Amsterdam Business School
Board of Directors’ Characteristics and the Effect on Real
Earnings Management.
Name: Jeanine Visser Student number: 10884319
Thesis supervisor: B.G.D. O’Dwyer Date: 15 May 2016
Word count: 10665, 0
MSc Accountancy & Control, specialization: Accountancy Faculty of Economics and Business, University of Amsterdam
Statement of Originality
This document is written by student Jeanine Visser who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
This study investigates the relationship between board characteristics and their impact on real earnings management. The following board characteristics were investigated: board size, age diversity, gender diversity and nationality diversity. A sample of approximately 100 North American and Canadian manufacturing firms were selected for the period of 2009 up to and including 2013. Contrary to theories predicting that smaller boards are more effective, the empirical results reveal that boards with more than 8 members exhibit a lower level of real earnings management for firms in the manufacturing industry. Manufacturing firms are more difficult to monitor due to the complex processes manufacturing firms are facing, indicating that more members are necessary to keep track of these processes. Regarding the diversity of the board composition, both gender diversity and nationality diversity show a negative relationship on real earnings management. Thus boards with more women and boards with more nationalities are better able to reduce real earnings management. The results in this study indicate that heterogeneous boards are more effective than homogeneous boards. Regarding age diversity, this study shows contrary results: boards with more age diversity exhibit a higher level of real earnings management. A higher diversity of age in the board of directors can contribute to a greater age-discrimination climate, which, in turn, leads to a higher level of real earnings management.
Table of contents
1 Introduction ... 6
2 Literature review ... 8
2.1 Agency theory and Corporate Governance ... 8
2.2 The role of the board of directors ... 9
2.3 Real earnings management ... 10
2.4 Prior research of board member characteristics and the impact on real earnings management ... 11
3 Hypotheses development ... 13
3.1 Hypothesis related to board size ... 13
3.2 Hypotheses related to board diversity ... 14
3.2.1 Hypothesis related to age diversity ... 14
3.2.2 Hypothesis related to gender diversity ... 15
3.2.3 Hypothesis related to nationality diversity ... 16
4 Research Design ... 18 4.1 Dependent Variable: ... 19 4.2 Independent Variables ... 20 4.2.1 Board size ... 20 4.2.2 Age diversity ... 20 4.2.3 Gender diversity ... 21 4.2.4 Nationality diversity ... 22 4.2.5 Control variables ... 22 4.3 Sample selection ... 22 5 Empirical Analysis ... 23 5.1 Descriptive Statistics ... 23 5.2 Empirical results ... 25
5.3 Sensitivity analysis ... 28 6 Conclusion ... 30 7 References ... 32
1 Introduction
In this paper the effectiveness of corporate governance on financial reporting quality is examined. According to Lin et al. (2010) financial reporting is of high quality when the information in financial reports will provide timely, relevant, and transparent information that can help minimize uncertainty. Conversely, they argue that low-quality financial reporting is associated with ambiguous, misleading, or unreliable information, which is likely to increase information asymmetry.
A way to generate misleading and unreliable information is to use earnings management. Hadini et al. (2011) argue that earnings management serves as a signalling mechanism and can either be treated or not treated as an opportunistic form of behaviour. Managers often practice two forms of activities to manage earnings: accrual-based activities and real activities (Heshemi & Rabiee, 2011). Managers are inclined to make choices between the two earnings management policies. A large body of academic research examines the causes and consequences of earnings management through accrual-based activities. Research of the causes and consequences of real earnings management activities are sparse. Graham et al. (2005) acknowledge that the aftermath of accounting scandals and the certification requirements imposed by the Sarbanes-Oxley Act have changed managers’ preferences for the mix between taking accounting versus real actions to manage earnings. He found that firms switched form accrual-based to real earnings management methods after the passing of SOX. It is therefore important to expand on existing literature by examining if financial reports are of high quality with regard to the extent real earnings management is used by listed firms.
This study focuses on the composition of the board of directors and the different board member characteristics since, according to Adams et al.(2010), these have been treated as the most salient aspects of corporate governance mechanisms in mitigating earnings management. This study tries to give answers to the following research question: how do different board member characteristics affect real earnings management?
Since board member quality plays an important role in depressing real earnings management (Lui and Tsai, 2015), it is important to investigate which board member characteristics have a negative effect on real earnings management. Some characteristics of the board of directors have been examined in previous studies. These studies were mainly concerned with the following characteristics: independence, ownership, professionalism, education, busyness, meeting attendance and pledges. Outcomes of these studies are discussed in paragraph 2.4.
This study expands on previous studies by examining characteristics such as board diversity (in age, gender and nationality) and board size and their effect on real earnings management. This is the first study which examines the impact of these characteristics on real earnings management. The outcomes of this study provide some insights into the importance of the different board characteristics and their quality in mitigating opportunistic real earnings management behaviour, thus generating better financial reporting quality for the benefit of investors and other stakeholders.
A sample of approximately 100 North American and Canadian manufacturing firms are selected for the period of 2009 up to and including 2013. A sample of 525 firm-year observations remain to examine the hypotheses related to board size, age diversity, gender diversity and nationality diversity.
Contrary to theories predicting that smaller boards are more effective, the empirical results in this study reveal that boards with a higher number of members exhibit a lower level of real earnings management for firms in the manufacturing industry. Following the results of this study it appears that boards with more than 8 directors are likely to be more effective for firms in the manufacturing industry. Manufacturing firms are more difficult to monitor due to the complex processes manufacturing firms are facing, indicating that more members are necessary to keep track of these complex processes. Regarding the diversity of the board composition, both gender diversity and nationality diversity show a negative relationship on real earnings management. Thus boards with more women and boards with more nationalities are better able to reduce real earnings management. Regarding age diversity, this study shows contrary results: boards with more age diversity exhibit a higher level of real earnings management. A higher diversity of age in the board of directors can contribute to a greater age-discrimination climate, which, in turn, leads to a higher level of real earnings management.
The study is divided into six sections, including this introduction. The second section covers the theoretical framework. The third section presents the development of the hypotheses. In Section 4 the research design and the sample selection are described. In section 5 the empirical results and additional analyses are given. Finally, the sixth section provides the conclusion.
2 Literature review
This section starts with explaining agency theory and the relation with corporate governance. In paragraph 2.2 the role of the board of directors will be discussed. In paragraph 2.3 the main reasons will be given why it is interesting to use real earnings management as proxy for financial reporting quality. Finally outcomes of prior research of different board member characteristics and the impact on real earnings management will be given.
2.1 Agency theory and Corporate Governance
According to Linder and Foss (2015) agency theory studies the problems and solutions dealing with delegation of tasks from principals to agents in the context of conflicting interests between the parties. Linder and Foss (2015) provide three examples of conflicting interests: ‘conflict in outcome-type preferences’ (1), ‘conflict in risk preferences’(2) and ‘conflict in time horizon’(3). Linder and Foss (2015) provide for each conflict the following examples:
- Top managers acting as agents of the firms’ shareholders may prefer ‘empire building,’ perks, leisure time, and so on, instead of maximizing shareholders’ returns (‘conflict in outcome-type preferences’);
- Managers may prefer engaging in capital expenditures that maximize the survival chances of the firm that pays their salaries, while shareholders likely prefer them to go for high returns, with risky investments as they can diversify risk (‘conflict in risk preferences’); - It is also possible that managers may not plan to stay on for a long time with the firm in
question and thus differ in the time horizon they consider in decision-making from (assumingly long-term oriented) shareholders (‘conflict in time horizon’).
According to Linder and Foss (2015) there are many other forms of conflict of interests between the principal and the agent which all imply that the agent may not act in the best interest of the principal. They argue that these conflicting interests only become problematic when information is asymmetrically distributed between the principal and the agent. The interests of various groups (e.g. shareholders, stakeholders) should be protected or balanced. This protection and balancing of the conflicting interests between these groups is part of the system of corporate governance.
According to McDonnell (2015) ‘corporate governance’ refers to the legal rules and institutional framework that structure decision-making within business corporations. He argues that corporate governance considers the means of various legal and non-legal mechanisms,
which help structure how decisions within corporations are made within the context of (the before mentioned) conflicting interests. McDonnel (2015) mentioned the following major means used to advance the interests of shareholders and/or other stakeholders: the board of directors, shareholder voting, monitoring by various gatekeepers, fiduciary duty rules, and incentives created by various markets. This study focusses on the role of the board of directors which will be discussed in chapter 2.2.
2.2 The role of the board of directors
According to McDonnell (2015) the board of directors is the core locus of authority within corporations as a matter of law. He argues that some justify this as the best way of protecting the interests of shareholders, who elect the board (Bainbridge, 2003), while others justify it as a way of balancing the interests of various constituencies (Blair and Stout, 1999). According to Dorado and Molz (2005) the responsibility of the board of directors is to clearly identify the core mission of the organization. They believe that the board of directors should help the organization to grow while remaining loyal to its mission. Therefore directors should participate in discussions which prepare or validate decisions, thus bringing their suggestions, opinions, and expertise.
Dorado and Molz (2005) emphasize the traditional role of boards in public corporations: to provide strategic guidance and advice. In recent decades, directors are responsible for monitoring, evaluating, and disciplining the management of a company on the one hand and oversight of financial reporting on the other hand. Oversight of financial reporting has become one of the most important responsibilities of the board (McDonnell; 2015). Contemporary boards of directors are charged with the task of monitoring the performance and activities of top management to ensure that the top managers act in the best interest of the owners (Jensen and Meckling, 1976). From this perspective boards have an important role to play in mitigating agency problems.
Bushman and Smith (2001) show that accounting can be used as a control mechanism through which shareholders and other stockholders can monitor managers actions. Accounting can contribute to a company’s corporate governance mechanisms by providing useful information to decision makers and thereby reducing information asymmetry and the impact of agency conflicts (Holtz and Neto; 2014). Beekes, Pope and Young (2004) believe that financial reports represent a particularly clear example of an activity in which the interest of managers and stockholders may not be perfectly aligned.
According to Lin et al. (2010) financial reporting is of high quality when the information in financial reports will provide timely, relevant, and transparent information that can help
minimize uncertainty. Conversely, they argue that low-quality financial reporting is associated with ambiguous, misleading, or unreliable information, which is likely to increase information asymmetry. According to Adams et al. (2010) the composition of the board of directors is treated as one of the most salient aspects of corporate governance mechanisms in mitigating real or accrual-based earnings management.
2.3 Real earnings management
Earnings management can be used by top managers to generate misleading and unreliable information. Hadini et al. (2011) argue that earnings management serves as a signalling mechanism and can either be treated or not treated as an opportunistic form of behaviour. Managers often practice two forms of activities to manage earnings: accrual-based activities and real activities (Hashemi & Rabiee, 2011).
High-profile accounting scandals in the USA such as Enron and WorldCom in 2001 and 2002 show how earnings management and corporate governance failures adversely affect firms and entire economies (Cho & Chun, 2015). Managers are inclined to make choices between the two earnings management policies in the post-Enron and post Sarbanes-Oxley Act era (Badertscher, 2011). A large body of academic research examines the causes and consequences of earnings management through accrual-based activities. A growing literature also examines how managers use real earnings management to achieve earnings targets. To examine whether financial reports are of high quality this study will focus on the extent real earnings management is used by listed firms in the manufacturing industry.
This study focusses on real earnings management for two reasons. First, prior studies show that accrual-based earnings management increased steadily from 1987 until the passing of the Sarbanes-Oxley Act (SOX) in 2002, followed by a significant decline after the passing of SOX. Conversely, the level of real earnings management activities declined prior to SOX and increased significantly after the passing of SOX, suggesting that firms switched form accrual-based to real earnings management methods after the passing of SOX (Graham, Harvey, and Rajgopal 2005; Cohen, Dey and Lys, 2008). Graham et al. (2005) acknowledge that the aftermath of accounting scandals and the certification requirements imposed by the Sarbanes-Oxley Act may have changed managers’ preferences for the mix between taking accounting versus real actions to manage earnings. They also argue that managers are more willing to make use of real decisions rather than accounting decisions. They interviewed a CFO which offered an interesting insight into the choice between real and accounting-based earnings management. The CFO said that auditors can second-guess the firm’s accounting policies but they cannot readily challenge
real economic actions to meet earnings targets that are taken in the ordinary course of business. Another executive emphasized that firms now go out of their way to assure stakeholders that there is no accounting based earnings management in their books. He stated that there seems to be a corporate fear that even an appropriate accounting choice runs the risk of an overzealous regulator concluding ex post that accounting treatment was driven by an attempt to manage earnings.
Second, according to Cho and Chun (2015) real earnings management can be more harmful to a firm than accrual-based earnings management because real earnings management can adversely affect firm value by distorting normal operating activities and thereby harming the relationship with key stakeholders such as customers, employees, and communities. Real earnings management accompanies cash expenditure to manage earnings because the firm manipulates normal operating, investing, and financing activities during the fiscal year, and thereby can damage firm value (Gunny, 2005; Kim and Sohn, 2013 as cited in Cho and Chun, 2015). Accrual-based earnings management does not accompany cash transactions, uses just timing differences in recognizing accrual revenues and expenses, and is likely to damage firm value in a minimal level (Subramanyam, 1996 as cited in Cho and Chun, 2015).
2.4 Prior research of board member characteristics and the impact on real
earnings management
Lui and Tsai (2015) state that to mitigate the agency problems resulting from earnings management, empirical research has identified some of the corporate governance mechanisms and has shown that earnings management through real activities or accounting accruals is constrained by these mechanisms such as board characteristics or ownership structure. Kang and Kim (2012) found that managers’ discretionary activities, such as sales manipulation, overproducing, and cutting expenses, can be controlled effectively, if corporate governance plays a more prominent role in operational and investment decisions. Therefore it is important to identify which mechanisms of corporate governance are most influential in mitigating the problems which may result from real earnings management. They found that managers are less likely to be engaged in real activity-based earnings management when the board of directors consists of more external directors because in that case the board operates more independently. Visvanathan (2008) finds that most overall board characteristics which are significant in limiting accrual-based earnings management are not significant in limiting real earnings management except for the proportion of independent directors.
Lui and Tsai (2015) incorporated seven different factors of board member characteristics and created a “board member quality index”. They used this index as a proxy measure to test the impact of board member characteristics on real earnings management. They incorporated the following characteristics: independence, ownership, professionalism, education, busyness, meeting attendance and pledges. They found that firms with a better board member quality (according to their index) exhibit a lower level of real earnings management. Aside from having independent directors it is important that board members have an accounting/financial expertise to decrease the probability of real earnings management (Chen, Elder, and Hsieh, 2007; Lui and Tsai, 2015). A higher educational degree also increases the probability to decrease real earnings management (Lui and Tsai, 2015). Xie et al. (2003) conclude that a board that meets more often is able to devote more time to constraining accrual earnings management. Jiraporn et al. (2009) report that busy directors exhibit a higher tendency to be absent from board meetings. Lui and Tsai (2015) found that boards which have more meetings, a higher meeting attendance, and less busy directors have lower real earnings management. They also found evidence that firms with higher institutional ownership can provide monitoring protection from opportunistic real earnings management behaviour.
Since board member quality plays an important role in depressing real earnings management (Lui and Tsai, 2015), it is important to investigate which board member characteristics have a negative effect on real earnings management. This study expands on previous studies by examining other board characteristics such as board diversity (in age, gender and nationality) and board size and their effect on real earnings management.
3 Hypotheses development
In this paragraph hypotheses related to board size and board diversity will be developed. The first hypothesis deals with board size and the next three hypotheses will deal with board diversity. These hypotheses will be divided into the topics: age, gender and nationality. Each of these topics will be discussed separately.
3.1 Hypothesis related to board size
An important subject matter in the corporate governance literature is whether having more outside directors increases corporate performance. Prior literature documents a negative association between board size and corporate performance (Eisenberg, Sundgren, and Wells, 1998; Yermack, 1996; Xie, 2013). According to Cheng (2006) this association is consistent with the view that both agency problems and coordination/communication problems become more severe as a board grows larger.
Cheng (2006) argues that agency problems arise from dysfunctional norms of behaviour in boardrooms. According to Lipton and Lorsch (1992) it becomes difficult for board members to express their opinions and ideas in the limited time of board meetings when a board has more than ten members. Similarly, Jensen (1993) contends that when a board has more than seven or eight directors, the directors are less likely to function effectively and are easier for the CEO to control. Ahmed and Duellman (2007) believe that larger boards face the problem of “free riding” in the sense that the members of the board depend on each other to monitor management.
According to Cheng (2006) the coordination/communication problems are more straightforward. He believes that as a board becomes larger, it is more difficult for the firm to arrange board meetings and for the board to reach a consensus. As a result, larger boards are less efficient and slower in decision-making (Cheng, 2006). Jensen (1993) states that when a board increases in size, the costs of the agency problems and the coordination/communication problems become bigger than the potential advantages of having more directors to draw on, leading to a lower level of corporate performance. Based on this information the following hypothesis is developed related to board size:
Hypothesis 1:
Boards with a lower number of members have less real earnings management activities than boards with a higher number of members.
3.2 Hypotheses related to board diversity
According to Kang et al. (2007), board diversity is defined as a variety in the composition of the board of directors and may be categorized into directly observable factors (e.g. nationality, age, gender and ethnic background) and less visible features (educational, functional and occupational background). This study focusses on the observable factors: age, gender and nationality.
Diversity has been advocated as a means of improving organizational value and performance by providing the board with new insights and perspectives. Prior research suggests that board diversity leads to a greater knowledge base, creativity and innovation and thereby provides a competitive advantage to the organization (e.g. Carter et al., 2003; Erhardt et al., 2003). Heterogeneous groups are more likely to disagree, thereby weakening the team consensus (Hambrick et al., 1996). There is no hard evidence that board diversity improves or deteriorates the corporate governance performance. Hutchinson and Plastow (2015) argue that diversity both enhances performance by increasing decision-making capacity, but detracts from group performance by increasing conflict. In the following paragraphs will be discussed how the factors age, gender and nationality are related to corporate governance performance.
3.2.1 Hypothesis related to age diversity
Prior studies on the relationship between age diversity and corporate performance have inconsistent outcomes. Mahadeo et al. (2012) found a positive relation between age diversity and corporate performance. Other studies found either no significant relation (Bunderson and Sutcliffe 2002; Zimmerman 2008) or even a negative relation (Milliken and Martins 1996). According to Ferrero et al. (2015) a possible explanation of these inconclusive findings could be that previous research has omitted the interactions among different aspects of diversity. According to him a distinction of diversity as variety, separation and disparity needs to be applied. This study focusses on diversity as a variety where age is categorized to the different generations (described in paragraph 4).
Ferrero et al. (2015) conclude that age diversity, defined as generational diversity, positively impacts corporate performance. According to Kang et al. (2007) generational diversity means that teams whose members draw from different generations in a balanced way can translate greater information richness within a unit; the older group provides experience and wisdom, the middle group carries the major positions of active responsibilities in corporations and in society, and the younger group has the energy and plans for the future. According to
Ferrero et al. (2015) groups with greater information richness make more effective decisions. Based on these arguments, the following hypothesis is expected:
Hypothesis 2a
Boards with less age diversity members have more real earnings management activities than boards with more age diversity.
3.2.2 Hypothesis related to gender diversity
According to Khazanchi (1995) women respond differently to unethical actions in certain types of ethical dilemmas pertaining to disclosure, integrity and conflict of interest than men. Khazanchi states that women are more ethical in the workplace and are less likely to engage in unethical behaviour to gain financial rewards. Not only do women demonstrate greater risk aversion and ethical behaviour, they are also better at obtaining voluntary information which may reduce information asymmetry between female directors and managers (Gul, Fung, & Jaggi, 2009). According to Byrnes et al. (1999) women are more careful and less aggressive than men in a variety of decision-making contexts. Also women are less likely to take risks particularly in the financial decision environment (Powell and Ansic, 1997). Gul et al. (2009) argue there is a greater likelihood that women use a restrained approach to earnings management.
Prior research of the quality of earnings management and the association with women on boards show mixed results. Krishnan and Parsons (2008) found that the quality of earnings management is higher for firms with more female directors, and argued that women are likely to be more ethical in their judgement and behaviour than men. Conversely, Sun, Liu, and Lan (2011) found no evidence for the impact of female representation on audit committees and earnings management, while Thiruvadi and Huang (2011) found that the presence of female directors on the audit committee is negatively related to earnings management.
A possible explanation of these inconclusive findings can be found in the study of Kyaw et al. (2015). They found empirical evidence that in countries where women have similar empowerment as men in the workplace, female representation on the corporate boards mitigates earnings management through aggregate accruals. They argue that this result highlights the importance of female empowerment in the workplace. Based on these arguments, the following hypothesis is expected:
Hypothesis 2b
Boards with gender diversity have less real earnings management activities than boards with no gender diversity.
3.2.3 Hypothesis related to nationality diversity
According to Webster the general description of the term diversity in the context of organizations is: “The condition of being diverse: the inclusion of diverse people (as people of different race or culture) in a group or organization” (Merriam-Webster Online Dictionary, 2016). Diversity in nationality can be seen as an overall measure of diversity that captures differences in deep-rooted social, cultural, and institutional characteristics that were shaped many years ago (Delis, Gaganis, Hasan, & Pasiouras, 2015).
The national origin of individuals reflects the institutional environment of the country in which they spend the majority of their formative years (Hambrick et al., 1998). A combination of formal and informal institutions in a country guide individuals and organizations in dealing with uncertainty, deciphering the environment, and taking appropriate actions (Crossland and Hambrick, 2007). Formal institutions are explicit and codified, and consist of the political and economic rules and contracts that govern property rights and transactions in a society, whereas informal institutions consist of tacit norms, conventions, and values that shape social interaction (North, 1990). Informal institutions, or national culture, is defined as a system of collectively held beliefs and values (Hofstede, 1980 as cited in Nielsen and Nielsen, 2013). Cultural patterns of thinking, feeling, and acting are acquired in early childhood because at that time a person is most susceptible to learning and assimilation (Hofstede and Hofstede, 2005 as cited in Nielsen and Nielsen, 2013). According to them these patterns are deeply rooted and once established within a person’s mind, they are unlikely to substantially change through subsequent experiences. For this reason nationality can be seen as a superordinate construct that encompasses the influences of both formal and informal institutionally embedded experiences on executive orientation and decision making (Nielsen and Nielsen, 2013).
Consistent with the information-processing/decision-making perspective, nationally diverse teams bring a wide range of knowledge of, and experiences with different institutional environments. As board members will reconcile their diverse institutionally embedded experiences, they engage in in-depth discussions, consideration of various alternatives, and generation of new creative ideas (Hambrick et al., 1998). As a result, nationally diverse boards are better at solving complex tasks and arriving at more innovative solutions (Nielsen and Nielsen, 2013).
On the other hand, social categorization theory (Turner, 1987) suggests that diversity may also come at costs for team dynamics. As nationality, and particularly informal institutions (culture), determines communication patterns and interaction styles, boards with members with a different nationality may experience affective conflict, lower cohesiveness, and slower decision making (Earley and Mosakowski, 2000; Hambrick et al., 1998).
Blau (1977) suggests that a high degree of group heterogeneity can effectively weaken social barriers because of diffusion of diversity within the groups (Delis, Gaganis, Hasan, & Pasiouras, 2015). Swann et al. (2004) argue that combining the different ideas, knowledge, and skills of different cultures greatly enhances the potential for creative synthesis. These theories imply that organizational groups composed of members of several different nationalities will benefit from prosperous interactions, heightened cooperation, and improved outcomes (Delis, Gaganis, Hasan, & Pasiouras, 2015). Based on these arguments, the following hypothesis is expected:
Hypothesis 2c
Boards with nationality diversity have less real earnings management activities than boards with no nationality diversity.
4 Research Design
To investigate the hypotheses developed in this study, the following regression model will be used to test the relationship between board characteristics (size and diversity) and real earnings management. The specification of the variables are shown in Table 1.
REMproxy it = a0 + a1BSIZEit +a3AGEDIVit + a4GENDIVit + a5NATDIVit + a6LEVit + a7FSIZEit + a8DUALit + a9ROAit + a10BIG4it + a11MBit + δDYEARit + εit
Table 1. Variable Definitions
Variable Definition
REMproxy Real earnings management proxy, which equals the sum of the standardized measure of abnormal cash flows, abnormal inventory over-production, and abnormal discretionary expenses.
BSIZE Board size is measured as a dummy variable that takes a value of one if the board consists of more than eight members, and zero if the board consist of less than eight members.
AGEDIV Age diversity is calculated with Blau’s Index.
GENDIV Gender diversity is calculated as the percentage of women on the board of directors.
NATDIV Nationality diversity is calculated with Blau’s Index.
LEV Firm leverage is measured as total debt divided by total assets. FSIZE Firm size is defined as the natural logarithm of the firm’s total assets.
DUAL CEO duality is measured as a dummy variable that takes a value of one if the chairman and CEO positions are held by the same person, and zero otherwise.
ROA The ratio of the return on assets is measured as the sum of profit after tax plus interest expenses divided by total assets.
BIG4 Big 4 auditor is measured as a dummy variable that takes a value of one if the firm’s auditor is among one of the big 4, and zero otherwise.
MB The market-to-book value is measured as the market value of equity divided by the book value of equity.
DYEAR Firm year is measured as year dummy variables to account for the unobserved variation.
4.1 Dependent Variable:
Firms that manage earnings are likely to use multiple activities and to capture overall effects of abnormal real earnings activities, therefore this study constructs a real earnings management index (REMproxy) as the dependent variable in the regression model. The model developed by Dechow et al. (1998) and implemented by Zang (2005), Roychowdhurry (2006), Chi et al. (2001), and Kang and Kim (2012) will be used. Dechow et al. (1998) developed empirical models to estimate the normal levels of real business activities, as reflected in cash flow from operations, production costs, and discretionary expenditures. They use the residuals from these models as proxies for real earnings management. The residuals are the abnormal level of each measure calculated as the actual level of a variable minus its normal level.
Model 1 will be used to estimate the normal level of cash flow from operations: where t is a year indicatior, CFO is cash flow from operations, A is total assets, S is net sales, and ∆S is the change in sales compared with the sales amount in the previous year.
Model 1
OCF it/Ait-1 = a0 (1/Ait-1) + a1(Sit/Ait-1) + a2(∆Sit/Ait-1) + εit
Model 2 will be used to estimate the normal level of production costs: where PROD is the sum of the costs of goods sold (COGS) and the change in inventory.
Model 2
PRODit/Ait-1 = a0(1/Ait-1) + a1(Sit/Ait-1) + a2(∆Sit/Ait-1) + a3(∆Sit-1/Ait-1) + εit
Model 3 will be used to estimate the normal level of discretionary expenses: where DISEXP is discretionary expenses as measured by the sum of advertising, R&D and selling, general and administrative expenses.
Model 3
DISEXPit/Ait-1 = a0(1/Ait-1) + a1(Sit-1/Ait-1) + εit
The above models will be estimated by running cross-sectional regressions for every year. The regression residuals capture the abnormal level of real earnings management activities. The residuals from these models will be used as proxies for real earnings management. An overall
proxy of real earnings management activities (REMproxy) will be constructed by combining the three models. According to Kang and Kim (2012) it is important to combine each individual measure to capture overall effects of abnormal real activities because firms that manage earnings upwards are likely to use multiple activities.
The abnormal cash flows from operations and abnormal discretionary expences will be muliplied by -1 to make it is easier to interpret the results. According to Cohen and Zarowin (2010) higher values indicate a higher probalility of making real decisions to increase earnings. The abnormal level of production costs will not be multiplied by -1 since higher production costs are indicative of overproduction to reduce the costs of goods sold.
4.2 Independent Variables
In this section will be discussed how the independent variables: board size, age diversity, gender diversity, and nationality diversity will be constructed. Besides the independent variables this study uses a number of 7 control variables. These control variables will also be discussed in this section.
4.2.1 Board size
To measure board size the number of board members will be used. Boards with more than 8 members are considered to be boards with a higher number of members. Boards with a number of members lower than 8 are considered to be boards with a lower number. The variable board size (BSIZE) is captured by a dummy variable and coded as 1 if the boards consist of more than 8 members and 0 if the boards consist of less than 8 members.
4.2.2 Age diversity
The way Ferrero et al. (2015) categorized different age stages will be used to measure age diversity. According to Ferrero et al. (2015) focussing on age stages as a variable represents differences in personality, traits, skills, attitudes, mental health, work values, and behaviours. These differences will be categorized according to the generations, since Ferrero et al. (2015) believes that the social and historical experiences and circumstances from a respective generation have influenced the individuals’ behaviours. Ferrero et al. (2015) use four major generations of the 20th century:
- the Greatest Generation (1922–1945); - Boomers (1946–1964);
- Generation Y (1984–2002).
Suvillan et al. (2009) and Twenge et al. (2010) argue that members of the Greatest Generation are self-disciplined, believe in self-sacrifice and traditional values. They are extremely loyal employees. Boomers think that hard work and effort will lead to success, they value extrinsic measures of career success, develop a distrust of authority, and place a high value on independent thinking. Xers are influenced by the financial, family, and societal insecurities that dominated their childhoods. They lack solid traditions but are highly mobile and are accustomed to rapid change. They learn quickly, embrace diversity, and like informality. According to Twenge et al. (2010) the characteristics of the youngest generation are less clear, they remark that this generation has grown up with the Internet, and they are accustomed to getting access to information quickly.
This study uses Blau’s Index (1977) that is calculated by model 4. This equation is used to calculate the variable age diversity (AGEDIV) as variety: where p is the percentage of members in the group of i (generation)
Model 4
Age diversity (AGEDIV) = [1 −∑ (pi)²]
This index has been divided by its theoretical maximum with the aim of standardizing the results and making the interpretation of the index easier. The minimum theoretical variety occurs when all members belong to the same category. Harrison and Klein (2007) highlight that the maximum theoretical variety is when each member within a unit comes from a unique category. However, this maximum implies that all boards have the same size and there are as many categories as directors. Given that the data do not fulfil both conditions, the maximum empirical variety is maximized when the four categories are present on a board in equal proportions.
4.2.3 Gender diversity
To measure gender diversity the relative ratio of females on the board of directors will be used. This ratio will be calculated using model 5: where x is the number of women on the board divided by its total members of the board.
Model 5
4.2.4 Nationality diversity
Nationality diversity is also calculated with Blau’s Index (Blau 1977) using the same equation as how age diversity is calculated: where p is the percentage of members in the group of i (i.e., nationality).
Model 6
Nationality diversity (NATDIV) = [1 −∑ (pi)2]
4.2.5 Control variables
A number of firm-specific control variables included in the models are based on the existing literature (Kang and Kim, 2012; Roychowdhury, 2006; Chi et al., 2011; Lui & Tsai, 2015). This study uses the following control variables: firm leverage, firm size, CEO duality, the ratio of the return on assets, Big4 auditor, the market-to-book value, and a firm year dummy. The specification of the variables are shown in Table 1.
4.3 Sample selection
The sample selection process started with 11,517 firms listed globally for the period of 2009 up to and including 2013. The data were taken from Compustat North American, a database of U.S. and Canadian fundamental and market information on active and inactive publicly held companies. Manufacturing firms were selected based on the North American Industry Classification Code (NAICC) because in this study production costs play an important role in measuring real earnings management activities. Only manufacturing firms have these costs. In order to capture overall effects of real earnings management activities, the abnormal levels of cash flow from operations, production costs, and discretionary expenditures are used as proxy to measure real earnings management.
After removing incomplete data, a total of 1,742 firms remain. Of these firms 107 firms were randomly selected. The data concerning the board of directors (board size, age, gender and nationality of board members) were not available in Compustat North American. This information was gathered from the financial statements of the companies. 7 outlier observations with an extreme high value for the independent variable REMproxy were removed. After removing the outliers a sample of 525 firm-year observations remain to examine the hypotheses of this study. Although for some of these observations the information of the members’ nationality was not available, the other characteristics of the board were used to examine the hypotheses related to board size, age and gender.
Table 2
Descriptive Statistics
Variable N Mean Median Std. Deviation Minimum Maximum
REM PROXY 524 1.231 0.985 1.240 -1.408 9.830 BSIZE 524 0.359 0.000 0.480 0.000 1.000 AGEDIV 524 0.407 0.444 0.170 0.000 0.889 GENDIV 524 0.091 0.000 0.111 0.000 0.636 NATDIV 470 0.243 0.197 0.262 0.000 0.972 LEV 524 0.656 0.533 1.442 0.034 22.531 FSIZE 524 15,400 497 49,728 0.382 360,325 DUAL 523 0.554 1.000 0.497 0.000 1.000 ROA 524 -0.281 0.043 1.432 -25137 0.736 BIG4 515 0.701 1.000 0.458 0.000 1.000 MB 512 -3.984 1.904 182.272 -4027.244 470.114
Valid N (listwise) is 460. See Tabel 1 for variable definitions.
5 Empirical Analysis
5.1 Descriptive Statistics
The descriptive statistics of the sample used in this study are shown in table 2. Table 2 shows that on average the real earnings management proxy (REMproxy) is 1.231. The positive mean of REMproxy implicates that most real earnings management activities go through the abnormal level of production costs. The mean of the board size (BSIZE) is 0.359. The means of the age diversity (AGEDIV), gender diversity (GENDIV), and nationality diversity (NATDIV) are 0.407, 0.091, and 0.243, respectively. The total debt to total assets level (LEV) is on average 65.6%. The mean of the firm size (FSIZE) is 15,400 (in millions). On average, 55.4% of the sample have a CEO duality structure (DUAL). Both the return on assets (ROA) and the market-to-book ratio (MB) show a negative mean of -28.1% and -398.4%, respectively. On average, 70.1% of the sample are audited by a big 4 auditor (BIG4).
Table 3 provides the Pearsons correlations of variables. The real earnings management proxy is significantly negatively related to gender diversity and nationality diversity, implying that when the firms’ board of directors consists of men as well as women, and consists of people with different nationalities, they exhibit a lower level of real earnings management. Age diversity, in contrast with gender diversity and nationality diversity, is significantly positively related to the real earnings management proxy. Boards with more age diversity exhibit a higher level of real earnings management. Since a different outcome was expected, a closer look was taken on
Table 3
Pearson correlations
REMproxy BSIZE AGEDIV GENDIV NATDIV LEV FSIZE DUAL ROA BIG4 MB
REMproxy 1 BSIZE - 0,075 1 AGEDIV 0,144** 0,034 1 GENDIV - 0,125** 0,276** - 0,165** 1 NATDIV - 0,098* 0,003 0,04 - 0,051 1 LEV 0,205** - 0,028 0,006 - 0,028 0,090* 1 FSIZE 0,06 0,373** - 0,090* 0,114** 0,279** - 0,009 1 DUAL - 0,054 0,059 - 0,05 0,190** - 0,164** - 0,057 - 0,07 1 ROA 0,071 0,138** 0,057 0,049 - 0,069 - 0,507** 0,077 0,049 1 BIG4 - 0,086 0,361** - 0,027 0,219** 0,127** - 0,093* 0,199** - 0,007 0,191** 1 MB - 0,063 0,032 - 0,002 0,042 0,048 - 0,014 0,007 0,052 - 0,018 - 0,025 1
The symbols ** and * denote statistical significance at the 5% and 10% levels, respectively. See Tabel 1 for variable definitions.
proportions of the different age groups. Each age group variable has been included in the Pearsons correlations. The table shows that only the age group of the ‘Boomers’ have a significant influence on the decrease of real earnings management activities. This implies that the generation ‘Boomers’ is better able to lower the level of real earnings management activities than the other generations.
The size of the board is negatively related to real earnings management, which suggests that when the board of directors has more members, there is a lower level of real earnings management activities. This phenomenon is against the predetermined expectation that boards with a lower number of members exhibit a negative relation with real earnings management.
Of the control variables only the leverage, measured as total debt divided by total assets, is significantly positively related. Table 3 shows that besides the leverage, the firm size and the return on assets are also positively related, implying that firms with higher leverage and higher assets are more likely to engage in real earnings management activities. It is obvious that when firms show a higher return on assets, they exhibit a higher level of real earnings management activities since firms are inclined to use real earnings management to manage earnings upwards. According to table 3 real earnings management activities decline with CEO duality, when firms are audited by a big 4 auditor and when firms show a higher market-to-book ratio.
Since this study includes several dependent variables the problem of potential multicollinearity is present. To verify if there is not a strong correlation between two or more dependent variables the variable inflation factors (VIF) of the variables in the regression model has been tested. The variable inflation factors (VIF) amount to 1.503 and suggest that a severe multicollinearity problem does not exist.
5.2 Empirical results
Board size
The results of the regression analysis of the different board characteristics on real earnings management are provided in Table 3. The board size (BSIZE) coefficients in Columns 1 and 5 of Table 3 are all negative and significant at the 5 percent level. This significantly negative result indicates that firms with a higher number of board members, i.e. more than 8 members, exhibit a lower level of real earnings management. This finding does not support hypothesis 1. Contrary to theories predicting that smaller boards are more effective and thus have a lower degree of real earnings management, increasing the number of directors in manufacturing firms does not increase the volume of real earnings management activities. The evidence shows a positive relationship between board size and real earnings management.
‘The Bridgespan Group’ (2016), a non-profit advisor, states that the optimal board size may vary according to the moment in the board’s life cycle, its mission, and whether it is a national or a local board. According to them a larger size provides enough people to manage the workload of the board. They believe that more board members represent more perspectives. Also Ticker (2015) and Pozen (2010) found that the ideal number of board members depends on the circumstances in each case. They state that not only smaller boards are important in being effective, but having professional directors with sufficient amounts of time, expertise and experience are important as well. Since this study did not control for these variables, it might be that in the used sample larger boards consist of more professional directors, which could cause the lower level of real earnings management.
This positive relationship could also be caused by the depth and complexity of the issues which manufacturing corporations are facing. This implies that having more than 8 directors on the board of manufacturing firms is necessary to increase the effectiveness of the board in decreasing real earnings management activities. Since this study focusses only on manufacturing firms, further study is necessary to investigate if in different industries the size of the board has a similar effect on real earnings management.
Age diversity
The coefficients of age diversity (AGEDIV) in Columns 2 and 5 of Table 3 are all positive and significant at a 1 percent level. This significantly positive result indicates that boards with more age diversity exhibit a higher level of real earnings management. This result is in contrast to what
Table 4.
Regression analyses of board characteristics on real earnings managment (N=525). Panel A: Variable statistics
Predicted Column 1 Column 2 Column 3 Column 4 Column 5
Sign Coefficient Coefficient Coefficient Coefficient Coefficient
Intercept +/- 0.664 0.743 0.644 1.064 0.877 (0.555) (0.625) (0.540) (0.858) (0.722) BSIZE + -0.098 -0.117 (-2.012)** (-2.235)** AGEDIV - 0.132 0.137 (3.100)*** (3.028)*** GENDIV - -0.117 -0.098 (-2.659)*** (-2.089)** NATDIV - -0.127 -0.161 (-2.681)*** (-3.425)*** LEV + 0.317 0.305 0.312 0.308 0.305 (6.398)*** (6.190)*** (6.325)*** (5.894)*** (5.959)*** FSIZE + 0.095 0.078 0.075 0.087 0.155 (2.086)** (1.803)* (1.735)* (1.844)* (3.098)*** DUAL - -0.037 -0.038 -0.022 -0.088 -0.063 (-0.862) (-0.897) (-0.509) (-1.944)* (-1.407) ROA + 0.245 0.223 0.237 0.228 0.214 (4.890)*** (4.462)*** (4.740)*** (4.297)*** (4.095)*** BIG4 - -0.093 -0.120 -0.099 -0.109 -0.048 (-2.014)** (-2.762)*** (-2.218)** (-2.341)** (-0.979) MB - -0.056 -0.060 -0.056 -0.052 -0.043 (-1.328) (-1.435) (-1.320) (-1.173) (-0.985)
DYEAR YES YES YES YES YES
Panel B: Model statistics
R² Adjust. 9.5% 10.5% 10.0% 10.0% 13.6%
F Stat. 5.473 5.993 5.755 5.261 5.832
Prob. F 0.000 0.000 0.000 0.000 0.000
Variable
was expected: boards with more diversity exhibit a lower level of real earnings management. This result does not support the hypothesis 2a.
Coef. is the estimated coefficient. T-statistics are in parentheses. The symbols *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels , respectively.
Hasfi and Turgut (2012) also found a negative result in their study which investigated the relation between age diversity of directors on corporate social performance. They were puzzled that the age diversity has a significant but negative effect. According to them age diversity possibly leads to polarization, a sort of generation conflict. They stated this conflict might slow the decision-making process due to disagreement and inability to converge. According to Kunze et al. (2013) high levels of age diversity can lead to social fragmentation between different age groups. They state that if a company’s employees’ age composition is highly diverse, it can trigger higher levels of a perceived negative age-discrimination climate, which can negatively affect a company’s performance. Assuming that these higher levels of age-discrimination can also apply to a group of board members, this could negatively affect the corporate governance performance. This could explain the significant positive coefficient in Table 3.
Gender diversity
The coefficients of gender diversity (GENDIV) in Columns 3 and 5 of Table 3 are all negative and significant at a 1 percent level. This significantly negative result provides evidence in support of the third hypothesis, which stated that boards with more gender diversity exhibit a lower level of real earnings management. This result indicates that women play an important role in participating on the board of directors to improve the corporate governance performance. Although these results show that boards with women show a greater corporate governance performance, because the level of real earnings management is lower when the diversity in gender increases, it is worth mentioning that it is important that females are empowered in the workplace (Kyaw et al., 2015). This result also shows that women in North America and Canada have reached a sufficient level of empowerment to be influential.
Nationality diversity
The coefficients of nationality diversity (NATDIV) in Columns 4 and 5 of Table 3 are all negative and significant at a 1 percent level. This significantly negative result indicates that boards with more nationality diversity exhibit a lower level of real earnings management. This result provides evidence in support of the fourth hypothesis.
Control variables
In regard to the control variables, the results in Table 3 are generally in the predicted directions and are consistent with the prior literature (Cohen et al., 2008; Chi et al., 2011; Tsai and Lui, 2015). Collectively, these results suggest that firms that have a higher leverage, are large in size,
Table 5.
Regression analyses of board characteristics on real earnings management-subsamples Panel A: Variable statistics
Predicted
Sign Coefficient t-value Coefficient t-value
Intercept +/- 0.910 5.512*** 0.987 2.152 ** BSIZE + -0.141 -2.581*** -0.066 -0.700 AGEDIV - 0.079 1.505 0.178 1.994** GENDIV - -0.138 -2.609*** -0.164 -1.831* NATDIV - -0.214 -3.997*** -0.029 -0.304 LEV + 0.343 6.503*** 0.337 3.228*** FSIZE + 0.231 4.088*** -0.015 -0.161 DUAL - -0.116 -2.362** -0.024 -0.269 ROA + -0.031 -0.616 0.275 2.697*** MB - -0.056 -1.161 -0.009 -0.100
DYEAR YES YES YES YES
Panel B: Model statistics
R² Adjust. 23.8% 9.9%
F Stat. 12.621 2.519
Prob. F 0.000 0.011
Variable Column 1 (Big 4 N= 361) Column 2 (Non-big 4 N=154)
have higher returns on assets, and are not audited by a big 4 auditor, are more likely to engage in real earnings management activities.
5.3 Sensitivity analysis
Tsai and Lui (2005) mentioned that prior literature indicates that the big 4 auditors supply a higher quality in detecting earnings management than non-big 4 auditors. Therefore it is necessary to perform a supplementary test to examine the validity of the effects of board member characteristics on real earnings management for big 4 and non-big 4 audited firms. To do this additional analysis, a regression is done for two sub-samples: one of firms which are audited by the big 4 and one of firms which are audited by a non-big 4 auditor. The results of the additional analyses of board characteristics on real earnings management for the two sub-samples are provided in Table 5.
Coef. is the estimated coefficient. T-statistics are in parentheses. The symbols *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels , respectively.
Table 4 and Column 1 of Table 5 show that board size (BSIZE), gender diversity (GENDIV) and nationality diversity (NATDIV) are negative and significant at the 1 percent level. However, the coefficients of the variables BSIZE and NATDIV in Column 2 of Table 5 are not significant. The variable AGEDIV shows a positive relation on real earnings management in Table 4 and in Column 1 and 2 of Table 5. In table 5 the AGEDIV is only significant for the sub-sample of the firms audited by a non-big 4 auditor. According to nationality diversity, findings in Table 5 show that nationality diversity exhibits a lower level of real earnings management for firms audited by a big 4 auditor. According to gender diversity, findings in Table 5 show that gender diversity exhibits a lower level of real earnings management for firms audited by a big 4 auditor as well as for firms audited by a non-big 4 auditor. The positive relationship between age diversity and real earnings management is only significant in Column 2 of Table 5, indicating that boards with more age diversity exhibit more real earnings management for firms which are not audited by a big 4 auditor.
6 Conclusion
This study investigates the relationship between board characteristics and their impact on real earnings management. The following board characteristics were investigated: board size, age diversity, gender diversity and nationality diversity. Contrary to theories predicting that smaller boards are more effective, the empirical results reveal that boards with more than 8 members exhibit a lower level of real earnings management for firms in the manufacturing industry. Linda Crompton (“Size Matters”, 2016) states that there is no ‘ideal board size’. She states that as corporations grow in size and complexity, it is likely that the board will also grow. Manufacturing firms are more difficult to monitor due to the complex processes manufacturing firms are facing, indicating that more members are necessary to keep track of these complex processes. According to ‘The Bridgespan Group’ (2016) a larger board size provides enough people to manage the workload of the board more easily. They believe that more board members represent more perspectives and thereby increase the effectiveness of the board of directors. The results in this study indicate that boards with more members are more effective, since they are able to decrease real earnings management activities.
Regarding the diversity of the board composition, both gender diversity and nationality diversity show a negative relationship contrary to age diversity which shows a positive result on real earnings management. Regarding the gender diversity the empirical results in this study show a negative relationship on real earnings management. Thus boards with more women are better able to reduce real earnings management. This finding is consistent with prior studies and implies that women are less likely to take risks particularly in the financial decision environment and are more ethical in their judgement and behaviour than men and therefore use a restrained approach to earnings management (Kyaw et al.,2015; Krishnan and Parsons, 2008; Powell and Ansic, 1997; Gul et al., 2009).
This study shows a significant negative relationship between nationality diversity and real earnings management. This significantly negative result indicates that boards with more nationality diversity exhibit a lower level of real earnings management. This result is consistent with prior literature about nationality diversity, which states that nationally diverse boards will benefit from prosperous interactions, heightened cooperation, and improved outcome, because boards with several nationalities, which bring a wide range of knowledge and experiences with different institutional environments, are more likely to engage in in-depth discussions, take various alternatives into consideration, generate new creative ideas and solve complex tasks (Hambrick et al., 1998).
Thus the effectiveness of the board of directors increases as boards are more diverse in nationality and possess more diversity in gender, since these boards are better able to reduce real earnings management. These results indicate that heterogeneous boards are more effective than homogeneous boards.
Contrary to nationality and gender diversity, age diversity shows a significant positive relation with real earnings management, indicating that boards with more age diversity exhibit a higher level of real earnings management. According to Hasfi and Turgut (2012) age diversity leads to polarization, a sort of a generation conflict, which in turn can slow down the decision-making process caused by disagreement and inability to converge. Kunze et al. (2013) mentioned that a higher degree of age diversity in the board of directors can contribute to a greater age-discrimination climate, which leads to a higher level of real earnings management. Thus the effectiveness of the board of directors does not increase when there is more age diversity.
There are some limitations to this study. First, this study uses a sample of only manufacturing firms so it is not possible to account for unobserved variation between different industries. Further study is necessary to investigate whether in different industries the board size has a similar positive effect on real earnings management. Second, the sample period only covers a 5-year period. This is a relatively short period of time and therefore might limit the generalizability of the empirical results, since real earnings management regularly imposes greater long-term costs on shareholders. Finally, this study investigated 525 firm-year observations of approximately 100 firms, which may be too small to be a representative population. A larger sample is better suited to limit the influence of outliers or extreme observations.
Nevertheless, this study provides insights into the importance of the different board characteristics and their quality in mitigating opportunistic real earnings management behaviour, thus generating better financial reporting quality. For investors and other stakeholders it is important to be aware of these board member characteristics in order to better evaluate the effectiveness of the board of directors. This study is also applicable for policy makers who aim to improve the corporate governance practices of listed firms.
7 References
Adams, R. B., Hermalin, B.E. & Weisbach, M.S. (2010). The role of boards of directors in corporate governance: A conceptual framework and survey. Journal of Economic
Literature, 48(1), 58-107.
Ahmed, A.S., & Duellman, S. (2007). Accounting conservatism and board of director characteristics: an empirical analysis. Journal of Accounting and Economics, 43(2-3), 411-437. Badertscher, B. (2011). Overvaluation and the choice of alternative earnings management
mechanisms. The Accounting Review, 86 (September), 1491-1518.
Bainbridge, S.M., 2003. Director Primacy: The Means and Ends of Corporate Governance. Northwestern University Law Review 97, 547–606.
Beekes, W., Pope, P. & Young, S. (2004). The link between earnings timeliness, earnings conservatism and board composition: evidence from the UK. Corporate Governance, 12 (1), 47-59
Blair, M.M., Stout, L.A., 1999. A team production theory of corporate law. Virginia Law Review 85, 247–328.
Blau, P.M., 1977. Inequality and Heterogeneity: a primitive theory of social structure. New York: Free Press. 677-683
Boehm, S., Bruch, H., and Kunze, F., (2013). Organizational Performance Consequences of Age Diversity: Inspecting the Role of Diversity-Friendly HR Policies and Top Managers’ Negative Age Stereotypes. Florian Kunze, Stephan Boehm & Heike Bruch. Journal of Management Studies (May 2013).
Bunderson, J.S., & Sutcliffe, K.M., (2002). Comparing Alternative Conceptualizations of Functional Diversity in Management Teams: Process and Performance Effects.’’ Academy of Management Journal 45, no. 5 875–893.
Byrnes, J.P., Miller, D.C., and, Schafer, W.D., 1999. Gender differences in risk taking: a meta-analysis, Psychological Bulletin 125, 367–383.
Carter, D. A., B. J. Simkins, and W. G. Simpson, 2003, Corporate governance, board diversity and firm value, The Financial Review 38, 33–53.
Chen, K.Y., Elder, R. & Hsieh, Y.M. (2007). Corporate governance and earnings management: The implication of corporate governance best-practice principles for Taiwanese listed firms. Journal of Contemporary Accounting and Economics, 3(2), 73-105
Cheng, S., (2006). Board size and the variability of corporate performance. Journal of Financial Economics 87 (2008) 157–176
Chi, W.C., Linsic, L.L., and Pevzner, M. (2011). Is enhanced audit quality associated with greater real earnings management? Accounting Horizons, 25(2), 315-335.
Cho, E., & Chun, S., (2015). Corporate social responsibility, real activities earnings management, and corporate governance: evidence form Korea. Asia-Pacific Journal of Accounting & Economics , 2015
Cohen, D.A., Dey, A., & Lys, Z.T., 2008. Real and accrual-based earnings management in the pre- and post –Sarbanas- Oxley periods. The Accounting Review , Vol. 83, No. 3. 757-787 Cohen, D., Zarowin, P. 2010. Accrual based and real earnings management activities around
seasoned equity offerings, Journal of Accounting and Economics 50: 2–19.
Crossland, C., Hambrick, D.C., 2007. How national systems differ in their constraints on corporate executives: a study of CEO effects in three countries. Strategic Management Journal 28(8): 767–789.
Dechow, P.; Kothari, S.; Watts, R. 1998. The relation between earnings and cash flows, Journal of Accounting and Economics 25: 133–168.
Delis, M.D., Gaganis, C., Hasan, I., and Pasiouras, F., (2015). The effect of Board Directors from Countries with Different Genetic Diversity Levels on Corporate Performance. Article in Advance, pp. 1-19.
Earley P.C., Mosakowski E., 2000. Creating hybrid team cultures: an empirical test of transnational team functioning. Academy of Management Journal 43: 26–49.
Eisenberg, T., Sundgren, S., Wells, M., 1998. Larger board size and decreasing firm value in small firms. Journal of Financial Economics 48, 35–54.
Erhardt, N. L., J. D. Werbel, and C. B. Shrader, 2003, Board of director diversity and firm financial performance, Corporate Governance: An International Review 11, 102–111. Ferrero, I.F., Izquierdo, M. Á. F. & Torres, M.J.M., (2015). Age Diversity: An Empirical Study in
the Board of Directors, Cybernetics and Systems, 46:3-4, 249-270
Graham, J.R., Harvey, C.R., & Rajgopal, S, 2005. The Economic Implications of Corporate Financial Reporting. Journal of Accounting and Economics 40 (1): 3–73.
Gul, F. A., Fung, S. Y. K., & Jaggi, B. (2009). Earnings quality: Some evidence on the role of auditor tenure and auditors' industry expertise. Journal of Accounting and Economics,
47(3), 265–287.
Hadini, M., Goranova, M. & Khan, R. (2011) Institutional investors, shareholder activism, and earnings management. Journal of business research, 64(12), 1352-1360.
Hambrick, D. C., T. Cho, and M. Chen, 1996, The influence of top management team heterogeneity on firms’ competitive moves, Administrative Science Quarterly 41, 659–684.
