<span class='text_page_counter'>(1)</span><div class='page_container' data-page=1>

<b>THE IMPACT OF BOARD STRUCTURE ON FINANCIAL </b>

<b>LEVERAGE OF VIETNAMESE LISTED FIRMS </b>

<b>Hoang Mai Phuonga*_{, Nguyen Thanh Hong An}a</b>

<i>a_{The Faculty of Economics and Business Administration, Dalat University, Lam Dong, Vietnam}</i>
<i>*_{Corresponding author: Email:}</i>

<b>Article history </b>

Received: November 1st_{, 2020}

Received in revised form: November 22nd_{, 2020 | Accepted: November 30}th_{, 2020}

<b>Abstract </b>

<i>This study examines the impact of board structure on financial leverage decisions, as </i>
<i>measured by the ratio of total debt to total assets, of nonfinancial firms listed on the Ho Chi </i>
<i>Minh City Stock Exchange. Based on a dataset of 1,592 observations collected from 199 firms </i>
<i>for the period from 2012 to 2019, the analysis shows a correlation between board </i>
<i>characteristics and a firm’s financial leverage. Specifically, the higher the number of annual </i>
<i>board meetings or the larger the number of female members on the board of directors, the </i>
<i>lower the rate of financial leverage. On the other hand, the size of the board and the presence </i>
<i>of CEOs on the board do not have a significant influence on financial leverage decisions. A </i>
<i>robust test using the system of generalized method of moments (GMM) to control for </i>
<i>endogeneity generally confirms the results. </i>

<b>Keywords: Agency theory; Board structure; Capital structure; Corporate governance; </b>

Financial leverage; Vietnamese listed firms.

DOI:
Article type: (peer-reviewed) Full-length research article
Copyright © 2020 The author(s).

</div>
<span class='text_page_counter'>(2)</span><div class='page_container' data-page=2>

<b>TÁC ĐỘNG CỦA CẤU TRÚC HỘI ĐỒNG QUẢN TRỊ LÊN ĐỊN </b>

<b>BẨY TÀI CHÍNH CỦA CÁC DOANH NGHIỆP NIÊM YẾT TẠI </b>

<b>VIỆT NAM </b>

<b>Hoàng Mai Phươnga*_{, Nguyễn Thanh Hồng Ân}a</b>

<i>a_{Khoa Kinh tế - Quản trị kinh doanh, Trường Đại học Đà Lạt, Lâm Đồng, Việt Nam}</i>
<i>*_{Tác giả liên hệ: Email:}</i>

<b>Lịch sử bài báo </b>

Nhận ngày 01 tháng 11 năm 2020

Chỉnh sửa ngày 22 tháng 11 năm 2020 | Chấp nhận đăng ngày 30 tháng 11 năm 2020

<b>Tóm tắt </b>

<i>Nghiên cứu này kiểm chứng sự tác động của cấu trúc hội đồng quản trị tới quyết định địn </i>
<i>bẩy tài chính, cụ thể là tỷ lệ tổng nợ trên tổng tài sản, của các doanh nghiệp phi tài chính </i>
<i>niêm yết tại Sở giao dịch chứng khốn Thành phố Hồ Chí Minh trong vòng tám năm từ năm </i>
<i>2012 đến 2019. Dựa trên bộ dữ liệu gồm 1,592 quan sát thu thập từ 199 doanh nghiệp, kết </i>
<i>quả phân tích cho thấy có mối tương quan giữa đặc điểm hội đồng quản trị và địn bẩy tài </i>
<i>chính. Cụ thể, số lượng cuộc họp hội đồng quản trị hàng năm càng nhiều hay số lượng thành </i>
<i>viên nữ trong hội đồng quản trị càng lớn thì tỷ lệ địn bẩy tài chính càng thấp. Trong khi đó, </i>
<i>quy mơ hội đồng quản trị và vai trò kiêm nhiệm của giám đốc điều hành khơng có ảnh hưởng </i>

<i>đáng kể tới các quyết định tài chính. Các phương pháp kiểm định tăng cường, mơ hình động </i>
<i>và phương pháp ước lượng system-GMM được sử dụng để ước lượng hệ số hồi quy, tăng tính </i>
<i>chính xác và khẳng định kết quả thu được từ mơ hình nghiên cứu. </i>

<b>Từ khóa: Cấu trúc hội đồng quản trị; Cấu trúc vốn; Công ty niêm yết tại Việt Nam; Địn bẩy </b>

tài chính; Lý thuyết người đại diện; Quản trị doanh nghiệp.

DOI:
Loại bài báo: Bài báo nghiên cứu gốc có bình duyệt

Bản quyền © 2020 (Các) Tác giả.

</div>
<span class='text_page_counter'>(3)</span><div class='page_container' data-page=3>

<b>1. </b> <b>INTRODUCTION </b>

In recent years, along with the development of the stock market in Vietnam, the
number and quality of listed companies have continuously improved. Today's businesses
are not only larger in scale but also increasingly professional and diversified in their
operations. In this context, abundant capital is a prerequisite for businesses to maintain
and expand production and to promptly meet their growth needs. Therefore, choosing an
appropriate capital structure is an important financial decision that businesses need to
consider to achieve the expected performance.

Corporate governance is an emerging field of research in Vietnam. In recent years,
empirical work has mainly focused on (1) finding the factors that affect the governance
structure or capital structure, and (2) examining the effects of ownership structure,
governance structure, and capital structure on financial performance, leaving the
relationship between governance structure and capital structure underexplored.
Therefore, examining the relationship between the governance structure and capital
structure of listed companies in Vietnam is necessary and can provide useful insights.

This is especially important in the context of an integrated economy, where Vietnamese
businesses are in dire need of effective management strategies to improve
competitiveness as well as corporate value.

Based on the agency theory of Jensen and Meckling (1976), one of the
fundamental theories of corporate governance, we argue that the characteristics of the
agent, in this case the board of directors, can influence the financial decisions of
businesses, particularly financial leverage decisions. Using a dataset of 1592 observations
collected from 199 nonfinancial companies listed on the Ho Chi Minh City Stock
Exchange from 2012 to 2019, our research results show that, in general, the characteristics
of the board of directors have an impact on the financial leverage ratio of businesses.
Firms with active boards of directors, represented by the number of meetings per year,
often control their financial leverage at a lower level than businesses with less active
boards of directors. In addition, the more women present on the board of directors, the
lower the level of leverage in general. On the other hand, board size and the presence of
the Chief Executive Officer (CEO) on the board have no significant impact on financial
leverage.

</div>
<span class='text_page_counter'>(4)</span><div class='page_container' data-page=4>

The paper is organized as follows: First, we briefly present the basic theory and
empirical studies on the relationship between governance structure and capital structure,
from which research hypotheses are developed. Then, the next section presents the
research method and the Models. Finally, we present and discuss the empirical results and
their implications.

<b>2. </b> <b>LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT </b>

<b>2.1. Literature review </b>

The seminal study of Modigliani and Miller (1958) is one of the initial studies on
capital structure. Their proposed theory is formulated in two important propositions

related to firm value and the cost of capital. Modigliani and Miller (1958) show that the
use of debt gives owners a higher rate of return, and their later theory (Modigliani &
Miller, 1963) shows that, with the existence of corporate income tax, the use of debt will
increase the value of the business. In other words, a reasonable level of financial leverage
will satisfy the requirements of managers (about the value of the business) as well as those
of shareholders (about income).

Following Modigliani and Miller’s (1963) research, a series of theories were built
with different perspectives on the corporate capital financing Model. Of these, the agency
theory of Jensen and Meckling (1976) is one of the prominent theories on the relationship
between optimal capital structure and governance structure in controlling conflicts of
interest between shareholders and managers. Conflicts of interest arise from the transfer
of certain decision-making powers to the agent in the relationship between the principal
(shareholders) and the agent (managers). Both sides want to advance their interests, and
there are always reasons to believe that the agent does not always act in the best interest
of the principal. In other words, the managers will be motivated to use the resources of
the business for their personal benefit instead of for the benefit of the shareholders.

</div>
<span class='text_page_counter'>(5)</span><div class='page_container' data-page=5>

structure. Their research shows that executives strive to avoid debt, and when there is no
demand from the shareholders, the debt ratio is kept at a lower-than-optimal level.
Consequently, if measures to minimize management entrenchment are applied, the
leverage ratio tends to increase.

Taking a closer look at the theoretical discussion on the relationship between
managers and shareholders addressed by agency theory, the divergence of interests
between shareholders and managers can be reduced by establishing effective monitoring
mechanisms to limit the managers' self-interested behavior. More specifically, the
decision to borrow is one of the monitoring options that makes managers hesitate because
this will place the business under the supervision of many outside parties. Hence,
understanding the effects of governance characteristics on capital structure decisions can

help us understand effective monitoring mechanisms. Empirical studies have shown that
a number of corporate governance characteristics, including board size, frequency of
meetings during the year, number of female members, and CEO duality, influence
corporate debt decisions. A brief discussion of board characteristics and their relationship
to financial leverage is given below.

<i>2.1.1. Board size and capital structure decisions </i>

A company's financing decisions are governed by its board of directors (BOD).
The operational efficiency of the BOD is the key to the success of the business. According
to Adams and Mehran (2003), a large board of directors can effectively monitor the
company's operations and provide better expertise. On the other hand, Lipton and Lorsch
(1992) claim that large boards perform less efficiently than small ones. Board size should
be limited to a maximum of ten members and boards with eight or nine members is the
most reasonable.

Existing studies on the relationship between board size and financial leverage
yield inconclusive findings. Berger et al. (1997) and Anderson, Mansi, and Reeb (2004)
find a significant, negative correlation between board size and financial leverage.
Conversely, the studies of Kyereboah-Coleman and Biekpe (2006), Abor (2007), Bokpin
and Arko (2009), and Rose, Munch-Madsen, and Funch (2013) find a positive
relationship between board size and the capital structure of businesses. This shows that
the effectiveness of the board in monitoring management behaviors can directly, or
indirectly, improve a company's access to debt. Finally, Wiwattanakantang (1999) and
Wen, Rwegasira, and Bilderbeek (2002) find no relationship between board size and
financial leverage.

<i>2.1.2. Number of board meetings and a firm’s financial leverage </i>

</div>
<span class='text_page_counter'>(6)</span><div class='page_container' data-page=6>

Chidambaran (2007) and Ntim and Osei (2011) find a positive relationship between the

frequency of board meetings and firm performance. Regular BOD meetings often tend to
produce better financial performance (Johl, Kaur, & Cooper, 2015) and the number of
board meetings should be at least four meetings each year according to Eluyela et al.
(2018). Buchdadi, Ulupui, Dalimunthe, & Pamungkas (2019) and Kajananthan (2012)
indicate that regular board meetings can lead to more debt decisions, thereby taking
advantage of outside capital to modernize, expand, exploit investment opportunities, and
increase the market value of the business. The study also uncovers the important role of
supervision through board meetings in agency theory. Firms with high leverage are also
likely to have more frequent board meetings (Al-Najjar, 2011). A study by Francis,
Hassan, and Wu (2015) finds that companies with infrequent board meetings performed
significantly worse during the financial crisis. Stephanus, Anastasia, and Toto (2014) find
a significant negative relationship between the frequency of board meetings and debt
ratio. Frequent board meetings can increase costs, time, and administrative support
requirements for a company.

<i>2.1.3. Board gender diversity and a firm’s financial leverage </i>

Globally, over the past two decades, female representation on corporate boards of
directors has increased significantly in a number of markets. At the same time, the issue
of board gender diversity has also been debated and is the basis to consider the impact of
female directors on a company's operations, including whether a greater presence of
women on a BOD affects corporate financial decisions, and why, in fact, few women are
on boards. The pioneering research on this topic was conducted by Morrison, White, and
Velsor (1987), and this topic has increasingly attracted the attention of many researchers
globally, both in developed and developing countries.

First, it has been shown that the maturity of a firm affects the composition of its
board. A high degree of board diversity is positively related to corporate financial results.
Looking at recent empirical studies, Tran, Hoang, and Tran (2015) investigate the impact
of gender diversity in the BOD on company performance and find that the proportion of

women directors on the board had a significant positive effect on the financial results of
banks in ASEAN from 2009 to 2013. A BOD is more active in the presence of at least
three female representatives. Gender-balanced boards are also more likely to replace
ineffective managers (Schwartz-Ziv, 2017). Rose et al. (2013) and Marinova, Plantenga,
and Remery (2015) studied the effects of women directors on the activities of companies
in Germany and the Nordic bloc in 2010. However, their results show that the proportion
of women on a BOD has no apparent influence on financial decisions. In two other
studies, Harris (2014) and Abobakr and Elgiziry (2015) find a significant negative
relationship between the proportion of female directors and financial leverage, especially
on boards where the presence of females accounts for 25% or more.

<i>2.1.4. CEO dual roles and a firm’s financial leverage </i>

</div>
<span class='text_page_counter'>(7)</span><div class='page_container' data-page=7>

decisions. Duality exists when the CEO of a company is also the chairman of the board.
On the one hand, according to Sheikh and Wang (2012), duality provides clear direction
from a single leader who can react more quickly to outside events. On the other hand,
duality increases the CEO's decision-making power by providing a broader base of power
and strengthening control (Boyd, 1995). As a result, assigning both tasks to the CEO can
weaken the board's control and influence financial decisions.

Empirical research on this relationship gives mixed results. Fosberg (2004) asserts
that a dual leadership structure is effective in increasing the amount of debt in a firm's
capital structure. Abor (2007) finds a significant positive relationship between CEO
duality and financial leverage. A CEO often tries to finance the company's operations
using debt capital instead of issuing new equity (Bokpin, & Arko, 2009). Meanwhile,
Kyereboah-Coleman and Biekpe (2006) find a clear negative relationship between CEO
duality and short-term and total leverage, asserting that agency costs increase when a
CEO is chairman of the board, which discourages investors from investing in the business.
They also report a positive link between CEO duality and long-term leverage, but this
relationship is not statistically significant. Research by Tarus and Ayabei (2016) also

confirms the negative relationship between CEO duality and financial leverage. CEOs
who are also chairs of the BOD are given too much power and have the ability to use less
financial leverage to avoid risks associated with borrowing. The study by Simpson and
Gleason (1999) examines the effect of CEO duality on the use of financial leverage at
300 banks. The results show that CEOs can influence the internal control system in a way
that reduces the likelihood of financial difficulties for the company. This means that they
take less risk, resulting in underuse of financial leverage.

<b>2.2. Research in Vietnam </b>

In Vietnam, recently published studies focus on finding factors that influence the
capital structure of firms listed on the Vietnamese stock markets. Specifically, Đặng and
Quách (2014) identify three factors that have a strong impact on the capital structure of a
firm, namely, firm size, profitability (positive impact), and taxes (negative impact).
Previously, Trương and Võ (2008) affirm that capital structure is positively correlated
with company size, industry, and revenue growth and is inversely correlated with
profitability. In addition, capital structure is positively correlated with the number of
directors. Recently, a series of studies on factors affecting the capital structure of firms in
specific industries was also conducted. The studies included firms in the logistics industry
(Lương, Phạm, Nguyễn, Nguyễn, Nguyễn, & Phạm, 2020), the food industry (Lê, Bùi, &
Lê, 2020), the Vietnam Oil and Gas Group (Vũ & Nguyễn, 2013), and the seafood
industry (Nguyễn, 2008). Most studies show that firm size, growth rate, and profitability
are positively correlated with capital structure. Some other factors that are negatively
correlated are also mentioned, including taxes, liquidity, profits, and business risk.

</div>
<span class='text_page_counter'>(8)</span><div class='page_container' data-page=8>

(2016) examine the impact of corporate governance (state ownership, financial
institutions, foreign investors, members of the BOD, and the largest shareholders of the
firm) and firm characteristics (size, profitability, tangible assets, tax shield, and the gap
between optimal leverage and observed leverage) on capital structure decisions. Research
results show that corporate capital structure not only depends on the characteristics of the

business, but is also influenced by enterprise ownership characteristics. In another study
by Phan, Trần, and Trần (2017), the role of CEO duality is examined. Their study
confirms that firms with a dual leadership structure performed more effectively.

As previous research is still inconclusive, a comprehensive study with a large set
of data on companies listed on the Vietnamese stock market would provide significant
insights into the relationship between BOD structure and capital structure (financial
leverage).

Based on the theory and previous research results, this study hypothesizes that:

• H1: Board size has an impact on financial leverage.

• H2: The number of annual board meetings has an impact on financial

leverage.

• H3: The number of female directors on the BOD has an impact on financial

leverage.

• H4: CEO duality has an influence on financial leverage.

<b>3. </b> <b>DATA AND RESEARCH METHOD </b>

<b>3.1. </b> <b>Definitions of variables and data collection method </b>

This study examines the relationship between the BOD structure and the financial
leverage of companies listed on the Vietnamese stock market. For quantitative analysis,
the authors use leverage ratio, which is defined as the ratio of total debt to total assets of

the firm, similar to the study by Haque, Arun, and Kirkpatrick (2011).

The independent variables used in this study include the size of the board, the
number of board meetings per year, the number of female directors on the board, and an
indicator variable indicating whether the chairman also holds a CEO position. To increase
the effectiveness of the estimate, two variables representing board size and the number of
board meetings were converted to logarithms prior to analysis. In addition to the number
of female directors on the board, the authors also use two other definitions of gender
diversity of the BOD, namely, the percentage of female directors on the board and an
indicator variable indicating the presence of female directors. Using different definitions
of the gender variable in the analysis will help increase the reliability of the results.

</div>
<span class='text_page_counter'>(9)</span><div class='page_container' data-page=9>

similar to studies on the effect of firm characteristics on financial leverage according to
Bradley, Jarrell, and Kim (1984), Castanias (1983), Long and Malitz (1985), and Titman
and Wessels (1988). These studies generally agree that financial leverage has a positive
relationship with firm size, fixed assets, and growth rate, and an inverse relationship with
returns and liquidity.

Detailed definitions of the variables are presented in Table 1.

<b>Table 1. Variable definitions </b>

Variable Code Definition/Formula
Dependent variable

<b>Financial leverage </b> Lev <b>Total debt/Total assets </b>
Independent variables

<b>BOD size </b> Lbsize <b>Logarithm (Number of directors on the board) </b>
<b>BOD meeting frequency </b> Lmeet <b>Logarithm (Number of BOD meetings per year) </b>

<b>Female directors </b> Female <b>Number of female directors on the BOD </b>

<b>CEO duality </b> Ceodual <b>Equals 1 if the CEO is also the BOD chairman, 0 otherwise </b>
Control variables

<b>Firm size </b> Lfsize <b>Logarithm (Total assets) </b>

<b>Fixed assets </b> Fixed_assets <b>(Total assets - Short-term assets)/Total assets </b>
<b>Liquidity </b> Liquidity <b>Short-term assets/Short-term liabilities </b>
<b>Profitability </b> ROA <b>Net profit/Total assets </b>

<b>Growth </b> Salegrowth <b>Annual sales growth </b>

The data are collected from audited financial reports, annual reports, and annual
executive reports of nonfinancial companies listed on the Ho Chi Minh City Stock
Exchange from 2012 to 2019. Companies with insufficient data are excluded from the
sample.

<b>3.2. </b> <b>Research method </b>

To analyze the relationship between the variables representing the characteristics
of the BOD and the leverage ratio, the authors propose the following research Model:

𝐿𝑒𝑣_𝑖𝑡 = 𝛽₁+ 𝛽₂𝐿𝑏𝑠𝑖𝑧𝑒_𝑖𝑡+ 𝛽₃𝐿𝑚𝑒𝑒𝑡_𝑖𝑡+ 𝛽₄𝐹𝑒𝑚𝑎𝑙𝑒_𝑖𝑡+ 𝛽₅𝐶𝑒𝑜𝑑𝑢𝑎𝑙_𝑖𝑡

+ 𝛽₆𝐿𝑓𝑠𝑖𝑧𝑒_𝑖𝑡+ 𝛽₇𝐹𝑖𝑥𝑒𝑑_𝑎𝑠𝑠𝑒𝑡𝑠_𝑖𝑡+ 𝛽₈𝑅𝑂𝐴_𝑖𝑡+ 𝛽₉𝐿𝑖𝑞𝑢𝑖𝑑𝑖𝑡𝑦_𝑖𝑡

+ 𝛽₁₀𝑆𝑎𝑙𝑒𝑔𝑟𝑜𝑤𝑡ℎ_𝑖𝑡+ 𝜀_𝑖𝑡 (1)

</div>
<span class='text_page_counter'>(10)</span><div class='page_container' data-page=10>

variables in the Model, causing the estimate to be biased and inconsistent. Second, since

this Model does not take advantage of information from the differences between firms,
the estimates may be less accurate (Wooldridge, 2002).

As a remedy, the authors restructure Model (1) to incorporate the differences

among companies (representing by <i> </i>in the new Model) in the dataset:

𝐿𝑒𝑣𝑖𝑡 = 𝛽1+ 𝛽2𝐿𝑏𝑠𝑖𝑧𝑒𝑖𝑡+ 𝛽3𝐿𝑚𝑒𝑒𝑡𝑖𝑡+ 𝛽4𝐹𝑒𝑚𝑎𝑙𝑒𝑖𝑡+ 𝛽5𝐶𝑒𝑜𝑑𝑢𝑎𝑙𝑖𝑡

+ 𝛽₆𝐿𝑓𝑠𝑖𝑧𝑒_𝑖𝑡+ 𝛽₇𝐹𝑖𝑥𝑒𝑑_𝑎𝑠𝑠𝑒𝑡𝑠_𝑖𝑡+ 𝛽₈𝑅𝑂𝐴_𝑖𝑡 + 𝛽₉𝐿𝑖𝑞𝑢𝑖𝑑𝑖𝑡𝑦_𝑖𝑡

+ 𝛽₁₀𝑆𝑎𝑙𝑒𝑔𝑟𝑜𝑤𝑡ℎ_𝑖𝑡+ 𝜇_𝑖 + 𝜀_𝑖𝑡

(2)

Model (2) is estimated by the regression method with random and fixed effects,
respectively. The LM test is used to choose between the POLS regression Model and the
regression Model with random effects. Then, the Hausman test is used to choose between
the regression Model with random effects and the regression Model with fixed effects.

However, the recent research of Liao, Mukherjee, and Wang (2015) indicates that
firms tend to adjust their leverage toward an optimal value over time. As discussed, if the
BOD actually impacts financial leverage decisions, the adjustment effect implies that the
BOD would refer to the past leverage level when deciding the future leverage ratio. In

<i>other words, Levit and Levit-1</i> are correlated. The fact that Model (2) omits this important

<i>variable (i.e., Levit-1) reduces the accuracy of the estimates. Furthermore, if Levit-1 </i>is

correlated with the present structure and operation of the BOD, a case which is raised in

<i>previous research by Berger et al. (1997), the omission of Levit </i>in Model (2) would render

the estimates inefficient and inconsistent. As a remedy, Model (2) is restructured as follows:
𝐿𝑒𝑣_𝑖𝑡 = 𝛽₁+ 𝛽₂𝐿. 𝐿𝑒𝑣_𝑖𝑡+ 𝛽₃𝐿𝑏𝑠𝑖𝑧𝑒_𝑖𝑡+ 𝛽₄𝐿𝑚𝑒𝑒𝑡_𝑖𝑡 + 𝛽₅𝐹𝑒𝑚𝑎𝑙𝑒_𝑖𝑡

+ 𝛽₆𝐶𝑒𝑜𝑑𝑢𝑎𝑙_𝑖𝑡+ 𝛽₇𝐿𝑓𝑠𝑖𝑧𝑒_𝑖𝑡 + 𝛽₈𝐹𝑖𝑥𝑒𝑑_𝑎𝑠𝑠𝑒𝑡𝑠_𝑖𝑡+ 𝛽₉𝑅𝑂𝐴_𝑖𝑡

+ 𝛽₁₀𝐿𝑖𝑞𝑢𝑖𝑑𝑖𝑡𝑦_𝑖𝑡+ 𝛽₁₁𝑆𝑎𝑙𝑒𝑔𝑟𝑜𝑤𝑡ℎ_𝑖𝑡 + 𝜇_𝑖 + 𝜀_𝑖𝑡

(3)

Model (3) cannot be consistently estimated by the methods used for Models (1)

<i>and (2) because the endogeneity problem caused by the inclusion of the variable Levit-1</i>.

</div>
<span class='text_page_counter'>(11)</span><div class='page_container' data-page=11>

<b>4. </b> <b>RESULTS AND DISCUSSION </b>
<b>4.1. Descriptive statistics </b>

The authors collected data based on the variable definitions and methodology
presented in Section 3. The final dataset included 1,592 observations collected from 199
firms from 2012 to 2019. Summary statistics are presented in Table 2.

<b>Table 2. Descriptive statistics </b>

Number of

observations Mean Median Std. dev. Min Max
Lev 1,592 0.49 0.50 0.21 0.03 1.29
Bsize 1,592 5.76 5.00 1.33 1.00 11.00

Meet 1,592 10.97 7.00 11.61 1.00 170.00
Female 1,592 0.90 1.00 1.04 0.00 6.00
Female_% 1,592 0.15 0.14 0.17 0.00 1.00
Female_dummy 1,592 0.55 1.00 0.50 0.00 1.00
Ceodual 1,592 0.28 0.00 0.45 0.00 1.00
Lfsize 1,592 21.17 21.08 1.26 18.46 26.72
Fixed_assets 1,592 0.41 0.37 0.23 0.01 0.98
ROA 1,592 0.06 0.05 0.08 -0.85 0.78
Liquidity 1,592 2.28 1.58 2.46 0.05 39.37
Salegrowth 1,592 0.18 0.07 1.53 -24.16 29.56

<i>Notes: The analysis is performed based on 1,592 observations collected from 199 nonfinancial companies </i>

listed on the Ho Chi Minh City Stock Exchange from 2012 to 2019; Bsize is the number of board members;
Lev is the ratio of total debt to total assets of the business; Meet is the number of board meetings per year;
Female is the number of female members on the Board of Directors. Female_% is the percentage of female
directors; Female_dummy is an indicator variable, with the value equal to 1 if the board has female
members and 0 if not; Ceodual is an indicator variable with value equal to 1 if the chairman of the BOD
also holds the position of CEO and 0 if not. Lfsize is the logarithm of the firm's total book value;
Fixed_assets is the ratio of fixed assets to total assets; ROA is the ratio of after-tax profit to total assets;
Liquidity is the ratio of short-term assets to short-term liabilities. Salegrowth is the rate of sales growth.

</div>
<span class='text_page_counter'>(12)</span><div class='page_container' data-page=12>

estimating the regression in the following section. As for the independent variables, the
surveyed firms' boards of directors have an average of about six members, with the
majority of the firms having a BOD consisting of five members. This number is almost
equivalent to that found by Nguyen et al. (2015) for 2008 to 2011, showing that the size
of the BOD of listed companies in Vietnam remained quite stable through the years.
Boards usually hold an average of 11 meetings per year, but most of them only hold about
seven meetings. Regarding the proportion of women, the BODs of the surveyed
companies average 15% female members (equivalent to about one person) and more than

55% of the companies surveyed have female directors. This figure is higher than in the
study of Nguyen et al. (2015), in which the proportion of women directors was 12% and
about 51% of surveyed firms had female board members. This is a sign that there has
been an improvement in gender diversity on the boards of Vietnamese-listed companies
in recent years. Finally, about 28% of the surveyed businesses have a chairman of the
board who concurrently holds the position of CEO. This figure has decreased by 4%
compared with the corresponding study of Nguyen et al. (2015) for 2008 to 2012.

Table 4 presents the correlation coefficients among the variables. The results show
that the correlation coefficients between the independent and control variables are below
0.5, implying that multicollinearity is not a serious problem in the regression analysis.
This conclusion is supported by the calculated VIF for the independent and control
variables of less than 2. In addition, the correlation analysis results show that the number
of board meetings is positively correlated, while board size is negatively correlated, with
firm leverage. However, this result is not reliable because the correlation analysis does
not take into account the impact of other covariates on the relationship between the two
considered variables. To analyze the relationship between the independent variables and a
firm's leverage ratio consistently and effectively, regression analysis should be performed.

<b>Figure 1. Financial leverage by industries </b>

</div>
<span class='text_page_counter'>(13)</span><div class='page_container' data-page=13>

<b>Table 3. Financial leverage by various industries </b>

2012 2013 2014 2015 2016 2017 2018 2019 Mean
Wholesale 0.6462 0.6185 0.6081 0.5859 0.5629 0.5573 0.5487 0.5538 0.5852
Retail 0.5491 0.5644 0.5196 0.5360 0.5484 0.5732 0.5040 0.5209 0.5394
Information Technology 0.4706 0.4939 0.5120 0.4918 0.4852 0.4343 0.4619 0.4766 0.4783
Accommodation and

Catering 0.3022 0.2541 0.2810 0.5585 0.5690 0.3811 0.4367 0.5068 0.4112

Mining 0.3388 0.3002 0.2861 0.3183 0.2498 0.2512 0.2940 0.2571 0.2869
Arts and Entertainment 0.1309 0.1498 0.1558 0.1304 0.1313 0.1153 0.1214 0.1107 0.1307
Production 0.4431 0.4583 0.4496 0.4423 0.4502 0.4606 0.4665 0.4582 0.4536
Agriculture 0.4282 0.3625 0.3295 0.3733 0.2975 0.3361 0.3333 0.4119 0.3590
Utilities 0.4936 0.4735 0.4576 0.4550 0.4566 0.4606 0.4430 0.4493 0.4611
Transportation and

Warehousing 0.4058 0.4002 0.3994 0.4076 0.3926 0.3777 0.3668 0.3408 0.3864
Construction and Real

Estate 0.5845 0.5856 0.5855 0.5859 0.5839 0.5862 0.5876 0.5658 0.5831

<b>Mean </b> <b>0.4953 0.4959 0.4880 0.4874 0.4839 0.4872 </b> <b>0.4859 </b> <b>0.4774 </b> <b>0.4876 </b>

Notes: Companies are classified by their first registered type of business. Industries are classified in
accordance with NAICS (North American Industry Classification System) 2007.

<b>Table 4. Correlation coefficients of research variables </b>

Lev Lbsize Lmeet Female Ceodual Lfsize Fixasset ROA Liquidity

Sale-growth
Lev 1.00

Lbsize -0.05** 1.00

Lmeet 0.12*** 0.09*** 1.00

Female -0.03 0.25*** 0.02 1.00

Ceodual 0.02 -0.01 0.01 0.10*** 1.00

Lfsize 0.28*** 0.29*** 0.20*** 0.17*** -0.01 1.00

Fixasset -0.16*** 0.17*** -0.11*** -0.05* -0.11*** 0.10*** 1.00

ROA -0.46*** 0.10*** -0.04 0.04 -0.06** -0.04* -0.05* 1.00

Liquidity -0.46*** -0.02 -0.03 -0.03 0.07*** -0.14*** -0.15*** 0.29*** 1.00

</div>
<span class='text_page_counter'>(14)</span><div class='page_container' data-page=14>

<b>4.2. Regression analysis </b>

To analyze the relationship between the characteristics of the BOD and financial
leverage ratio, the authors conducted a preliminary empirical analysis using a static Model
structured as in Models (1) and (2) of Section 3. The results are shown in Table 5.

Before interpreting the results, a number of standard tests were performed to select
the optimal Model. First, Breusch-Pagan’s LM test was performed to choose between the
POLS Model (Column 1) and the regression Model with random effects (Column 2). The
results show that the test statistic, Chibar-square (1) = 22.800, corresponds to the value
Prob (chibar-square) = 0.000, indicating that the Model with random effects (Column 2)
was more effective than the POLS Model (Column 1). To choose between the random
effects Model (Column 2) and the fixed effects Model (Column 3), the Hausman test was
performed. The results show that the test statistic Chi-square (16) = 1,055.300, which
corresponds to Prob (chi-square) = 0.000, and that the fixed effects Model (Column 3) is
more consistent and efficient than the random effects Model (Column 2). In conclusion,
the two tests indicate that the Model with fixed effects should be used.

First, the regression results with the POLS Model (Column 1) suggest that the
makeup of the BOD has an impact on the firm's leverage decisions. Specifically, the

<i>variable Female has a negative coefficient and is statistically significant at 1%, implying </i>
that having more female directors is associated with a lower rate of financial leverage.
<i>The remaining three characteristics of the board, namely, size of the board (Lbsize), </i>
<i>number of meetings per year (Lmeet), and a CEO who is also the chairman of the board </i>
<i>(Ceodual), have no impact on a firm’s leverage ratio. However, the results are not reliable </i>
because estimation by the POLS method can be biased and inconsistent due to the
omission of firm-level characteristics.

To take into account unobserved factors at the firm level, the tests show that the
regression Model with fixed effects (Column 3) is optimal. The regression results in
Column (3) show that, after taking into account other unobservable characteristics at the
firm level, the variables represent the performance of the BOD, including the size of the
<i>board (Lbsize), the number of meetings per year (Lmeet), the number of female members </i>
<i>(Female), and whether the CEO is also the chairman of the BOD (Ceodual), have no </i>
impact on the leverage ratio of the firms. In Columns (4) and (5), the authors re-estimate
Model (2) with two other common definitions of board gender diversity, namely, the
<i>proportion of female directors (Female_%) and presence of female directors </i>
<i>(Female_dummy). </i>

</div>
<span class='text_page_counter'>(15)</span><div class='page_container' data-page=15>

<b>4.3. </b> <b>Robust tests </b>

Some recent studies have shown that firms tend to keep their leverage at a target
level they consider optimal. This means that when the leverage is higher than the target,
the business will adjust it downward and when the leverage is lower than the target, they
will adjust it upward (Liao et al., 2015). This adjustment mechanism implies that leverage

<i>in the present (Levit) is correlated with leverage in the past (L.Levit</i>). To the extent that

<i>this is true, Models (1) and (2) have omitted an important factor (L.Levit</i>), and this would

lead to inefficiencies in the estimates presented in Table 5.

More seriously, if the omitted variable and the independent variables (in this case,
the quality of the board's performance) are correlated (the so-called endogenous
phenomenon), estimates for the relationship between the BOD’s quality and the firm’s
financial leverage would be biased and inconsistent, implying that the conclusions based
on the static Model in Table 5 do not reflect reality. This possibility is even more evident
as some past studies, such as Kyereboah-Coleman and Biekpe (2006), Abor (2007),
Bokpin and Arko (2009), and Rose et al. (2013), have shown that the BOD really has a

<i>voice in deciding a firm’s financial leverage. Correlation analysis between L.Levit</i> and the

independent variables in Model (2) also shows that the correlation coefficient between

<i>L.Levit and Lbsizeit</i> is -0.05 and statistically significant at 10%, the correlation coefficient

<i>between L.Levit and Lmeetit</i> is 0.13 and statistically significant at 1%, the correlation

<i>coefficient between L.Levit and Lfsizeit</i> is 0.25 and statistically significant at 1%, the

<i>correlation coefficient between L.Levit and Fixed_assetsit</i> is -0.16 and statistically

<i>significant at 1%, the correlation coefficient between L.Levit and ROAit</i> is -0.39 and

<i>statistically significant at 1%, and the correlation coefficient between L.Levit</i> and

<i>Liquidityit</i> is -0.42 and statistically significant at 1%. Based on the results of previous

studies and preliminary empirical evidence, we find that endogeneity is very likely to
exist, and the hypothesis testing results based on the estimates in Table 5 may not be

correct. Therefore, the authors reconstructed Model (2) by adding the one-step lag
<i>variable of the dependent variable (L.Levit</i>) to the list of independent variables. The

modified Model is shown in Model (3).

Model (3) was re-estimated using the POLS method (Column (1)) and the
regression method with fixed effects (Column (2)). However, estimates using POLS or
fixed effects regression cannot consistently estimate the regression coefficients in the
presence of endogeneity due to dynamic Model structures (Blundell & Bond, 1998).
Therefore, Model (3) was re-estimated by the system GMM method, which is capable of
estimating the regression coefficients consistently in the presence of endogeneity. The
results are presented in Column (3). Columns (4) and (5) present the estimation results of
<i>Model (3) by the system GMM method with the variable Female replaced by Female_% </i>
<i>and Female_dummy. </i>

<i>Estimated results in all Columns in Table 6 show that L.Levit</i> has a positive and

<i>statistically significant correlation of 1% with Levit</i>. This means that the dynamic Model

</div>
<span class='text_page_counter'>(16)</span><div class='page_container' data-page=16>

Columns (3), (4), and (5) present the estimates for Model (3) by the system GMM method
<i>with the variables Female, Female_%, and Female_dummy, respectively. The </i>
Arellano-Bond test statistic for series correlation shows that the second-order series correlation,
AR (2), does not exist. Therefore, the authors chose lag variables of order 3 onwards as
instrumental variables. This eliminates the correlation between the instrumental variables
and the error and remedies the endogeneity problem completely. The Hansen J-test
statistic in all three Columns is not statistically significant, i.e., the instrumental variables
are not correlated with the error term, suggesting that the endogeneity problem is fixed
(Wintoki, Linck, & Netter, 2012; Roodman, 2009).

The results in Columns (3), (4), and (5) show that, in general, the quality of the

board of directors has an impact on the firm's financial leverage decisions, with two out
of four variables reflecting the characteristics of the board of directors having statistically
significant coefficients. Specifically, the number of board meetings during the year is
negatively correlated and statistically significant in the Model at the 5% and 10% levels
<i>with the variables Female and Female_dummy. This implies that enterprises with a large </i>
number of BOD meetings per year often have low financial leverage. Conversely,
leverage ratios are higher in businesses where BODs are less active. Vafeas (1999)
suggests that the number of board meetings demonstrates a positive effect on the
performance of the board and is a good representation of managerial oversight. Therefore,
the research results show that companies with more-active BODs often control their
leverage at a lower level than firms with less-active boards. The frequency of board
meetings reduces the debt ratio, indicating the effort of the BOD in actively monitoring
financial operations (Anderson et al., 2004). This result is similar to that of Stephanus et
al. (2014) and Vafeas (1999).

</div>
<span class='text_page_counter'>(17)</span><div class='page_container' data-page=17>

<b>Table 5. Regression results for static Models </b>

Variable POLS_static
(1)
Random_static
(2)
Fixed_static 1
(3)
Fixed_static 2
(4)
Fixed_static 3
(5)

Lbsize -0.0090
[0.650]

-0.0040
[0.820]
-0.0010
[0.960]
-0.0040
[0.860]
-0.0020
[0.930]

Lmeet 0.0050
[0.420]
-0.0030
[0.530]
-0.0060
[0.400]
-0.0060
[0.410]
-0.0060
[0.400]

Female -0.0120***
[0.000]

-0.0050
[0.210]

-0.0020
[0.730]

Female_% -0.0250

[0.530]

Female_dummy -0.0050

[0.710]

Ceodual 0.0080
[0.370]
0.0000
[0.955]
-0.0020
[0.880]
-0.0020
[0.890]
-0.0020
[0.890]

Lfsize 0.0420***
[0.000]
0.0780***
[0.000]
0.1010***
[0.000]
0.1010***
[0.000]
0.1010***
[0.000]

Fixed_assets -0.2150***

[0.000]
-0.1300***
[0.000]
-0.1040**
[0.030]
-0.1040**
[0.030]
-0.1040**
[0.030]

ROA -0.7720***
[0.000]
-0.4970***
[0.000]
-0.4720***
[0.000]
-0.4720***
[0.000]
-0.4720***
[0.000]

Liquidity -0.0310***
[0.000]
-0.0190
[0.000]
-0.0170***
[0.000]
-0.0170***
[0.000]
-0.0170***

[0.000]

Salegrowth -0.0030
[0.100]
0.0020*
[0.060]
0.0020**
[0.040]
0.0020**
[0.040]
0.0020**
[0.050]

Constant -0.1330*
[0.090]
-0.9120***
[0.000]
-1.4910***
[0.000]
-1.4920***
[0.000]
-1.4880***
[0.000]

Year dummies Yes Yes Yes Yes Yes

Industry

dummies Yes Yes No No No
Number of

observations 1,592 1,592 1,592 1,592 1,592

R2 0.4950 0.3853 0.3330 0.3330 0.3330

F statistic (or

Wald statistic) 49.7260 811.6100 9.2830 9.4630 9.2930

P-value 0.0000 0.0000 0.0000 0.000 0.0000

</div>
<span class='text_page_counter'>(18)</span><div class='page_container' data-page=18>

The analysis is performed on 1,592 observations collected from 199 nonfinancial
companies listed on the Ho Chi Minh City Stock Exchange from 2012 to 2019. Lev is the
dependent variable, defined as the ratio of total debt to total assets of the business. Bsize
is the number of board members. Meet is the number of board meetings per year. Female
is the number of female members of the Board of Directors. Female_% is the percentage
of female directors. Female_dummy is an indicator variable equal to 1 if the board has
female members, and 0 if not. Ceodual is an indicator variable equal to 1 if the chairman
of the BOD also holds the position of CEO, and 0 if not. Lfsize is the logarithm of the
firm's total book value. Fixed_assets is the ratio of fixed assets to total assets. ROA is the
ratio of after-tax profit to total assets. Liquidity is the ratio of term assets to
short-term liabilities. Salegrowth is the rate of sales growth. Column (1) presents the regression
results for Model (1), using the POLS regression method. Column (2) presents the
regression results for Model (2) using the regression method with random effects. Column
(3) presents the regression results for Model (2), using the regression method with fixed
effects. Column (4) presents the regression results for Model (2), using the regression
method with fixed effects and variable Female_% replacing variable Female. Column (5)
presents the regression results for Model (2), using the regression method with fixed
effects and variable Female_dummy replacing variable Female. The test statistic of the
LM test for random effects is Chibar-square(1) = 22.8 (Prob(chibar-square)=0.000),

indicating that the Model with random effects (Column 2) is more efficient than the POLS
Model (Column 1). The test statistic for the Hausman test is Chi-square(16)=1055.3
(Prob(chi-square)=0.000), indicating that the Model with fixed effects (Column 3) is
consistent and more efficient than the Model with random effects (Column 2). In general,
the Model used for analysis is the fixed effects Model.

Model (1) is structured as follows:

𝐿𝑒𝑣𝑖𝑡 = 𝛽1+ 𝛽2𝐿𝑏𝑠𝑖𝑧𝑒𝑖𝑡+ 𝛽3𝐿𝑚𝑒𝑒𝑡𝑖𝑡+ 𝛽4𝐹𝑒𝑚𝑎𝑙𝑒𝑖𝑡+ 𝛽5𝐶𝑒𝑜𝑑𝑢𝑎𝑙𝑖𝑡

+ 𝛽6𝐿𝑓𝑠𝑖𝑧𝑒𝑖𝑡+ 𝛽7𝐹𝑖𝑥𝑒𝑑_𝑎𝑠𝑠𝑒𝑡𝑠𝑖𝑡+ 𝛽8𝑅𝑂𝐴𝑖𝑡

+ 𝛽9𝐿𝑖𝑞𝑢𝑖𝑑𝑖𝑡𝑦𝑖𝑡 + 𝛽10𝑆𝑎𝑙𝑒𝑔𝑟𝑜𝑤𝑡ℎ𝑖𝑡+ 𝜀𝑖𝑡

(4)
Model (2) is structured as follows:

𝐿𝑒𝑣_𝑖𝑡 = 𝛽₁+ 𝛽₂𝐿𝑏𝑠𝑖𝑧𝑒_𝑖𝑡+ 𝛽₃𝐿𝑚𝑒𝑒𝑡_𝑖𝑡+ 𝛽₄𝐹𝑒𝑚𝑎𝑙𝑒_𝑖𝑡+ 𝛽₅𝐶𝑒𝑜𝑑𝑢𝑎𝑙_𝑖𝑡

+ 𝛽6𝐿𝑓𝑠𝑖𝑧𝑒𝑖𝑡+ 𝛽7𝐹𝑖𝑥𝑒𝑑_𝑎𝑠𝑠𝑒𝑡𝑠𝑖𝑡+ 𝛽8𝑅𝑂𝐴𝑖𝑡

+ 𝛽9𝐿𝑖𝑞𝑢𝑖𝑑𝑖𝑡𝑦𝑖𝑡+ 𝛽10𝑆𝑎𝑙𝑒𝑔𝑟𝑜𝑤𝑡ℎ𝑖𝑡+ 𝜇𝑖 + 𝜀𝑖𝑡

(5)

<b>Table 6. Regression results for dynamic Models </b>

Variable POLS-dynamic
(1)

Fixed-dynamic
(2)

GMM-Dynamic 1
(3)

GMM-Dynamic 2
(4)

GMM-Dynamic 3
(5)

L.Lev 0.820***
[0.000]

0.410***
[0.000]

0.752***
[0.000]

0.759***
[0.000]

0.705***
[0.000]
Lbsize -0.010

[0.330]

-0.012
[0.480]

-0.006
[0.820]

-0.021
[0.430]

</div>
<span class='text_page_counter'>(19)</span><div class='page_container' data-page=19>

<b>Table 6. Regression results for dynamic Models (cont.) </b>

Variable POLS-dynamic
(1)
Fixed-dynamic
(2)
GMM-Dynamic 1
(3)
GMM-Dynamic 2
(4)
GMM-Dynamic 3
(5)

Lmeet -0.002
[0.520]
-0.002
[0.660]
-0.019**
[0.040]
-0.015
[0.012]

-0.014*
[0.090]
Female -0.003

[0.130]

-0.001
[0.840]

-0.018**
[0.020]

Female_% -0.103**

[0.020]

Female_dummy -0.032**

[0.030]
Ceodual 0.005

[0.270]
-0.003
[0.710]
-0.010
[0.340]
-0.008
[0.420]
-0.009
[0.410]

Lfsize 0.012***

[0.000]
0.078***
[0.000]
0.044***
[0.000]
0.045***
[0.000]
0.036***
[0.000]
Fixed_assets -0.069***

[0.000]
-0.120***
[0.000]
-0.080*
[0.070]
-0.076*
[0.090]
-0.081**
[0.050]
ROA -0.283***

[0.000]
-0.415***
[0.000]
-0.218**
[0.050]
-0.226**

[0.050]
-0.347***
[0.000]
Liquidity -0.008***

[0.000]
-0.013***
[0.000]
-0.001
[0.510]
-0.001
[0.680]
-0.002
[0.340]
Salegrowth -0.001

[0.420]
0.001
[0.330]
0.003
[0.190]
0.002
[0.240]
0.003
[0.160]
Constant -0.083**

[0.040]
-1.221***
[0.000]

0.000
[.]
-0.709***
[0.000]
0.000
[.]
Year dummies Yes Yes Yes Yes Yes
Industry dummies Yes No Yes Yes Yes
Number of

observations

1,393 1,393 1,393 1,393 1,393

R2 _0.875 _0.448 _- _- _-

F statistic (or
Wald statistic)

453.598 34.288 29,205.64 15,832.75 30,388.74

P-value 0.000 0.000 0.000 0.000 0.000
Number of

instruments

87.000 87.000 87.000

Arellano-Bond
AR(1) (P-value)

0.000 0.000 0.000

Arellano-Bond
AR(2) (P-value)

0.630 0.562 0.562

Hansen J
(P-value)

0.288 0.312 0.312

</div>
<span class='text_page_counter'>(20)</span><div class='page_container' data-page=20>

The analysis is performed on 1,592 observations collected from 199 nonfinancial
companies listed on the Ho Chi Minh City Stock Exchange from 2012 to 2019. The use
of lagged variables (1 period) reduces the number of observations to 1393. Lev is the
dependent variable, defined as the ratio of total debt to total assets of the business. L.Lev
is lagged one period from Lev. Bsize is the number of board members. Meet is the number
of board meetings per year. Female is the number of female members of the Board of
Directors. Female_% is the percentage of female directors. Female_dummy is an
indicator variable equal to 1 if the board has female members, and 0 if not. Ceodual is an
indicator variable equal to 1 if the chairman of the BOD also holds the position of CEO,
and 0 if not. Lfsize is the logarithm of the firm's total book value. Fixed_assets is the ratio
of fixed assets to total assets. ROA is the ratio of after-tax profit to total assets. Liquidity
is the ratio of short-term assets to short-term liabilities. Salegrowth is the rate of sales
growth. Column (1) presents the regression results for Model (3), using the POLS method.
Column (2) presents the regression results for Model (3) using the regression method with
fixed effects. Column (3) presents the regression results for Model (3) using the system
GMM method. Column (4) presents the regression results for Model (3) using the system
GMM method and variable Female_% replacing variable Female. Column (5) presents

the regression results for Model (3) using the system GMM method and variable
Female_dummy replacing variable Female.

Model (3) is structured as follows:

𝐿𝑒𝑣_𝑖𝑡 = 𝛽₁+ 𝛽₂𝐿. 𝐿𝑒𝑣_𝑖𝑡+ 𝛽₃𝐿𝑏𝑠𝑖𝑧𝑒_𝑖𝑡+ 𝛽₄𝐿𝑚𝑒𝑒𝑡_𝑖𝑡+ 𝛽₅𝐹𝑒𝑚𝑎𝑙𝑒_𝑖𝑡

+ 𝛽₆𝐶𝑒𝑜𝑑𝑢𝑎𝑙_𝑖𝑡+ 𝛽₇𝐿𝑓𝑠𝑖𝑧𝑒_𝑖𝑡+ 𝛽₈𝐹𝑖𝑥𝑒𝑑_𝑎𝑠𝑠𝑒𝑡𝑠_𝑖𝑡

+ 𝛽₉𝑅𝑂𝐴_𝑖𝑡+ 𝛽₁₀𝐿𝑖𝑞𝑢𝑖𝑑𝑖𝑡𝑦_𝑖𝑡+ 𝛽₁₁𝑆𝑎𝑙𝑒𝑔𝑟𝑜𝑤𝑡ℎ_𝑖𝑡+ 𝜇_𝑖 + 𝜀_𝑖𝑡 (6)

<b>5. </b> <b>CONCLUSIONS </b>

This study examines the impact of BOD characteristics on the financial leverage
ratio of nonfinancial companies listed on the Ho Chi Minh City Stock Exchange for eight
years from 2012 to 2019. The dataset includes 199 companies with 1,592 observations.
The system GMM used to analyze the data helps avoid correlation between the
instrumental variables and the errors, overcomes the endogeneity problem, and estimates
the regression coefficients consistently.

</div>
<span class='text_page_counter'>(21)</span><div class='page_container' data-page=21>

lower debt ratios compared to boards without female directors. The size of the BOD and
CEOs with dual roles were found to have no impact on the leverage ratio.

This result complements and follows previous studies on the influence of
governance structure on corporate capital structure, which is the basis for providing
suggestions to managers on operating and managing their businesses.

<b>ACKNOWLEDGMENTS </b>

This research is funded by a Dalat University research grant.

<b>REFERENCES </b>

Abobakr, M., & Elgiziry, K. (2015). The effects of board characteristics and ownership
<i>structure on the corporate financial leverage. Accounting and Finance Research. </i>
<i>5(1), 1-14. </i>

Abor, J. (2007). Corporate governance and financing decisions of Ghanaian listed firms.
<i>Corporate Governance, 7(1), 83-92. </i>

Adam, R. B., & Ferreira, D. (2009). Women in the boardroom and their impact on
<i>governance and performance. Journal of Financial Economics, 94(2), 291-309. </i>
Adams, R., & Mehran, H. (2003). Is corporate governance different for bank holding

<i>companies? Economic Policy Review, 9(1), 123-142. </i>

Adams, R., & Funk, P. (2012). Beyond the glass ceiling: Does gender matter?
<i>Management Science, 58(2), 219-235. </i>

Adams, R., de Haan, J., Terjesen, S., & van Ees, H. (2015). Board diversity: Moving the
<i>field forward. Corporate Governance: An International Review, 23(2), 77-82. </i>
Al-Najjar, B. (2011). The determinants of the frequency of board meetings: Evidence

<i>from categorical analysis. Journal of Applied Accounting Research, 13(2) 178-190. </i>
Anderson, R. C., Mansi, S. A., & Reeb, D. M. (2004). Board characteristics, accounting
<i>report integrity, and the cost of debt. Journal of Accounting and Economics, 37(3), </i>
315-342.

Berger, P. G., Ofek, E., & Yermack, D. L. (1997). Managerial entrenchment and capital
<i>structure. Journal of Finance, 4, 1411-1438. </i>

Blundell, R., & Bond, S. (1998). Initial conditions and moment restrictions in dynamic
<i>panel data Models. Journal of Econometrics, 87, 115-143. </i>

Bokpin, G. A. & Arko, A. C. (2009). Ownership structure, corporate governance, and
<i>capital structure decisions of firms: Empirical evidence from Ghana. Studies in </i>
<i>Economics and Finance, 26(4), 246-256. </i>

</div>
<span class='text_page_counter'>(22)</span><div class='page_container' data-page=22>

Bradley, M., Jarrell, G., & Kim, E. H. (1984). On the existence of an optimal capital
<i>structure: Theory and evidence. Journal of Finance, 39, 857-878. </i>

Brick, I., & Chidambaran, N. K. (2007). Board meetings, committee structure, and firm
<i>value. Journal of Corporate Finance, 16(4), 533-553. </i>

Buchdadi, A. D., Ulupui, I. G. K. A, Dalimunthe, S., Pamungkas, B. G., & Fauziyyah, Y.
<i>(2019). Board of director meeting and firm performance. Academy of Accounting </i>
<i>and Financial Studies Journal, 23(2), 1-7. </i>

<i>Castanias, R. (1983). Bankruptcy risk and optimal capital structure. Journal of Finance </i>
<i>38, 1617-1635. </i>

Đặng, T. Q. A., & Quách, T. H. Y. (2014). Các nhân tố ảnh hưởng tới cấu trúc vốn của
các doanh nghiệp niêm yết tại Sở giao dịch chứng khoán Thành phố Hồ Chí Minh.
<i>Tạp chí Phát triển và hội nhập, 18(28), 34-39. </i>

<i>Demsetz, H. (1983). The structure of ownership and the theory of the firm. Journal of </i>
<i>Law & Economics, 26(2), 375-390. </i>

Eluyela, D. F., Akintimehin, O. O., Okere, W., Ozordi, E., Osuma, G. E., Ilogho, S. O.,
& Oladipo, O. A. (2018). Board meeting frequency and firm performance:

<i>Examining the nexus in Nigerian deposit money banks. Heliyon, 4(10), 1-14. </i>
<i>Fama, E. F., & Jensen, M. C. (1983). Separation of ownership and control. Journal of </i>

<i>Law & Economics, 26(2), 301-325. </i>

Fosberg, R. (2004). Agency problems and debt financing: Leadership structure effects.
<i>Corporate Governance, 4, 31-38. </i>

Francis, B., Hasan, I., & Wu, Q. (2015). Professors in the boardroom and their impact
<i>on corporate governance and firm performance. Financial Management, 44(3), </i>
547-581.

<i>Friend, I., & Hasbrouck, J. (1985). Determinants of capital structure. Philadelphia, USA: </i>
Rodney L. White Center for Financial Research Publishing.

Friend, I., & Lang, L. H. P. (1988). An empirical test of the impact of managerial
<i>self-interest on corporate capital structure. Journal of Finance, 43(2), 271-281. </i>
Haque, F., Arun, T. G., & Kirkpatrick, C. (2011). Corporate governance and capital

<i>structure in developing countries: A case study of Bangladesh. Applied </i>
<i>Economics, 43, 673-681. </i>

Harris, C. (2014). Women directors on public company boards: Does a critical mass affect
<i>leverage? Business and Economics Faculty Publications, 29, 1-18. </i>

Jensen, M. (1986). Agency costs of free cash flow, corporate finance, and takeovers.
<i>American Economic Review, 76(2), 323-329. </i>

Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency
<i>costs and ownership structure. Journal of Financial Economics, 3, 306-360. </i>

<i>Jiankoplos, N. A., & Bernasek, A. (1998). Are women more risk averse? Economic </i>

</div>
<span class='text_page_counter'>(23)</span><div class='page_container' data-page=23>

Johl, S. K., Kaur, S., & Cooper B. J. (2015). Board characteristics and firm performance:
<i>Evidence from Malaysian public listed firms. Journal of Economics, Business and </i>
<i>Management, 3(2), 239-243. </i>

Kajananthan, R. (2012). Effect of corporate governance on capital structure: Case of the
<i>Sri Lankan listed manufacturing companies. Researchers World, 3(4), 63. </i>

Kyereboah-Coleman, A., & Biekpe, N. (2006). Corporate governance and financing
<i>choices of firms: A panel data analysis. South African Journal of Economics, </i>
<i>74(4), 670-681. </i>

Lê, T. D., Bùi, Đ. T., & Lê, T. H. (2020). Nhân tố ảnh hưởng đến cấu trúc vốn của doanh
<i>nghiệp ngành thực phẩm trên thị trường chứng khốn. Tạp chí Tài chính. </i>
Retrieved from
/>cau-truc-von-cua-doanh-nghiep-nganh-thuc-pham-tren-thi-truong-chung-khoan-325679.html.

Liao, L. K., Mukherjee, T., & Wang, W. (2015). Corporate governance and capital
<i>structure dynamics: An empirical study. The Journal of Financial Research, </i>
<i>38(2), 169-191. </i>

Lipton, M., & Lorsch, J. (1992). A modest proposal for improved corporate governance.
<i>The Business Lawyer, 48(1), 59-77. </i>

Long, M., & Malitz, I. (1985). The investment-financing nexus: Some empirical
<i>evidence. Midland Corporate Finance Journal, 3, 53-59. </i>

Lương, T. T. D., Phạm, N. U. N., Nguyễn, T. T., Nguyễn, T. K. L, Nguyễn, T. T. T., &
Phạm, B. T. (2020). Nhân tố ảnh hưởng đến cấu trúc vốn của doanh nghiệp

<i>logistics trên thị trường chứng khốn Việt Nam. Tạp chí Tài chính. Retrieved from </i>


Marinova, J., Plantenga, J., & Remery, C. (2015). Gender diversity and firm performance:

<i>Evidence from Dutch and Danish boardrooms. The International Journal of </i>
<i>Human Resource Management, 27(15), 1777-1790. </i>

Modiglani, F., & Miller, M. H. (1958). The cost of capital, corporate finance and theory
<i>of investment. The American Economic Review, 48(3), 261-297. </i>

Modiglani, F., & Miller, M. H. (1963). Corporate income taxes and the cost of capital: A
<i>correction. The American Economic Review, 53(3), 433-443. </i>

<i>Morrison, A. M., White, R. P., & Velsor, E. V. (1987). The narrow band. Leadership in </i>
<i>Action, 7(2), 1-7. </i>

Nguyễn, T. (2008). Các nhân tố ảnh hưởng đến cấu trúc tài chính của các doanh nghiệp
<i>chế biến thủy sản Khánh Hòa. Tạp chí Khoa học-Cơng nghệ Thủy sản, 3, 54-59. </i>
Nguyen, T., Locke, S., & Reddy, K. (2015). Ownership concentration and corporate

</div>
<span class='text_page_counter'>(24)</span><div class='page_container' data-page=24>

Nguyễn, T. H., Trần, D. T., Nguyễn, H. N. H., Võ, T. T. N., & Nguyễn, T. T. (2016).
Quản trị công ty và quá trình điều chỉnh động cấu trúc vốn - quan sát từ các doanh
<i>nghiệp niêm yết trên thị trường chứng khốn Việt Nam. Tạp chí Khoa học Đại </i>
<i>học Mở, 50(5), 25-40. </i>

Ntim, C. G., & Osei, K. A. (2011). The impact of corporate board meetings on corporate
<i>performance in South Africa. African Review of Economics and Finance, 2(2), </i>
83-103.

Phan, B. G. T., Trần. Đ. T., & Trần, T. T. A. (2017). Ảnh hưởng của đặc điểm Tổng giám
<i>đốc điều hành đến hiệu quả hoạt động của doanh nghiệp. Tạp chí Khoa học Đại </i>
<i>học Mở, 55(4), 51-63. </i>

Roodman, D. (2009). How to do xtabond2: An introduction to difference and system
<i>GMM in Stata. The Stata Journal, 9(1), 86-136. </i>

Rose, C., Munch-Madsen, P., & Funch, M. (2013). Does board diversity really matter?
<i>Gender does not, but citizenship does. International Journal of Business Science </i>
<i>and Applied Management, 8(1), 15-27. </i>

Schwartz-Ziv, M. (2017). Gender and board activeness: The role of a critical mass.
<i>Cambridge University Press, 52(2), 751-780. </i>

Sheikh, N. A., & Wang, Z. (2012). Effects on corporate governance on capital structure:
<i>Empirical evidence from Pakistan. Corporate Governance, 12(5), 629-641. </i>
Simpson, W. G., & Gleason, A. E. (1999). Board structure, ownership, and financial

<i>distress in banking firms. International Review of Economics & Finance, 8(3), </i>
281-292.

Stephanus, R. W., Anastasia, K. N. F. T., & Toto, R. (2014). The effect of corporate
<i>governance on capital structure in public listed companies in Indonesia. Abstract </i>
<i>of Emerging Trends in Scientific Research, 1, 1-16. </i>

Tarus, D. K., & Ayabei, E. (2016). Board composition and capital structure: Evidence
<i>from Kenya. Management Research Review, 39(9), 1056-1079. </i>

<i>Titman, S., & Wessels, R. (1988). The determinants of capital structure choice. The </i>
<i>Journal of Finance, 43(1), 1-19. </i>

Tran, T. T. T., Hoang, H. L., & Tran, T. H. Y. (2015). Relationship between gender
diversity on boards and firm's performance - Case study about ASEAN banking
<i>sector. International Journal of Financial Research, 6(2), 150-159. </i>

Trương, Đ. L., & Võ, K. T. (2008). Các yếu tố ảnh hưởng đến cấu trúc vốn của các công
<i>ty cổ phần niêm yết trên thị trường chứng khoán Việt Nam. Tạp chí Nghiên cứu </i>
<i>kinh tế, 361, 21-26. </i>

<i>Vafeas, N. (1999). Board meeting frequency and firm performance. Journal of Financial </i>
<i>Economics, 53(1), 113-142. </i>

</div>
<span class='text_page_counter'>(25)</span><div class='page_container' data-page=25>

Wen, Y., Rwegasira, K., & Bilderbeek, J. (2002). Corporate governance and capital
<i>structure decisions of the Chinese listed firms. Corporate Governance: An </i>
<i>International Review, 10(2), 75-83. </i>

Wintoki, M. B., Linck, J. S., & Netter, J. M. (2012). Endogeneity and the dynamics of
<i>internal corporate governance. Journal of Financial Economics, 105(3), 581-606. </i>
Wiwattanakantang, Y. (1999). An empirical study on the determinants of the capital

<i>structure of Thai firms. Pacific-Basin Finance Journal, 7, 371-403. </i>

</div>

<!--links-->