CHAPTER 8 CONCLUSIONS, IMPLICATIONS AND LIMITATIONS . 239
4.3.2.2 Firm-level explanatory variables
This study employs six firm-level explanatory variables to control for corporate governance characteristics39 which are well-documented in the literature. They consist of: (i) board gender diversity; (ii) board composition; (iii) board leadership structure; (iv) board size; (v) ownership concentration; and (vi) capital structure.
The theoretical framework for including these variables has been mentioned in Chapter 2. The following subsections discuss in turn how they are calculated.
39 In this study, the terms ‘corporate governance structures’, ‘corporate governance characteristics’, and ‘corporate governance mechanisms’ are all interchangeable with each other.
86
Board gender diversity
This research uses gender diversity as a proxy for board diversity. According to Walt and Ingley (2003), board diversity comprises the various characteristics of boards that are associated with decision-making and other administrative processes within the board. These characteristics are categorised as: (i) observable characteristics such as ethnicity, nationality, gender and age; and (ii) unobservable characteristics such as knowledge, educational and professional background, industry experience, among others (Erhardt et al., 2003).
According to Erhardt et al. (2003), the recent empirical studies on the relationship between board diversity and financial performance have concentrated on observable demographic characteristics, including gender and ethnicity.
Therefore, there is a lack of consensus among researchers regarding what board diversity actually is. In line with prior studies (e.g., Adams & Ferreira, 2009;
Ahern & Dittmar, 2012), this research uses gender diversity as a proxy for board diversity. This is also consistent with the suggestion of Srinidhi, Gul, and Tsui (2011) that research about the relationship between board governance and firm performance should consider explicitly female director representation.
As mentioned in Subsection 2.3.1.1 of Chapter 2, both agency theory (Jensen &
Meckling, 1976) and resource dependence theory (Goodstein et al., 1994; Pfeffer, 1973) imply that board gender diversity is value-relevant (Carter et al., 2010).
However, prior empirical studies on this relationship provide inconclusive results due to differences in the way corporate governance empiricists deal with the endogenous nature of the board diversity variable. Following Adams and Ferreira (2009); and Dezsử and Ross (2012), this research treats the board gender diversity
87
variable, defined by the percentage of female directors on BOD (female)40, as an endogenous variable.
Board composition and board leadership structure
As mentioned in Chapter 2, board composition and board leadership structure are key characteristics of board independence. It is common in corporate governance literature to separately use the percentage of independent directors or the percentage of non-executive directors as alternative proxies for board composition. In line with prior research, this study uses the percentage of non- executive directors (denoted as nonexe) as a main proxy for board composition of companies in both the Singaporean and Vietnamese markets.
Given that the Singaporean Code (2005) distinguishes between non-executive directors and independent directors41, it would be expected that the presence of independent directors on corporate boards will have a different impact on board effectiveness and firm performance. For this reason, and to check the robustness of the findings, this study also uses the percentage of independent directors (denoted as indep) as an alternative proxy for board composition of Singaporean companies.
However, data on independent directors are not available for the Vietnamese market for the sampling period because the Vietnamese Code 2007 (MOF, 2007) does not distinguish between non-executive directors and independent directors.
40 Additionally, three other proxies for board gender diversity are included in the model designed for the Vietnamese market. See Subsection 4.3.3.2 for more details.
41 The Guideline 2.1 of the Singaporean Code (2005, p. 2) defines that “an independent director is one who has no relationship with the company, its related companies or its officers that could interfere, or be reasonably perceived to interfere, with the exercise of the director’s independent business judgment with a view to the best interests of the company”.
88
For comparative purposes, the current study does not differentiate between independent and non-executive directors in comparative analyses implemented in Chapter 7. Accordingly, board composition – in Chapter 7 – is measured as the percentage of independent and/or non-executive directors (denoted as indep_nonexe).
Board leadership structure is defined by whether the roles of CEO and board chairperson are separated (non-dual leadership structure) or combined (dual leadership structure). To measure board leadership structure, this study uses a dummy variable (denoted as dual) that takes a value of one if the chairperson of BOD is also the CEO, and zero otherwise. Following Schultz et al. (2010);
Wintoki et al. (2012), this study considers board composition and board leadership structure variables endogenous.
Board size
Board size is measured by the total number of directors on the board. The natural logarithmic form of board size (denoted as lnbsize) is used in the models. As mentioned in Chapter 2, from the perspective of agency theory, good governance prescriptions assume smaller boards are more effective (Yermack, 1996) and thus may contribute positively to firm performance (Jensen, 1993). However, resource dependence theorists suggest that larger board size is positively related to performance (Dalton et al., 1999). Prior empirical evidence is mixed, and hence, no consensus has been reached. As suggested by Schultz et al. (2010); Wintoki et al. (2012), among others, board size is treated as an endogenous variable in this study.
89
Ownership structure
Following Holderness (2009); Munisi, Hermes, and Randứy (2014); and Thomsen, Pedersen, and Kvist (2006), among others, this research defines ownership concentration as the percentage of common stocks held by shareholders who own at least 5% of the total number of a firm’s common stocks (denoted as block). As argued in Subsection 2.3.2 of Chapter 2, ownership concentration is treated as an endogenous variable.
Given data availability, this study follows Reddy, Locke, Scrimgeour, and Gunasekarage (2008) in using another proxy for ownership concentration of Singaporean listed companies to check the robustness of the results. Specifically, this proxy is measured by the ratio of ordinary shares held by twenty largest shareholders to the total number of ordinary shares of a company, named as
‘ownership concentration top 20’ (denoted as blockktop20).
Capital structure
This study also takes account of the potential performance effect of financial leverage (denoted as lev), measured by total debt over total assets. According to Jensen and Meckling (1976), the nature of the agency problem and thus the performance impact of ownership structure may be affected by capital structure.
In more details, leverage is supposed to have an effect on firm value because it can help to discourage managers’ over-investment of free cash flow (Hoechle et al., 2012). In support, Black et al. (2014) also argue that leverage is mechanically associated with Tobin’s Q by its effects on reducing income tax and free cash flow problems. In line with Antoniou et al. (2008), capital structure is considered to be an endogenous variable in the current research.
90