0.0004 0.002 -0.010 0.842 -0.066 1.444 -0.791 -0.014 -0.691 -0.008 0.009 -0.814 -14.398** -0.916 0.084 -0.485 -0.243 -0.333 2.986 -2.261* -0.419 -3.206** 2.395* -0.005 0.039 0.573 -0.071 -0.227 0.656 53.110** -0.546 0.091 0.495 0.929 0.099 0.055 0.475 0.926 0.064 Std Error (Regression) 0.564 1.523 0.273 3.278 F-Statistic 2.579 25.328 338.756 2.846 Probability (F-Statistic) 0.008 0.000 0.000 0.003 Durbin-Watson 2.019 2.047 1.990 1.917 FIND SIZE BLOCK SEGMT DIVR2 AGE GEAR2 EQED RISK2 R2 Adjusted R * 2 Significance at the 5% level ** 230 Significance at the 1% level Appendix 11 OLS Regressions: Audit Committee Independence and Subsequent Performance Sample Period: 2003-2006 Included Observations: 243 Coefficient t-Statistic ROA2 ROE2 SHERT2 TOBQ2 Intercept -0.187 0.574 -0.377 3.477 -0.981 1.119 -4.061** 3.119** -0.118 -0.660 0.046 0.520 -1.029 -2.142* 0.818 0.777 0.023 0.006 0.024 -0.233 1.081 0.101 2.368* -1.878 -0.024 -1.108 0.174 0.739 -0.112 -1.888 1.632 0.579 0.006 0.120 0.005 -0.035 0.326 2.335* 0.510 -0.311 0.231 0.628 0.121 -0.889 2.840** 2.861** 3.044** -1.865 0.002 -0.0005 0.002 -0.009 0.791 -0.087 1.531 -0.678 -0.015 -0.694 -0.008 0.013 -0.851 -14.472** -0.870 0.122 -0.468 -0.190 -0.345 2.839 -2.180* -0.328 -3.296** 2.259* -0.004 0.041 0.572 -0.078 -0.194 0.685 52.636** -0.599 0.085 0.495 0.928 0.082 0.050 0.475 0.925 0.047 Std Error (Regression) 0.565 1.522 0.276 3.308 F-Statistic 2.402 25.376 332.748 2.321 Probability (F-Statistic) 0.013 0.000 0.000 0.016 Durbin-Watson 2.007 2.023 2.005 1.928 ACIND SIZE BLOCK SEGMT DIVR2 AGE GEAR2 EQED RISK2 R2 Adjusted R * 2 Significance at the 5% level ** 231 Significance at the 1% level Appendix 12 OLS Regressions: Nomination Committee Independence and Subsequent Performance Sample Period: 2003-2006 Included Observations: 243 Coefficient t-Statistic ROA2 ROE2 SHERT2 TOBQ2 Intercept -0.255 0.206 -0.351 3.783 -1.423 0.422 -4.006** 3.607** -0.182 -0.252 0.036 1.027 -1.520 -0.773 0.624 1.470 0.026 -0.006 0.025 -0.257 1.232 -0.111 2.351* -2.060* -0.004 -0.890 0.161 0.677 -0.017 -1.530 1.540 0.542 0.010 0.126 0.004 -0.055 0.511 2.416* 0.425 -0.487 0.226 0.594 0.123 -0.869 2.794** 2.629** 3.106** -1.834 0.002 -0.001 0.002 -0.009 0.790 -0.179 1.555 -0.690 -0.013 -0.689 -0.008 -0.0004 -0.715 -14.208** -0.932 -0.004 -0.486 -0.202 -0.342 2.947 -2.267* -0.346 -3.258** 2.348* -0.006 0.041 0.572 -0.070 -0.253 0.679 52.560** -0.539 0.090 0.486 0.928 0.088 0.055 0.467 0.925 0.053 Std Error (Regression) 0.564 1.535 0.276 3.297 F-Statistic 2.554 24.514 332.318 2.510 Probability (F-Statistic) 0.008 0.000 0.000 0.009 Durbin-Watson 2.009 2.008 1.997 1.944 NCIND SIZE BLOCK SEGMT DIVR2 AGE GEAR2 EQED RISK2 R2 Adjusted R * 2 Significance at the 5% level ** 232 Significance at the 1% level Appendix 13 OLS Regressions: Remuneration Committee Independence and Subsequent Performance Sample Period: 2003-2006 Included Observations: 243 Coefficient t-Statistic ROA2 ROE2 SHERT2 TOBQ2 Intercept -0.195 0.372 -0.355 3.491 -1.072 0.753 -3.979** 3.274** -0.188 -0.536 0.010 0.909 -1.739 -1.824 0.182 1.432 0.028 0.010 0.026 -0.260 1.318 0.170 2.431* -2.073* -0.054 -1.068 0.160 0.900 -0.249 -1.816 1.503 0.708 0.007 0.124 0.005 -0.041 0.392 2.409* 0.493 -0.364 0.236 0.623 0.123 -0.913 2.906** 2.831** 3.087** -1.921 0.002 -0.001 0.002 -0.008 0.757 -0.187 1.576 -0.654 -0.015 -0.694 -0.008 0.014 -0.870 -14.425** -0.884 0.137 -0.465 -0.172 -0.346 2.825 -2.174* -0.296 -3.304** 2.255* -0.004 0.044 0.572 -0.082 -0.160 0.739 52.555** -0.628 0.093 0.492 0.928 0.088 0.057 0.473 0.925 0.053 Std Error (Regression) 0.563 1.526 0.276 3.298 F-Statistic 2.640 25.103 331.772 2.496 Probability (F-Statistic) 0.006 0.000 0.000 0.010 Durbin-Watson 2.003 2.041 2.002 1.930 RCIND SIZE BLOCK SEGMT DIVR2 AGE GEAR2 EQED RISK2 R2 Adjusted R * 2 Significance at the 5% level ** 233 Significance at the 1% level Appendix 14 OLS Regressions: Chairman Independence and Subsequent Performance Sample Period: 2003-2006 Included Observations: 243 Coefficient t-Statistic ROA2 ROE2 SHERT2 TOBQ2 Intercept -0.173 0.356 -0.370 3.229 -0.928 0.704 -4.068** 2.973** -0.115 -0.213 0.027 0.780 -1.572 -1.067 0.746 1.824 0.019 -0.017 0.026 -0.214 0.906 -0.297 2.563* -1.771 -0.015 -0.921 0.164 0.778 -0.068 -1.581 1.567 0.622 0.006 0.120 0.005 -0.031 0.301 2.318* 0.518 -0.277 0.235 0.610 0.121 -0.925 2.887** 2.759** 3.050** -1.952 0.002 -0.001 0.002 -0.008 0.721 -0.217 1.587 -0.622 -0.014 -0.690 -0.008 0.003 -0.766 -14.266** -0.918 0.032 -0.484 -0.207 -0.342 2.953 -2.257* -0.355 -3.260** 2.360* -0.004 0.044 0.572 -0.081 -0.167 0.726 52.612** -0.627 0.090 0.485 0.928 0.093 0.055 0.468 0.925 0.058 Std Error (Regression) 0.564 1.533 0.276 3.289 F-Statistic 2.573 24.631 332.574 2.650 Probability (F-Statistic) 0.008 0.000 0.000 0.006 Durbin-Watson 2.015 2.008 2.002 1.926 CMIND SIZE BLOCK SEGMT DIVR2 AGE GEAR2 EQED RISK2 R2 Adjusted R * 2 Significance at the 5% level ** 234 Significance at the 1% level Appendix 15 OLS Regressions: Board Independence and Firm Risk Sample Period: 2003-2006 Included Observations: 243 Coefficient t-Statistic Intercept FIND RISK2 0.766 0.622 0.616 0.621 0.646 4.779** 4.261** 4.197** 4.161** 4.240** -0.307 -2.199* ACIND -0.081 -0.867 NCIND -0.080 - -0.821 RCIND -0.012 -0.139 CMIND -0.042 -0.694 SIZE BLOCK SEGMT DIVR2 AGE GEAR2 EQED SHERT2 -0.046 -0.045 -0.045 -0.047 -0.048 -2.700** -2.618** -2.562* -2.699** -2.828** -0.207 -0.157 -0.136 -0.130 -0.138 -1.161 -0.874 -0.773 -0.726 -0.783 -0.008 -0.011 -0.010 -0.011 -0.011 -0.534 -0.716 -0.606 -0.699 -0.721 -0.240 -0.241 -0.245 -0.245 -0.242 -3.675** -3.648** -3.714** -3.701** -3.663** -0.002 -0.002 -0.003 -0.003 -0.003 -1.315 -1.402 -1.422 -1.449 -1.460 0.015 0.014 0.015 0.014 0.015 1.009 0.962 1.040 0.979 1.010 0.440 0.458 0.451 0.460 0.454 2.501* 2.585* 2.536* 2.592* 2.554* 1.612 1.612 1.611 1.612 1.612 53.110** 52.636** 52.560** 52.555** 52.612** 235 R2 0.930 0.929 0.928 0.928 0.928 0.927 0.926 0.926 0.926 0.926 Std Error (Regression) 0.459 0.463 0.463 0.463 0.463 F-Statistic 342.611 336.286 336.164 335.151 335.868 Probability (F-Statistic) 0.000 0.000 0.000 0.000 0.000 Durbin-Watson 1.976 1.989 1.980 1.986 1.990 Adjusted R * 2 Significance at the 5% level ** 236 Significance at the 1% level