The Relationship between Corporate Social Responsibility Disclosure and Corporate Governance Characteristics: Evidence from VN100.

Le Hoang Nguyen Phuong Linh Nguyen International University, Vietnam National University-HCMC, Vietnam Abstract Due to the increasing importance of Corporate Social Responsibility CSR

The Relationship between Corporate Social Responsibility Disclosure and

Corporate Governance Characteristics: Evidence from VN100

Le Hoang Nguyen Phuong Linh Nguyen

International University, Vietnam National University-HCMC, Vietnam

Abstract

Due to the increasing importance of Corporate Social Responsibility (CSR) practice especially in developing countries where there is a lack of research about CSR and factors affecting it, this study aims to provide a better understanding of CSR disclosure and it relationship with firm corporate governance characteristics in Vietnam Using a sample of 100 biggest market capitalization firms (VN100) from 2014 to 2016, our study reveals a moderate extent of CSR disclosures Board of director size and audit type are found to have a positive significant link with CSR reporting while the meetings by board of director and number of outside directors demonstrate a negative connection with CSR disclosure However, the CEO duality and presence of women directors in the board show no relationship with CSR disclosure Findings from this research are expected to extend the current literature regarding CSR disclosure in Vietnam They can also be helpful for investors in making decisions to invest in corporate citizens as well as policy makers in considering regulations to improve CSR practices in Vietnam

Keywords: Corporate Social Responsibility, Corporate Citizenship, Corporate Governance, Vietnam Stock

Market

1 Introduction

Business and the public raise more awareness about corporate social responsibility (CSR) or corporate citizenship because of the collapse of numerous corporations such as Enron, WorldCom, and Global last twenty years ago and the application of the corporate social responsibility to analyze the equity of firms (Holder -Webb, Cohen, Nath & Wood, 2009) Although CSR is important, CSR disclosure is still optional in many countries including Vietnam On publishing such information, firms want to reduce the information asymmetry issue Moreover, they also would like to commit society to social, ethical, and environmental problems for gaining a good reputation However, there existed little research on the link between the corporate social responsibility (CSR) practice and the governance structure (Akhtaruddin et al., 2009, p.1) Consequently, there is a strong desire to analyze the impact of the governance mechanism on firm’s corporate social responsibility (Rao & Tilt, 2016)

This research can be helpful to markets where the knowledge of CSR is still poor because of the perception that such practices are considered being as philanthropic activities Bui (2010) also showed that in the past, Vietnamese firms paid little attention to corporate ethics for three main reasons: the firms do not understand the disclosure impacts on society, lack both financial resources, and an enforcing legal framework These factors have prevented firms from adopting corporate social responsibility in their policies In Vietnam, most studies about the corporate social responsibility practices are qualitative To the best knowledge of the authors, there existed no extensive studies on CSR reporting and the governance structure of listed firms on Vietnam stock market Prior studies in Vietnam (e.g., Vu, 2012) mainly focused on institutional, firm or individual

determinants of CSR disclosure Thus, there is a scarcity of research on the association between CSR disclosure and corporate governance

Thus, by employing a combined method of qualitative and quantitative analysis to understand the extent

to which firms disclosure CSR and the relationship between CSR disclosure and firm governance structure, our paper hopes to address the gap in the literature Particularly, our paper aims to examine the following research questions:

RQ1 To what extent do listed firms in Vietnam disclose corporate social responsibility?

RQ2 How do corporate governance characteristics influence corporate social responsibility disclosure of listed firms

in Vietnam?

The paper is structured as follows: Section 2 reviews relevant literature Hypotheses are also presented in Section 2 Section 3 presents data and methodology Section 4 discusses the results and Section 5 concludes the study

2 Literature Review

In the early stage, corporate social responsibility is first known as an obligation of businessmen to follow their policies, their decision-making, and desirable actions in favor of the objectives and values of our society

by Bowen (1953) Later, Friedman (2007) established the concept based on two main question: for what are corporations responsible for? and to whom are they responsible? Overall, disclosing corporate citizenship provides companies with financial and strategic gains Money & Schepers (2007) suggested that through social activities and its reporting, firms can gain the trust and goodwill from stakeholders to obtain the competitive benefits; promote their image and their name (Li et al., 2010) Corporate citizenship participation with reporting practice is able to lower the risk that labor quarrels, product safety disgrace, and frauds can affect the firm’ performance and reputation (Waddock & Graves,1997)

2.1 Relationship between the Corporate Social Responsibility and the Corporate Governance Characteristics

There are various theories that can be applied to explain why corporations are willing to disclose their activities relating to community In this study, the authors would like to discuss four main theories that are relevant to the association between the corporate social responsibility disclosure and corporate governance First, the agency theory indicates a conflict of interest between managers and shareholders The agency relationship was considered as a contract in which the principal assigned the work to the agent, but they are unable to examine how the agents behave (Jensen and Meckling, 1976) causing an agency problem as there existed the contradiction between the goal and the interest among them (Eisenhardt,1989) This indicated that the agent could perform in the best interest of them instead of amplifying the shareholders’ wealth Therefore, there existed three forms of costs involving agency problems Firstly, there is monitoring cost that the owners used to mitigate the conflict of interest with the agencies due to different goals (Hill & Jones, 1992) Secondly, bonding cost is the cost utilized by managers to make sure that his actions benefit the owner’s wealth Thirdly, the residual loss leads to the reduction of the principals’ advantage because of different pursued objectives and attitude toward bearing risk Thus, it cannot be denied that adopting the corporate social responsibility reporting is the main way to reduce information asymmetry, mitigate the agency problem and reduce the cost as well

Second, stakeholder theory (Freeman, 1984) indicates that firms operate ethically and responsibly to satisfy the expectation of the public and the stakeholders have right to be updated information through the sustainability performance about how the firm affects the surrounding environment (Freeman, 2010) Moreover, the reputation and the image of the corporation can be improved by both the corporate ethics and the structure

of governance together Therefore, it can sharpen the relations with main financial stockholders and cool down the conflict

of interest among stakeholders named as agency theory By this way, Aguilera, Williams, Conley, & Rupp (2006) found

that participating in the corporate social responsibility reporting would have a positive relation to an efficient mechanism for corporate governance Moreover, according to Barako & Brown (2008), the theory of stakeholder posited that an entity attempts to harmonize its activities with expectations of stakeholders through the social responsibility Stakeholder theory thus introduced the link between corporate governance and the corporate social responsibility practice (Driver & Thompson, 2002; Michelon & Parbonetti, 2012)

Third, legitimacy theory (Suchman, 1995) expands the function of corporate governance by adding more different stakeholders to the principle-agent relationship Moreover, this theory based on the contract between corporations and society Specificly, corporations bear responsibility for the quantity and quality of social resouces and generate the benefits from economic activities for the society because the community provided firms with natural and human resources and permission as well to produce goods Therefore, managers are likely to release their corporate social responsibility activities and other related information to present and sustain the legitimacy of the firms or as a remedial strategy of the firm (Hooghiemstra, 2000)

Fourth, stewardship (Donaldson & Davis, 1991) is one of many theories that can enlighten the relationship between shareholders and managers Pursuant to Davis (1997), the agency theory assumes the discrepancy of interest between the principals and the agents, while the stewardship assumes the agreement of interest between the principals and the managers Despite starting from the different assumption, both theories concentrated on the goal alignments and the rationality of the two parties (Davis, 1997) Stewardship theory suggested that steward managers concentrated on the collective benefit because, in their belief, their personal success is minor to the success of the corporation Moreover, due

to the rationality of steward managers, they identified greater utility in the behavior of cooperation rather than the self-serving behavior Donaldson and Davis (1991) claimed that managers are believed to manage the resources of business effectively by using their intelligence and experiences, so managers are stewards and trustworthy to take actions for protecting the owners’ wealth (Donaldson & Davis, 1991; McWilliams, Siegel, & Wright, 2006) One of these actions was the participation in the corporate social responsibility activities and disclosure, as it is preferred by societies to have long-term benefit on the financial performance (McWilliams et al., 2006) and promotes the position of the organization

as well based on the notion of “doing well by doing good” (McWilliams et al., 2006)

2.1.1 Board of Director Size

From the agency perspective, the board of director size was one of the most vital components of corporate governance in supervising the management of the company by the agents, impacting the transparency as well

as the performance of firms (Fama & Jensen, 1983) Members on board of directors represent shareholders and protect shareholders’ wealth by encouraging the commitment to the corporate social responsibility initiatives

as strategic business activities and promoting a higher level of reporting McWilliams et al (2006); Ahmed Haji (2013); Kathy Rao, Tilt, & Lester (2012b) suggested that the number of members on board decided the effectiveness and efficiency of such mechanism

Board of director size has been found to positively affect the corporate social responsibility reporting in many different nations (e.g., Kathy Rao et al 2012a; Esa & Anum Mohd Ghazali 2012; De Villiers et al 2011 and Jizi et al 2014a) For example, Kathy Rao et al (2012a) found that the extent of environmental information and the board of director size were positively conjoined basing on data collected for the largest 100 firms on Australia stock exchange in 2008 Given the support from prior papers, the following hypothesis is proposed:

H1: There is a positive relationship between the board of director size and corporate social responsibility disclosure 2.1.2 Board of Director Meetings

According to Laksmana, (2008), the number of meetings per year is one of the indicators of the governance structure Laksmana (2008) argued that regular meetings gave directors chances to discuss more information, improve decision-making process, and increase firm value, deter firm growth, consequently promoting the corporate ethics practices (Ahmed Haji, 2013) Previous empirical studies suggested a positive link between the board of director meeting and the corporate citizenship practices (e.g., O’Sullivan et al 2008; Allegrini & Greco 2013a; Chen et al 2010) For example, Allegrini & Greco (2013a) supported a positive influence of board

of director meetings on the degree of the voluntary social responsibility release with the sample size of 177 non-financial Italian listed firms in 2007 Following that, H2 is posited as below:

H2: There is a positive relationship between the number of board meetings and corporate social responsibility disclosure

2.1.3 Audit Type

As argued by agency theory, many companies would like to disclose and become more transparent about their financial reporting and the corporate enterprise performance as well to attract more customers and investors They can obtain their desire to ensure their existence and success (Brown & Deegan 1998; Deegan 2002) by hiring a good auditor (Al-Gamrh et al., 2015) because the large and international audit firms require their client to provide more main information in the reports and websites A review of literature shows evidence of the positive association between audit type and corporate social responsibility reporting (e.g., Raffournier, 1995; Xiao et al., 2004) Take the research of Raffournier in 1995 as an example, he collected data from 161 firms on researching the component of voluntary financial reporting in Switzerland and concluded that audit type linked positively to the corporate citizenship disclosure Given the support from prior papers, this study develops hypothesis

H3: There is a positive relationship between audit type and corporate social responsibility disclosure

2.1.4 Women Directors

The percentage of women directors plays an important role in the corporate governance literature (Carter, Simkins, & Simpson, 2003) Female directors might constrain the board to satisfy the stakeholders’ requirements, thus, the execution of the corporate social responsibility and its performance were more practicable (Zhang, 2012) Moreover, with more female directors, the board can make the most of multiple stakeholders to increase level of the corporate social responsibility Furthermore, women directors can associate their interests with other stakeholders and higher demand from foreign shareholders for the corporate social responsibility disclosure The female directors might constrain the board to satisfy the stakeholders’ requirements thus the execution of the social enterprise and its performance were more practicable (Zhang, 2012) The presence of women directors has been found to positively associate firm corporate social responsibility practice (e.g., Webb, 2004; Mallin & Michelon, 2011; Liao et al, 2015; Krüger, 2009) For instance, Webb (2004) reported a positive link between the percentage of women directors and the corporate social responsibility practice by investigating the structure of board of 394 social firms in the US Hence, this study expects that

H4: There is a positive relationship between the proportion of women directors in the board and corporate social responsibility disclosure

2.1.5 CEO Duality

Agency theory argued that CEO duality concentrates all the authority of the managers and the chairperson

in the same person leading to the interest dispute between individual and firm (Haniffa & Hudaib, 2006) Stewardship theory suggested that duality provides merged command structure boosting the making decision procedure, fasten the enactment of corporate social responsibility performance (Vo, 2010) Webb (2004) investigated the structure of board in 394 social firms in the U.S and found a negative link between the CEO duality and corporate social responsibility reporting Mallin & Michelon (2011) reported that CEO duality is negatively associated with the social performance of 176 corporations Following these papers, this study expects:

H5: There is a negative relationship between CEO duality and corporate social responsibility disclosure

2.1.6 Outside Directors (Independent non-executive directors)

Agency theory claimed that the presence of outside directors is considered being crucial in the corporate governance for efficient management and inspection of the board (Rozaini Mohd Haniffa & Cooke, 2002; Michelon & Parbonetti, 2012) due to their authority and dominance to take the gauge of management performance more fairly than executive directors and take action to reduce the egocentricity behavior (Post, Rahman, & Rubow, 2011) Furthermore, based on the stewardship and the stakeholder theories, independent nonexecutive board members ensure protections for not only the shareholders’ wealth and other stakeholders’ interest as well by pursuing the corporate social responsibility activities to reply to social concerns and meet social obligations (Haniffa & Cooke, 2005a; Ntim & Soobaroyen, 2013)

Webb (2004) investigated the structure of board in 394 social firms in the U.S and reported a positive link between outside directors and the corporate citizenship reporting Similarly, in another context, Ntim & Soobaroyen (2013) examined 75 firms on the stock exchange of Africa and reported a consistent finding Given the support from prior papers, our study proposes the following hypothesis:

H6: There is a positive relationship between the number of outside directors and corporate social responsibility disclosure

2.2 The Relationship between Corporate Social Responsibility Disclosure and Control Variables

As informed from the literature, out study also control for the impact of firm size, listing duration, and profitability on the CSR disclosure There is consistent evidence for the significant positive link between the firm size and the corporate social responsibility disclosure (e.g., Muttakin et al 2015; Reverte, 2009; Mahadeo

et al., 2011) For example, Muttakin et al (2015) examined 116 non-financial corporations on Bangladesh Stock exchange and found a positive link between the firm size and CSR disclosure Similarly, Owusu-Ansah (1998) investigated 49 corporations listed in Zimbabwe and documented a positive association between the listing duration and CSR Most prior empirical studies regarding the link between profitability and the corporate citizenship disclosure report a positive result (e.g., Ntim & Soobaroyen, 2013; Jizi et al., 2014a; Khan, Muttakin,

& Siddiqui, 2013b; Gamerschlag et al., 2011) For example, Gamerschlag et al (2011) comprising 470 observations from selected corporations in Germany found that profitability was significantly and positively associated with CSR reporting

3 Methodology

3.1 Data

The sample include non-financial listed firms in VN100 in the period of 2014-2016 excluding insurance companies and financial institutions due to different regulations, firms that have limit/ no relevant information; delisted firms during 2014-2016; listed firms after January 1 2014 Data are collected from firm annual reports and official websites and carefully examined before conducting analysis

3.2 Model Specification

3.2.1 The relationship between Corporate Social Responsibility Disclosure and Corporate Governance Characteristics

Panel data regression analysis was employed to analyze the data Specification tests (F-Test, Hausman test, and Breusch-Pagan LM test) are used to determine which model (Pooled OLS, REM, FEM) is the most appropriate Breusch-Pagan test and Breusch-Godfrey LM test are employed to check the heteroscedasticity and autocorrelation problems The following regression model is used to test the association between Corporate Social Responsibility Disclosure and Corporate Governance Characteristics

Model 1: CSRD it= β0 + β1 Dual it + β2 BS it + β3 IND it + β4 AT it + β5 BM it + β6 WD it + β7 LNTA it + β8 LD

it + β3 PROF it + ε it

Where:

CSRD it = Corporate Social Responsibility Disclosure for ith firm at the time t

Dual it = CEO Duality for ith firm at time t

BS it = Size of Board of Directors of ith firm at time t

IND it = Independent non-executive directors for ith firm at time t

AT it = Audit Type for ith firm at time t

BM it = Board Meetings for ith firms at time t

LNTA it = Company Size of ith firm at time t

LD it = Listing Duration for ith firm at time t

PROF it = Profitability for ith firm at time t

WD it = Proportion of Women Directors on Board for ith firm at time t

Beta = the regression coefficient

ε it = Error term

3.2.1 The Relationship between Each Theme of Corporate Social Responsibility Disclosure and Corporate

Governance Characteristics

To provide further insights into the link between CSR disclosure and firm corporate governance, corporate governance characteristics are regressed on each CSR theme as below:

Model2: General CSRD it = β0 + β1 Dual it + β2 BS it + β3 IND it + β4 AT it + β5 BM it + β6 WD it + β7 LNTA it + β8 LD it + β3 PROF it + ε it

Model 3: Community CSRD it = β0 + β1 Dual it + β2 BS it + β3 IND it + β4 AT it + β5 BM it + β6 WD it + β7 LNTA it + β8 LD it + β3 PROF it + ε it

Model 4: Environment CSRD it = β0 + β1 Dual it + β2 BS it + β3 IND it + β4 AT it + β5 BM it + β6 WD it + β7 LNTA it + β8 LD it + β3 PROF it + ε it

Model 5: Workplace CSRD it = β0 + β1 Dual it + β2 BS it + β3 IND it + β4 AT it + β5 BM it + β6 WD it + β7 LNTA it + β8 LD it + β3 PROF it + ε it

Model 6: Marketplace CSRD it = β0 + β1 Dual it + β2 BS it + β3 IND it + β4 AT it + β5 BM it + β6 WD it + β7 LNTA it + β8 LD it + β3 PROF it + ε it

3.2.2 The Relationship between Total Corporate Social Responsibility Disclosure (Lagged) and Corporate

Governance Characteristics:

To investigate if there is any relationship between corporate governance characteristics and lagged CSR disclosure, the following model is examined:

Model 7: CSRD it = β0 + β1 𝐷𝑈𝐴𝐿𝑖𝑡−1+ β2 𝐵𝑆𝑖𝑡−1 + β3 𝐼𝑁𝐷𝑖𝑡−1 + β4 𝐴𝑇𝑖𝑡−1 + β5 𝐵𝑀𝑖𝑡−1 + β6 𝑊𝐷𝑖𝑡−1+ β7

𝐿𝑁𝑇𝐴𝑖𝑡−1+ β8 𝐿𝐷𝑖𝑡−1+ β9 𝑃𝑅𝑂𝐹𝑖𝑡−1+ ε

where

CSRD it = Corporate Social Responsibility Disclosure for ith firm at the time t

𝐷𝑈𝐴𝐿𝑖𝑡−1= CEO Duality for ith firm at time t-1

𝐵𝑆𝑖𝑡−1= Size of Board of Directors of ith firm at time t-1

𝐼𝑁𝐷𝑖𝑡−1= Independent Non-executive directors for ith firm at time t-1

𝐴𝑇𝑖𝑡−1= Audit Size for ith firm at time t-1

𝐵𝑀𝑖𝑡−1= Board Meetings for ith firms at time t-1

𝑊𝐷𝑖𝑡−1= Proportion of Women Directors on Board for ith firm at time t-1

LNTAit−1 = Company Size of ith firm at time t

LDit−1 = Listing Duration for ith firm at time t

PROFit−1 = Protitability for ith firm at time t

Beta = the regression coefficient

ε it = Error term

3.3 Measurement of Variables

Our study adopts a comprehensive 51-item checklist by Ahmad, Rashid, and Gow (2017) to measure CSR disclosure Almad et al (2017) developed the check-list to be applied in Malaysia, a neighbor market of Vietnam Thus, we believe that the check-list is also highly applicable in Vietnam context Content analysis has been commonly utilized to investigate CSR reporting (Chan et al., 2014; Abudulah et al., 2011) Annual reports and company websites are read carefully and one (1) point is given for firm that discloses ith item within the checklist and zero (0) otherwise Following Ahmad et al (2017), CSR index is calculated as below:

𝐶𝑆𝑅𝐼𝑖=∑ 𝑋𝑖𝑗

𝑛𝑗 𝑡=1

𝑛𝑗 Where

CSRI= Corporate Social Responsibility Index

nj= Quantity of items expected for jth company

Xij= 1 if firm disclosed ith item; 0 if the firm did not disclose ith item

The measurement of other variables is presented in Table 1:

Table 5: Measurement of variables

CSR Disclosure CSRD Scoring of CSR Disclosure index Ahmad, Rashid & Gow (2017) Independent

Non-executive directors

IND Percentage of outside directors to

total directors

Haniffa & Cooke (2002)

Board Size BS Ln (Total quantity of directors) Jizi et al (2014)

0: Others

Jizi et al (2014)

Board size

H Khan (2010)

0 otherwise

Uwuigbe & Egbide (2012)

Board of director

Meetings BM The quantity of meetings per financial year Prado- Lorenzo & Garcia- Sanchez (2010) and

Woochan Kim (2004)

Total Asset Ho and Yekini (2014)

4 Results

4.1 Corporate Social Responsibility Disclosure

As shown in Figure 1, the extent of CSR reporting of firms in Vietnam is moderate with a mean ranging

from 0.46 to 0.49 from 2014 to 2016 This level is higher than that of Malaysian firms (M=0.22) (Ahmad et al.,

2017) Importantly, CSR index has increased from year to year, suggesting the tendency to disclosure more CSR information by Vietnamese firms The level of CSR reporting in Vietnam may be due to the lack of standards or guidelines In practice, many companies have acknowledged the importance of disclosing CSR and followed international financial reporting standards According to Thompson & Zakaria (2004), firms decide not to report their CSR due to the scarcity of insights on the merits of sustainability reporting and the cost of releasing as well

Figure 2: Corporate Social Responsibility Reporting in Vietnam 2014-2016

Overall, among five main themes, workplace related disclosure has the highest score (M=0.57) on average

(Table 2) A potential explanation is the firms’ concern over the welfare of employees, and the good working environment would boost the efficiency, morale, loyalty of current employees and attract high quality candidates Hence, disclosing such information is important and vital to the growth and operation of firms

Conversely, environment related information accounts for a small proportion (M=0.34) and appears to be less accessible in the annual reports Information regarding marketplace is ranked second (M=0.55) and then followed by the general dimension (M=0.54)

Table 6: Descriptive statistics for firm CSR Reporting

Total CSDI

General Community Environment Workplace Market place Others

0%

10%

20%

30%

40%

50%

60%

70%

80%

46%53%

36%33%

56%53%

73%

47%

56%

38%

33%

57%53%

74%

49%56%

39%35%

58%59%

75%

Corporate Social Responsibility Disclosure in Vietnam 2014-2016

4.2 Descriptive Analysis

Table 3 displays mean, standard deviation, Pearson's correlations for the variables The correlation

coefficients among the variables ranged from r=0.10, p<0.1 to r=0.34, p<0.01 in their absolute values, which

does not indicate any collinearity problem Tolerance statistics and variance inflation factors also refuted the problem

Table 7: Mean, Standard deviation, Min, Max, and Pearson’s correlation

*, **, *** Correlation is significant at the 0.1, 0.05, 0.01 level BS= Board Size BM= Board Meeting AT= Audit Type WD= Women Directors on board IND= independent non-executive directors on board LNTA= Company Size LD= Listing Duration

PRO=Profitability4.3 Regression Results

4.3.1 The Relationship between Total CSR Disclosure and Corporate Governance (Model 1)

Detailed results are shown in Table 4 Both F-test and Hausman test confirm the Fixed-Effect Model (FEM)

as the most appropriate model for the data Additional tests for heteroskedasticity and serial correlation are also conducted Results show data suffers from both autocorrelation and heteroskedasticity Thus, the model

is corrected by Driscoll-Kraay standard error

Table 8: The relationship between total CSR disclosure and corporate governance (Model 1)

Standard error

P> |t|

Std

AT Big4 (1): 63.7%

Non-Big4(0): 36.3% 0 1 0.34*** 0.06 -0.15** 1

DUAL Duality (1) : 28.7%

Separation(0):71.3% 0 1 -0.03 -0.04 -0.10* 0.01 0.17*** 1

IND 1.35 28.84 25.43 32.83 0.15*** 0.11* -0.01 0.02 0.07 -0.14** 1

LNTA 0.22 0.21 0 1.86 0.23** 0.15** -0.1 0.2*** 0.12** -0.08 0.06 1

LD 0.41 1.92 0 2.77 0.14** 0.02 0.03 0.13** -0.07 -0.17*** 0.11* 0.02 1

RPO 0.08 0.08 -0.39 0.38 0.15* 0.06 -0.13** -0.00 0.24*** -0.03 -0.00 -0.06 -0.07 1

Probability 0.01

Breusch-Pagan/Cook-Weisberg for

heteroskedasticity

Chi2 (1) =5.8 p-value = 0.02 Breusch-Godfrey LM for autocorrelation Chi2 (1) =110.95

P-value = 0.02

P-value = 0.00 Breusch and Pagan LM for random

effects

Chibar2(01) = 240.13 P-value = 0.00

P-value = 0.00

*, **, *** significance of correlation at the level of 0.1, 0.05, 0.01

Our results reveal that board of director size is in a significant positive association with CSR reporting (ß= 0.01, p-value = 0.05), which is consistent with Esa & Anum Mohd Ghazali (2012); Jizi et al (2014a); Kathy Rao

et al (2012a) who reported that firms having more board members do engage to the greater extent in the corporate social responsibility activities and the corporate social responsibility reporting practice Moreover, number of board of director meetings are negatively related to CSR disclosure (ß=-0.001, p-value=0.03) This finding in line with Alhazaimeh et al (2014) who confirmed that the more frequently the meetings are held, the less the sustainability information is disclosed Audit type is also found to positively associated with CSR reporting (ß =0.03 and p-value=0.03), suggesting that firms audited by big4 companies are willing to boost their level of corporate social responsibility practice as argued by the agency theory (Raffournier, 1995; Xiao

et al , 2004) With ß= -0.01 and p-value = 0.01, although the finding does not support our hypothesis, it agrees with Eng & Mak (2003b); Haniffa & Cooke (2005b) who documented a negative relationship between CSR and outside directors Hence, with more outside directors in board, corporations tend to experience a decrease in the extent of CSR disclosure

However, there is no evidence of the relationship between the proportion of women directors and CSR disclosure Although the finding does not support our hypothesis, it is in line with the prior studies by Giannarakis (2014); H Khan (2010) One potential explanation is differences in Vietnam cultures Similarly, CEO duality and CSR disclosure shows insignificant relationship which is in line with Cheng & Courtenay (2006); Giannarakis (2014); Said et al (2009) It can possibly be explained that the existence but low rate of CEO duality in the corporate board is unable to boost the better perspective to ethical and environmental concerns Hence, the CEO duality does not have any significant impact effect on the corporate social responsibility reporting

As for control variables, the listing duration (ß= 0.08, p-value =0.05 < 0.1) and firm size (ß= 0.01, p-value

=0.07) are positively associated with the corporate social responsibility reporting Hence, large firms and firms with long listing duration on Vietnam Stock Exchange are likely to improve and upgrade their level of CSR reporting as claimed by agency theory However, there is not a significant link between the firm profitability and the corporate social responsibility practice

4.3.2 The Relationship between each Component of CSR Reporting and Corporate Governance Structure (Model 2

To Model 6)

The study considers not only the total score of CSR disclosure but also the score of each component: the general, the community, the environment, the workplace, the marketplace to provide further insights to the association between CSR reporting and the corporate governance structure Table 5 presents that FEM is the most proper model to be employed in model 2: General and model 5: Workplace while REM is suitable for model 3: Community, model 4: Environment, model 6: Workplace Moreover, this table also presents the statistical tests of six models Particularly, model 3, model 5, model 6 suffer from both heteroskedasticity and